College and Research Libraries CLIFTON BROCK The Quiet Crisis in Government Publishing Although the publishing activities of the federal government comprise the most extensive information distribution system in the world, they fall far short of meeting the "information requirements of a highly- educated, industrialized, complex, and space-age society." Enormous amounts of money are spent annually on printing-some in the GPO, and much outside it-and there is considerable bureaucratic battling over what branch or agency receives priority in getting its publications into print. Documents are sometimes sold and often deposited, but the depository system is so uncomprehensive in scope and so costly to administer in the recipient libraries as to make it at best a mixed blessing. High level attention is badly needed. ''T HE UNITED STATES GOVERNMENT is the biggest publisher in the world." This statement, or variations on the same theme, is encountered frequently in the speeches and writings of govern- ment officials, librarians, scholars, and others whose work causes them to have an interest in federal government publi- cations. The connotations are not always positive. Occasionally, a legislator will arise in wrath on the floor of Congress, assail the bureaucrats as a bunch of paper pushers, regale his colleagues with quotations from the latest absurdity emanating from the government presses, such as "The Official Girlwatcher' s Man- ual,"1 and urge a massive cut in printing 1 Congressional Record, September 13, 1963, p. 16955. The Girlwatcher's Guide was not a govern- ment publication, but a citation to it was buried in an 814-page bibliography of programed instruction Mr. Brock is Chief, Business Administra- tion and Social Sciences Division, Univer- sity of North Carolina Library. This article is preprinted from a book to be published soon by the Public Affairs Press, edited by M. B. Schnapper, concerning government information policies and practices. appropriations. Whether meant nega- tively or positively, however, the state- ment implies an enormous government publishing program, with books, pamph- lets, and periodicals on every con- ceivable subject being produced by the thousands and distributed widely throughout the land. That there is much truth in this pic- ture rna y be seen from a few figures on the government's printing and publica-. tion distribution operations during fiscal year 1963: 1. The Government Printing Office produced $127.1 million worth of print- ed matter, resulting in a total produc- tion of 1,022,840,498 copies of publica- tions. 2. The Public Documents Division of GPO sold 53,076,581 copies of publi- cations, with total sales receipts of $11,297,784.06. 3. The Public Documents Division made free distribution of 5,817,058 materials produced on contract with the U.S. Office of Education. The incident is significant primarily as an example of Congressional ingenuity in nee- dling the executive bureaucracy. I 477 4781 College & Research Libraries • November, 1965 copies of publications to more than six hundred libraries across the country. These figures, moreover, as will be seen below, represent only a part of the government's investment in printing and publishing. Despite this seemingly enor- mous effort, the contention here will be that the government's program of print- ing, publishing, and distribution of pub- lications is seriously deficient when mea- sured against the information require- ments of a highly-educated, industri- alized, complex, and space-age society. There are, moreover, not one but several government publication and distribution programs, and they overlap and inter- sect in a tangle of administrative and operating confusion beside which the much-criticized administration of the foreign aid program is a model of ration- ality and efficiency. There is, in fact, a "quiet crisis" in government publishing which has almost completely escaped at- tention, both within and without the government. In attempting to analyze this crisis be- low, the focus is on two sets of ques- tions: 1. How much does the government spend on printing and publishing? What has been the pattern of this expenditure in recent years? What proportion of the total federal budget goes for printing and publishing? Is this level of expendi- ture adequate for present-day needs? 2. Whatever the answer to these ques- tions, taking the present government publications output as given, what hap- pens to these publications? To what ex- tent, and under what conditions, are they available to the general public, li- braries, scientists, and others who have a legitimate need for the information contained in government publications? ExTENT OF GOVERNMENT PUBLISIDNG It is a fairly simple matter, using sta- tistical data compiled by the govern- ment, to find such esoteric bits of in- formation as the annual receipts of pool halls and bowling alleys ill Great Falls, Montana ( $695,000 in 1963); the num- ber of strapless or convertible brassieres shipped annually in interstate commerce ( 12,864,000 in 1964); or the number of four-month-old or older chickens in Sagadahoc County, Maine (there are 21,025). Given such plenty, it might seem no problem at all to find out how much the government spends for print- ing, how much of this goes for printing of publications rather than forms, sta- tionery, etc., and how many publica- tions the government issues each year. Such is not the case. The government does not compile this information in its massive statistical program, and officials who have offered estimates from time to time have stressed that their figures are largely guesswork. In its analysis of federal printing expenditures in 1954, for example, the Hoover Commission- despite its great resources and vast au- thority-confessed that its figures were only "rough estimates."2 In attempting to assess over-all gov- ernment expenditures for printing, how- ever, we do have two estimates which- whatever their shortcomings-at least were arrived at by strictly comparable methods. One is the Hoover Commission estimate mentioned above, the other a summary of 1964 expenditures prepared by the Budget Bureau at the author's request .. The Hoover Commission, using the lin~ item labeled "printing and repro- duction" in the federal budget for fiscal year 1954, placed total government printing expenditures at $370 million.3 The Budget Bureau, using the same line item of the budget for fiscal 1964, placed the total expenditure for that year at $27 4 million. 4 Thus we have what appears to be an incredible bu- 2 Commission on Organization of the Executive Branch, Business Enterprises (Washington: GPO, 1955). p. 101. 8 Ibid. 'Letter from Chief of Budget Preparation, Bureau of the Budget, February 23, 1965. The Quiet Crisis in Government Publishing 1 479 reaucratic fact: government printing ex- penditures decreased by $96 million, or 26 per cent, during a decade when total government expenditures were increas- ing by 46 per cent. One other estimate, moreover, allows us to pinpoint the years during which this most non-Parkinsonian situation oc- curred. In 1960 the staff of the Congres- sional Joint Committee on Printing esti- mated total printing expenditures at $197.7 million. 5 While this estimate ap- parently was arrived at by different methods than the other two, it is the most authoritative available and should be reasonably comparable. Accepting it, we find that during the eight years of the Eisenhower Administration, expendi- tures for printing decreased by $172.3 million, or 47 per cent, while 'total fed- eral expenditures were rising by 13 per cent. Lest too many partisan implications be drawn from this analysis, however, the table below shows the pattern of print- ing expenditures as a percentage of total government expenditures during this century. From this it is clear that the Eisen- hower years were unique only in that the economy drives of that period re- sulted in an absolute decrease in the amount expended for printing.' Relative- ly, however, the percentage of the fed- 5 U.S. Congress, House Committee on Appropria- tions, Legislative Branch Appropriations for 1962, Hearings, 87th Cong., 1st Sess. (Washington: GPO, 19~1), p. 303. eral budget going for printing has been dropping drastically for decades, and the real bite came between 1940 and 1954, for which Democratic administra- tions must get the credit or take the blame.6 f Is the present level of expenditure adequate? It will be obvious that the author is preparing to come down on the negative side of this question. Such a stance, of course, is almost un-Ameri- can. Anyone who takes it is put into the position of arguing that the "paper- work jungle" of Washington should be increased rather than decreased. There are, moreover, no standards in this area. No one has devised an "optimum" ratio of government printing expenditures to total expenditures. Earlier ratios are hardly valid guides, since the range of goods and services produced by the gov- ernment has changed too greatly. To take only the most obvious example, the budget for fiscal year 1900 did not have 6 It should be noted that estimates of total printing expenditures made by the three sources-Joint Com- mittee on Printing, the Hoover Commission, and the Budg et Bureau-have differed considerably. The Hoover Commissi.on put 1954 expenditures at $370 million, while the Joint Committee staff put the 1955 figure at $165 million. For 1964 the Budget Bureau offers an estimate of $274 million, while the Joint Committee staff reports $235 million for the same year. All figures purportedly represent "total printing expenditures" for the respective years. If the Joint Committee staff is right and the Hoover Commission is wrong, there would be no absolute decrease in printing expenditures during the Eisen- hower Administration, as indicated above. On the other hand, the level of current expenditure would be lower. The discrepancies are perhaps mOilt sig- nificant as an indication of the confusion and lack of accurate data in this area. PERCENTAGE OF GOVERNMENT EXPENDITURES FOR PRINTING Total Printing Per Cent of Total Date Expenditures* Expenditures* Expended for Printing 1900t 447 3.9 . 0.87 1930 2,746 13.3. .48 1940 8,798 20.0 .23 1954 67,537 370.0 .0055 1960 76,539 197.7 .0026 1964 98,405 274.0 .0028 * In millions of dollars. t Data for 1900 through 1940 taken from LeRoy C. Merritt, The United States Government as Publisher. (Chicago: University of Chicago Press, 1943) -./ ( / / 480 j College & Research Libraries • November, 1965 to accommodate millions for military hardware and billions for moon shots. Thus the argument that present printing expenditures are not adequate has to be made largely on a prima facie basis. In attempting to make this case, and as background for analysis of problems of publication distribution, it is neces- sary to go into some detail on the physi- cal and administrative aspects of gov- ernment printing. THE GoVERNMENT PRINTING OFFICE According to the United States Code, "All printing, binding, and blank-book work for Congress, the Executive Office ... and every executive department, in- dependent office, and establishment of the government, shall be done at the Government Printing Office." The law then provides for certain exceptions which, as will be seen below, have re- sulted in the establishment of not one but more than three hundred and forty government printing plants. During fiscal year 1963 the GPO pro- duced a total of $127.1 million worth of printed matter. Analysis of the Annual Report of the Public Printer shows that $70 million of this (or 55 per cent) went for publications, the rest for forms, no- tices, postal cards, and other miscellane- ous printint, An additional $5.2 million of printing was done in six field plants of the GPO located in New York, San Francisco, and other cities. Reports do not allow a breakdown of this amount between publications and other printing. Assuming that the ratio was the same as for the main GPO plant, however, we have an additional $2.9 million for pub- lications, giving a total of $72.9 million. Unfortunately, it is impossible to translate expenditures for publication printing into number of publications produced.'The GPO, reflecting its orien- tation as "printer" rather than "publish- er," reports the number of copies of pub- lications printed, and even the number of pages, but not the number of different publications or titles. Administratively, GPO is a creature of Congress, not of the executive branch, and is specifically under the supervision of the Congressional Joint Committee on Printing, which consists of three Sena- tors and three Representatives. Legally, the President appoints its chief officer, the Public Printer, but actually the Joint Committee governs the appointment (the present Public Printer was formerly staff director of the Committee) and the President exerts no } control whatever over GPO operations. This curious ad- ministrative structure has long intrigued students of government and occasioned some criticism. In 1912 Senator Elihu Root noted: The GPO is not in any executive de- partment and has no supervision, except the supervision of Congress. . . . Now, either Congress ought to make its own supervision adequate, if it is going to per- form that duty, and create for it adequate machinery and fix upon somebody the re- sponsibility, or else it ought to put this bureau in an executive department. Which- ever Congress chooses to do is all right, but the office is today a lost child, and has been ever since I have known anything about the administration of the Govern- ment of the United States.7 This administrative pattern may have been justified a century ago. When Con- gress established the GPO in 1861 pri- marily in order to curb the graft and corruption which resulted from its previ- ous practice of contracting out printing to private firms, the majority of govern- ment printing was done for Congress rather th~n the executive branch. Today, however, only 12 per cent of GPO print- ing is Congressional, the remaining 88 per cent being done for the various exec- utive ~ncies. This anomalous admin- istratiVe legacy of the nineteenth cen- tury, however, is a primary cause of the present government printing problems and of the great gaps in the govern- ment's system of publication distribu- tion. 7 U.S. Government Printing Office, 100 GPO Years, 1861-1961 (Washington: GPO, 1961), p, 168. The Quiet Crisis in Government Publishing I 481 NoN-GPO PRINTING As noted earlier, the Government Printing Office in Washington actually is only one of more than three hundred and forty government printing plants scattered across the country and over- seas. The non-GPO plants, called "de- partmental" or "field" plants, are run by the various executive departments, agen- cies, bureaus, commissions, etc., and are completely unrelated to GPO. The development of these plants is part of the long and complex ~ist?ry of centralization versus decentrahzatwn of government printing, which need not be covered here. It is enough to say that they developed primarily as the re- sult of the vast expansion of the execu- tive branch under the impact of the New Deal World War II, the Cold War, and . the Space Age. Although other factors were present, they were authorized es- sentially because the GPO did not have the capacity to handle the tremendous executive branch printing requirements which stemmed from this growth. While precise figures are available for the amount of printing produced through GPO, no one really knows how much printing is done in the non-GPO plants. In 1954 the Hoover Commission noted: The total cost of printing done in these plants in the executive branch can only ~e guessed at. Estimates run from $100 mil- lion to $350 million annually. One govern- ment official estimated the expenditure at $250 million, another official as high as $350 million annually. Both officials were familiar with government printing expend- itures. The most authoritative current esti- mate of non-GPO printing is made by the staff of the Joint Committee on Printing. For fiscal year 1963, this source put the total at $106.4 million. 8 Com- bined with the GPO total of $127.1 mil- lion, this would mean that 46 per cent of a "Staff Report to the Joint Committee on Print- ing," April 1964 (mimeographed). all government printing for fiscal year 1963 was done outside GPO. We found that 55 per cent of GPO ex- penditure went for publications rather that other printing. If this ratio is ap- plied to non-GPO printing, we would have $58.5 million going for "publica- tion printing" in non-GPO plants. Given the $72.9 million for publication printing by GPO, this would mean that 44 per V cent of government publications are pro- duced outside the Government Printing Office. There is considerable evidence, however that because of the differences in types' of printing involved, the ratio is much higher in the non-GPO plants. In 1962, for example, the staff director of the Joint Committee on Printing esti- mated that from 60 to 65 per cent of all government publications were pro- duced outside GP0. 9 The matter is of some importance in the analysis of pub- lication distribution programs below. In order to have a base from which to analyze distribution, we will split the difference in these figures and assume that 55 per cent of all government pub- lications are printed in the non-GPO plants. It was noted above that the three hundred and forty-odd non-GPO print- ing plants were run by the various exec- utive agencies and were unrelated to GPO. Though technically true, in opera- tional terms this is a very misleading picture; these plants actually are under very tight Congressional control. The legal basis for this control is found in a 1919 law which empowers the Congres- sional Joint Committee on Printing "to adopt and employ such measures_ as, in its discretion, may be deemed necessary to remedy any neglect, delay, duplica- tion, or waste in the public printing and binding and the distribution of govern- ment publications." Under this broad power the Joint Committee authorizes (and abolishes) non-GPO plants and regulates their operation down to the o U.S. Congress, Senate Committee on Rules and Administration, Depository Libraries, Hearings, 87th Cong., 2nd Sess. (Washington: GPO, 1962), p. 21. 482 1 College & Research Libraries • November, 1965 smallest detail, such as requiring com- mittee approval before any plant can buy a "power operated paper cornering machine."10 These factors provide sufficient back- ground for an analysis of the curious pattern of government publishing, the most significant aspect of which is Con- gressional manipulation and tight con- trol over executive publishing. THE CoNGRESSIONAL "SQUEEzE" oN GOVERNMENT PUBLISIDNG The comparative cost of GPO versus executive plant printing apparently has never been investigated. The only hard data on this question known to the au- thor cropped up accidently in 1956 in- cidental to a House committee study of government paperwork. From data sub- mitted to the committee by the Navy Department, we find that 18 per cent of Navy printing in 1956 was done at GPO, 80 per cent was done in Navy printing plants, and 2 per cent was done on contract by commercial printers. The From the reports of the Public Printer 18 per cent of GPO printing, however, and the Joint Committee on Printing, we accounted for 54 per cent of the total know that only $15.2 million of the total Navy printing bill for the year. The government printing bill for fiscal 1963 2 per cent of commercial printing ac- was expended for Congressional print- counted for 30 per cent of the cost, and ing. This leaves $218.3 million, or 94 the 80 per cent done by Navy plants per cent, for printing by the various ex- constituted only 16 per cent of total ecutive departments and agencies. At cost.12 first glance this looks like another ex- The great cost differential shown here ample of the "swollen executive bu- does not mean that GPO is inefficient or reaucracy" which Congressmen frequent- that its charges are excessive. The prob- ly complain about. Considering the fact lem lies primarily in the type of print- that total expenses for Congress consti- J ing done in GPO as contrasted with tuted less than one-fiftieth of one per the executive plants. Most GPO printing cent of total federal expenditures for is by traditional letterpress methods, 1963, however, it is evident that Con- which are far more costly than the vari- gress-with its six per cent of printing ous offset or lithographic methods devel- expenditures-does not lack means of oped in the last thirty-odd years. GPO expression through the printed word. has begun to shift a bit recently toward The money for Congressional printing the more modern methods, but it has is appropriated directly to the Govern- been very slow to make the change, ment Printing Office, and all Congres- while very little of the printing done in sional printing is done at GPO. Execu- the newer executive plants is by letter- tive agencies, theoretically at least, may press. The resulting cost differential have their printing done at GPO or in gives executive agencies a strong motiva- their own departmental and field tion for attempting to avoid GP0.13 plants.11 In either case the cost must be Considerations of time and admin- paid from their appropriation for "print- istrative control point in the same direc- ing and reproduction." The agencies pre- tion. Congress is GPO's boss, and Con- fer to route most printing to their own gressional printing takes precedence. As plants, for reasons of cost and admin- istrative control. 10 U.S. Congress, Joint Committee on Printing, Government Printing and Binding Regulations, April 1, 1963, p, 8. 11 The executive agencies may contract out some printing to commercial printers, just as GPO also does, but the amount is limited and strictly con- trolled by the Joint Committee. 12 U.S. Congress, House Committee on House Ad- ministration, Paperwork Management and Printing Facilities in the United States Government, Part II, Report No. 2945, 84th Cong., 2nd Sess. (Washington: GPO, 1956), p. 130. 13 In fairness to GPO, it should be noted that GPO officials, while partially admitting the ex- istence of a cost differential, claim that the cost of printing in executive agencies is underestimated be- cause certain factors are not figured in. The Quiet Cr~is in Government Publ~hing I 483 the Public Printer put it in testimony before the House Appropriations Com- mittee, "the Printing Office was estab- lished primarily to serve the needs of Congress and this will always be our first consideration."14 GPO officials have also testified that printing the Congres- sional Record is their "single most im- portant job." The amazing job done on the Record under very difficult condi- tions probably makes it the world's prime example of printing speed and efficiency. At $98.00 a page, of course, the Record also is undoubtedly the most expensive printing job of its magnitude in the world. This emphasis on Congressional print- ing naturally means that printing for the executive agencies frequently is given short shrift. The Agricultural Re- search Service, for example, reported that the "average printing time" for its publications done by GPO was three to six months, while the average time in its own printing facility was one to two months.15 One agency official with sev- eral years' experience related to printing is convinced that a print priority list exists in GPO, "on which the Congress is at the top, the ,President is a poor sec- ond, and the rest of us are low men on the totem pole." Such factors tend to force executive printing into non-GPO plants, in which the agencies can set their own priorities. Thus a quiet bureaucratic war goes on over the placement of executive branch ptinting. The referee in this war is the Joint Committee on Printing, and the executive agencies fight a losing bat- tle. As background here, it is necessary too to note that GPO originally was set up to handle all government printing, and it did so until roughly the 1930's. During those years and since the GPO built up far more capacity than was 1 • Legislative Branch Appropriati