College and Research Libraries By RUTHERFORD D. ROGERS Appraising a Research Collection Mr. Rogers is director~ The Grosvenor Library~ Buffalo~ N.Y. Introduction This report on methods and results of an appraisal of the Grosvenor Library is offered for three reasons : 1. To present per volume insurance values, by Library of Congress classes, for a research collection of approxi- mately one-half million volumes, • 2. To present a new method of applying processing costs to total holdings in order to arrive at a practical per volume insurance value, 3· To present processing costs, particu- larly in regard to phonograph records, which may be of interest to other li- braries. The Grosvenor Library is a non-circulat- ing reference and research library estab· lished in I859. The collection is especially strong in bibliography, literature, history (with emphasis on local and American his- tory), genealogy, fine arts, and music (in- cluding sheet music, phonograph records, and definitive editions of major composers). There is also · a Medical Department con- taining approximately I 5,000 older books and journals, but current emphasis is now placed on an effective, up-to-date working collection of sooo-8ooo volumes in that field. Method The appraisal described here was com- pleted in July, I950, and was based on the "first method" recommended by Dorothea Singer on p. 33-34 of The Insurance of Libraries (American Library Association, I 946). Only one part of this method, that pertaining to binding, was · not considered applicable. For the most part, Grosvenor Library binding funds have been devoted to binding the current intake of journals, and inasmuch as the Grosvenor is a non- circulating library, its books are not subj.ect to the wear that those in a conventional circulating library are. Although binding expenditures were not specifically and ob- jectively considered, the physical condition of books was kept in mind in applying de- preciation factors, as may be seen in the footnotes to Table I. For those who may not have a copy of Miss Singer's book easily available, the procedure for computing insurance values involves: a. A per volume cou~t, by class or sub- class, of the collection, either from the shelves (if all books are in) or from the shelf list. Both methods were used in this appraisal. b. The careful spot pricing of books at equal intervals throughout each class. We priced 5,298 volumes in accord-ance with this sampling . technique, plus the volume by volume pricing of certain rare and valuable books. In most classes, every fiftieth volume was priced, but in a few homogeneous classes every one- hundredth volume. Prices were taken from accession and order records extend- ing from 1947 to 1863. Where such information was not available, volumes were priced from second-hand catalogs or, in a limited number of cases, by staff specialists in the various subject fields. Figures prepared by the Insurance Com- mittee of the American Library Associa- tion (ALA Bulletin, Vol. 38, Oct. 1944, 24 COLLEGE AND RESEARCH LIBRARIES / Table 1 Computation of per Volume and per Class Insurance Values Column I Column2 Column 3 Column 4 Column 5 Column6 Column 7 Column 8 Average Deprecia- Adjusted Proc- Per Original tion and Average essing Volume Total Num- Cost or Obsoles- Current Factor Insurance Insurance Library Classification her of Value cence or Value (See Value Values Vols. per Apprecia- per Table Including by Class Volume tion Volume Ill) Processing (Per Cent) Factor A- General Works 6,834 $2.38 -IS $2.02 $.116 $2. I36 $14,597 ·42 B- Philosophy 22,821 2.27 -40 1.36 .208 I.s68 35, 783 .32 C- His tory-Auxiliary 8,366 4·50 +rs 5. 17 .256 5·426 45,393.92 Sciences D- History and Topogra- 35' 836 2.42 -rs phy except America 2.os ·3~8 2.418 86,6sr .44 £ - America 12,868 2.14 None 2.14 ·5I3 2 . 653 34,138.8o F- America r8,9I5 3·33 +rr 3·70 .254 3 · 954 74,789.91 G- Geography, Anthro- 5,292 2.72 -45 r.so ·447 I ·947 10,303.52 pology H - Social Science 38' !64 1.75 -20 1.40 ·313 I · 713 65,374·93 ] - Political Science 13,539 r.6o None r.6o .J2I 1.921 26,oo8 .42 K- Law 1,541 1.73 -40 1.04 .245 I.28S I, 980. r8 L- Education 8,435 r.6o -IS I.J6 ·368 1.728 14,575·68 M- Music 12 , o26 2.9s None 2 . 9s .224 J . 174 J8 1 I70.52 N- Fine Arts r8,88r 6.88 None 6.88 · 747 7.627 144,005.38 P - Language & Literature 6,r ,2IO I. 55 -rs 1.32 .ro8 I .428 87,407.88 Q- Science 22,730 2 . 44 -20 1.95 .JI7 2.267 51,528.9I R- Medicine 8,829 3.62 -so r. 8r .279 2.089 r8 , 443.78 S- Agricul ture 8,552 I. 84 -rs r. s6 ·470 2.03~ r7,36o.s6 T - Technology 24,0I3 2.17 -40 I.JO ·402 1.702 40,870.12 U-V- Military and Naval 3,755 1.63 -ro 1.47 .J2J 1.793 6 , 732. 71 Science Z- Bibliography and Li- 8,609 brary Science J.28 -10 2.9s .116 J . o66 26,395·19 Explanation of Column 4. Table I Following are some of the important considerations which affected our decisions in regard to the figures in Column 4 · Notes are arranged by L. C. class letters: A Twenty-five per cent of this collection consists of almanacs which have maintained or increased in value since purchase. Many of our encyclopedias would also cost somewhat more than original purchase price. B Approximately ten per cent of this class comprises magazines which are valued at ALA insurance figures and therefore are not subject to further depreciation. Many standard works and the unusual and valuable Shaker collection are also in this class. These considerations are offset by the extensive amount of inspirational literature. c Many books in this class have doubled or more than doubled in value since time of purchase. The fifteen per cent appre- ciation is considered particularly moderate in view of current costs of books in this field. D&E The physical condition and basic nature of books in these classes led several department heads to advocate a five to fifteen per cent appreciation for them, but it was decided to use a more conservative valuation. F In view of current costs of books in this field, the excellence of our collection and the fact that these books are likely to in- crease rather than decrease in value, an eleven per cent appreciation is regarded as low rather than high. G The strength of this class has not been maintained. Many of the books are out of date. H Almost fifty per cent of this class is government document material priced with ALA insurance values not subject to fur- ther depreciation. The twenty per cent depreciation is therefore greater than it might appear. J This class is predominantly made up of government documents priced at ALA insurance values. K Despite the substantial amount of government document material in this class forty per cent depreciation was applied because of the weakness of the non-document material. · L Periodicals priced at ALA insurance values constitute much of this class; therefore, fifteen per cent depreciation applied to the whole class is larger than it might appear. M The depreciation in half of this class has been at least offset by appreciation of the remainder of the class. The excellence of our music collection and the current cost of books in this field make the evaluation conservative. N . This is an excellent collection containing an unusual number of oversize books. Our failure to appreciate this class is a compromise between current cost factors and the condition of the collection on one hand and the exceptional nature of the collection on the other. JANUARY1 1952 25 p The fifteen per cent depreciation of this class reduces the unit price well below the ALA average for adult non-fiction, but there is enough fiction represented in the class, despite the fact that we buy practically no fiction, to warrant the lower figure. On the other hand sets of standard authors and reference books have kept us from using a greater depreciation figure. Q A twenty per cent depreciation was considered adequate because of the number of documents and periodicals for which ALA insurance values were used and because many of the books are basic science books of historical significance. R Fifty per cent depreciation is excessive when applied to the current collection, but the high figure was used in view of the many older non-periodical volumes. s This class contains many government documents priced at ALA insurance values and several exceptional special collec- tions (early sporting books and books on gardening). T Standard works in technology and special collections, e.g., cook books, keep the value of this class as high as it is despite lack of current buying in this field. U-V Much of this material is of historical interest including military costume books of great value. . z The depreciation of this class represents a compromise between a substantial appreciation in the bibliography sectiun and a substantial depreciation in th; library science section. Table II Appraisal of Catalogs, Indexes and Specially Classified or Unclassified Library Materials Column I Column 2 Column 3 Column 4 Number of Vols. Unit Total Description of Items Unless Otherwise Insurance Insurance Stated i Value Value Congressional Series 7,4I6 $ I .oo* $ 7,4I6.oo U. S. Patent Specifications 9,309 2 ~ oo* I8,6I8.oo U. S. Patent Gazettes and Indexes 824 2.oo* I,648.oo British and Canadian Patents 792 I .oo* 792.00 Magazines (General and Medical not else- where classified) Bound 27,046 3.oo* 8I, I38 .oo Unbound I,768 2.00 3,s36.oo Newspapers Before I8oo I, 2 4 issues 3.oo* 3,672.00 r8oo-I82o 209 issues ·7S* rs6.7S 182o-I840 614 issues .so* 307.00 r84o-I86s 19,66s issues .2s* 4,9I6.2S I86S-I947 84, I88 issues . ro* 8,418.8o Fixed Classification 2,483 I .46* 3,62S.I8 League of Nations Documents 3,S9I 2, I39.004 Rare Books S,366l 81, 93S .6o Microfilm S04 spools S, S77 .oo4 Phonograph records I3, 262 records I ·392 18,434. I8 Boston Evening Transcript genealogy col- 141 3S-OO 4,93S-OO . umns mounted in scrapbooks (40 years) Index to Boston Evening Transcript geneal- 10 3SO.OO 3,soo.oo ogy column, in I 6-inch special oversize binders Hartford Times genealogy columns mounted 23 3s.oo 8os .oo in scrapbooks Index to Hartford Times genealogy column 6 soo.oo 3,ooo.oo in I6-inch special oversize binders Main Card Catalog and Shelflist3 Approximately $78,ooo with microfilm 1 , 300, ooo cards of shelflist ($496,ooo Departmental Card Catalogs and Indexes 400, ooo ca~ds without microfilm) Not Otherwise Listed3 24,000.00 Explanatory Notes-Table II * Figures in Column 3 followed by an asterisk are insurance values suggested by the Insurance Committee of the American Library Association (ALA Bulletin, Vol. 38 (October 1944), p. 369). Depreciation and obsolescence have been considered in arriving at these values. 1 Individually priced. 2 The unit insurance value for phonograph recortls comprises $z.oo per record average replacement cost plus $o.3 9'"per record for processing (see Table IV). · • The valuation of the library's card catalogs is based on figures as to card reproduction and filing costs cited in R. B. Downs, Union Catalogs in the United States, Chicago, ALA, 1942. • Cost price. 26 COLLEGE AND RESEARCH LIBRARIES Table III Computation of Processing Factor Column 2 3 4 s 6 -----I------------I-------·1------- I-------I----------- I 7 I ....!..L:: ~:_a l _____ s _ ____ I ______ 9 _____ 1-~I_o __ "' "' CIS 0 "' "' v '"' bD s:: 0 u ...... 0 :>. .... CIS ..a ~ A B c D E F G H J K L M N p Q R s T U-V z '$ I3,S04.68 JI ,036. 56 43,252.22 73,46J.SO 27,S37·S2 69,9ss.so 7,93S.oo S3,429.60 2I,662.40 1,6o2.64 I I ,47I. 6o 3S,476·70 I29,90I.2S S0,797·20 44,323 . SO IS,9S0 .49 13,34I.I2 3I,2I6.90 S,SI9.SS 2S,396 . ss '$S.oo 3.00 6. I3 4.00 3.00 S·93 3·29 3.22 3·07 4.SI 3·04 6.04 6.63 3·7I 4·42 6.6S 3 ·42 3·33 3.2S 7·7 2 I,72S I0,34S 7,oss IS,26s 9, I79 I I, Sol 2,4I2 I6,S93 7,o56 333 3,773 s,S73 19,593 2I 'ns IO,o2S 2,392 3,900 9,374 I,6S2 3,2S9 75% 7S 90 so so So 25 so 6o IO 40 7S so 90 so 20 20 20 50 90 '$ 2SS·75 I,S5I.7S 1,269·90 I,S36 . so 9I7 · 90 I,SSS.I6 12o.6o I,6S9·30 S46·72 6.66 30I. S4 SS0.95 I,9S9·30 3,920.04 I ,oo2. So 9S.6S IS6.oo 374·96 I6S.2o S92.02 '0~ ~-.t~ CIS -.:::l ~ c:·- 5s~ l:!.B ~ ~8Z "0 v !lc:P=l <11'- .§ ~~ .... Eo ~::l~ 25% '$ 25 IO so so 20 7S so 40 90 6o 2S so IO so So So So so 10 S34· 7S '$ 3,2o6 . 9S S74·S2 II,3S6.30 5,690 · 9s 2,926.64 2, 243. J6 I0,2S7.66 3,499·77 37I . 62 2,So7.II I,S2o.63 I2,I47·66 2,700,47 6,2I7·36 2,372.S~ 3,S6S.So 9,299·00 I,042.S4 407.S4 793·SO 4.75S.70 2,I44·72 I3,222.So 6,6oS.SS 4,SI4.SO 2,363·76 II, 946.96 4.346·49 37S . 2S 3,IoS.9s 2,70I.5S 14, I06 . 96 6,62o.si 7,22o.I6 2,46S._S'4 4,024.SO 9,673·96 I 1 2II .04 999 · S6 '$. u6 .2oS .2s6 ·36S ·SI3 . 2S4 ·447 ·3I3 .32I .24.S' ·36S .224 ·747 . IoS ·3I7 .279 ·470 ·402 ·323 .II6 Explanatory Notes for Table III Column 3--Figures are based on 1948 and I9-t9 purchases. Column 5- These estimated percentages are based on the combined judgment of five key staff members. The nature of the Grosvenor Library and the availability of books in the second hand market have been considered in arriving at these fig~cl~mn 6- Figures in this column are computed in the following manner: Number of volumes in Column 4 X Percentage in Column 5 X $o. 2owhere So. 2o is the average cost of preparing and adding a volume when the same or similar edition is in the library. (See Table IV.) This computation is independent of the value of the card catalog and is based on the assump- tion that the card catalog is suitably insured, with or without a microfilm of the shelflist. Column 8- Figures in this column are computed in the following manner: Number of volumes in Column 4 X Percentage in Column 7 X $1.24 where $1.24 is the average cost per volume for cataloging in the Grosvenor Library. (See Table IV.) Column 9-Figures in this column are the sum of figures in Columns 6 and 8. Column ro-Figures in this column are computed by dividing Column 9 of this table by the corresponding figures in Column 2 of Table I. This reduces the anticipated processing cost to the same terms as corresponding figures in Column 5 of Table I. thereby making possible a per volume insurance value which includes the processing factor and which can be applied to total present holdings as represented in Column 2, Table I. p. 369) were used for periodicals and government documents with certain ex- ceptions. In this connection, it is im- portant to note that our government documents and our periodicals, other than general and medical periodicals, are classified according to the L. C . system. c. Application of depreciation, obsolescence and appreciation factors. This involves JANUARY~ 1952 a certain amount of judgment, and in- formation such as is listed in the foot- notes to Table I. d. Addition of the cost of processing. For this purpose, we developed the processing factor in Table III, the results of which are recorded in Column 6 of Table I. The first step in Table III was to com- pute the approximate insurance which 27 Table IV Processing Costs- Grosvenor Library Books (Based on the entire year 1949) · Salaries and wages chargeable to cata- loging $17,804.86 Cost of Library of Congress printed cards Cost of plain cards Cost of stencils 2,140.00 77·00 58.oo $20,079·86 N urn her of volumes cataloged (Includes · . 2,322 volumes originally cataloged and 7,443 volumes recataloged) P er volume average cost Phonograph Records (Based on the processing of 89 records, June 1950) Cost per Total Operation Hours Hour Cost Cataloging 13· 75 $1 .6o '$22 . 00 Typing added entries 7.66 1 . 20 9· 19 Labeling 2 . 33 1.20 2.80 Revi sing ·75 1.6o 1.20 $35.19 Unit cost $ 0 .395 would be collected, assuming coverage based on the number of volumes in Column 2, Table I, and the per volume values listed in Column 5, Table I. The next steps were to compute the number of replacement volumes we could buy at present prices with the insurance money and to estimate as intelligently as possible the percentage of the replace- ment volumes which would be essentially the same as present editions. These estimates were made by staff members who were thoroughly familiar with their subject fields, who had 20-30 years ex- perience in this library, and who were well acquainted with the second-hand market. Once we had determined the degree of identity which was likely to obtain between a replacement collection and the present one, we made two com- putations (Column 6 and 8 of Table III), the one being the · total cost of processing replacement volumes similar to present editions, and the other, the cost of processing replacement volumes which would be new to the collection. For processing costs, we used current experience as computed in Table IV. Total estimated processing costs of re- placement volumes were then reduced to a factor (Column 10, Table III) which could be applied to total present holdings but which would reflect the cost of processing a smaller replacement collection. .Explanation of Method Our failure to use a flat $1.24 per present volume as a processing cost may be ques- tioned. However, this would be unrealistic in terms of the number of volumes which would be purchasable with the insurance money, assuming an extensive loss. It also makes no distinction between replacements which would be essentially the same as, and those which would be different from, pres- ent editions. No library can be careless about funds, and the use of the flat $1.24 per volume would send total insurance and total premiums skyrocketing. The attempt here is to be practical without grossly over- or under-insuring. Several situations are possible in the event of an extensive loss, and each contingency must be considered in developing the in- surance program. a. Part or all of the card catalog and shelf list might be destroyed and the book collection remain intact. b. Part or all of the collection might be lost without destruction of the card catalog. c. Both catalog and collections might be wholly or partially destroyed. It is especially important to understand that the use of the processing factor as developed in Table III is based on the assumption that the card catalog will be a. Protected by microfilm (in our case, a microfilm of the shelf list) .1 1 The microfilm of the shelf list also provides a basis for making a claim in t he event of loss. 28 COLLEGE AND RESEARCH LIBRARIES b. Separately insured with or without a microfilm. (In the latter instance, much more insurance is required. We decided that three years' premiums would be so much greater without the protection afforded by a film that the film would pay for itself in that time). Some of the per volume insurance values in Column 7, Table I, may appear high in comparison with some of the insurance values suggested by ALA, but considera- tion of two factors should modify this im- pression. First, one should compare the figures in Column 7, Table I, with the corresponding figures in Column 3, Table III, the latter being the present cost (after substantial discounts) of books purchased by this library. Secondly, there are _many valuable books standing on the general shelves of the Grosvenor Library which have been covered only by the sampling technique. In 19 50 alone, we transferre~ twenty-five titles with a total value of $3300.00 from our general collection to the Rare Book Room. This project, which has only been started, is continuing with equally impressive results. Sequence of Tables With this general introduction, it 1s hoped that the tables with their accompany- ing notes will be largely self-explanatory. It is suggested that Table I be read through Column 5; then Tables III and IV in their entirety; and finally, Columns 6-8 of Table I, and Table II. Insurance Evaluation of a University Library's Collections (Continued from page 23) in which library property is located for the purpose of keeping waste paper, pack- ing boxes, and other inflammables cleared out. Conclusion In insuring a university library's collec- tions, the primary consideration is the secur- ing of coverage which, in the event of the loss by fire of all or · part of the holdings, would provide funds for the replacement of the items lost, at or close to present purchase costs. The rarities and other i terns of high monetary value should be given all- risk protection by being placed under a Fine Arts policy. To secure the maximum benefits which may be desired and to secure JANUARY, 1952 them at the lowest premium rate which will be con~istent with security, it is advis- able to keep in close touch with a reputable, experienced insurance broker, agent, or firm. By so doing, helpful counsel will be assured during the planning stage, during the evalu- ation, and during the later months and years whenever . problems connected with in- surance arise. The responsibility laid on the Librarian for computing the insurance values of the collections of a large library is heavy. How- ever, with teamwork by the staff specialists in the various subject fields and: with expe~ t insurance counsel, a sound insurance pro- gram which will provide reasonable pro- tection against all major risks can readily be worked out. 29