MICHAEL F. DiMARIO

Mr. Chairman and Members of the Committee, thank you for inviting me to present the views of the Government Printing Office (GPO) on public access to Government information in the 21st century. My views concern proposals to reform the public printing and documents statutes of Title 44 of the U.S. Code, GPO's authorizing law.

For more than a century GPO has served as the Government's printer and publications distributor, providing prompt and economical printing and reproduction services as well as comprehensive and equitable dissemination of the Government's information products that have public interest or educational value.

Today, some observers are advocating a major overhaul of Title 44, but there is no consensus on the direction reform should take. Some want to privatize all Government printing. Others want to decentralize printing and distribution authority. This is the case not only with the executive branch, but with the House of Representatives. Still others advocate a new Federal information policy based principally, if not solely, on electronic technology.

To date, four bills have been introduced in the House of Representatives to accomplish these goals, in different ways, although none has emerged from committee. A draft bill to establish an electronic Federal Depository Library Program (FDLP) has been circulated by the Office of Management and Budget (OMB) to several Members of Congress, including the leadership of this Committee, but it has not been introduced. The House Oversight Committee recently approved a House Document Management Plan that would essentially result in the transfer of those GPO congressional printing functions that support the House to the House itself. Some agencies have apparently decided to take matters into their own hands and have started using joint ventures with private sector publishers to produce their publications. As the Superintendent of Documents described to this Committee in the hearing on June 18, 1996, these measures are resulting in copyright or copyright-like restrictions on dissemination, undermining the central purpose of Title 44 to make public information publicly available.

Recently, the Justice Department's Office of Legal Counsel issued a memorandum opinion claiming that GPO's "extensive control" over executive branch printing and duplicating violates the constitutional principle of separation of powers. The opinion underscores the goal of the National Performance Review (NPR) to allow the executive branch to establish its own printing policy.

I do not agree with this opinion. I think it was wrongly decided, and I think it sends the wrong message to agencies and the public. GPO does not have "extensive control" over executive branch printing. We perform an administrative function to ensure that executive branch printing is performed economically and in the interests of the taxpayer, and to ensure that the publications are made available to the public on a comprehensive and equitable basis. By signaling that the Justice Department will not uphold the law, the message is being sent that there is no need for economy in public printing, and that providing effective public access is a secondary concern.

In September 1994, the Administration reached an agreement with Congress to maintain the status quo regarding printing and duplicating arrangements for fiscal year 1995 until the leadership of the congressional committees of jurisdiction and the Administration worked out a legislative approach to reform Federal printing. It specified a number of policies, including that all printing procurement should be conducted through GPO to the extent required by law, that existing agency inplant operations could continue but not expand, and that all publications required by law should be made available to the FDLP. Because no new legislative approach was developed, in April 1996 the White House Chief of Staff issued a memorandum that substantively reiterated the September 1994 agreement for the ensuing 12 months. I think this is a far more constructive approach to a possible reform of Title 44 than the one taken by the Justice Department.

Simply because a law has been on the books for many years does not mean that it is a "painful anachronism," as one witness recently remarked to this Committee. I agree that there is room for improvement to the statutes. I am well aware of the views of GPO's critics, and I agree that we need to take the opportunity now to make changes that will address these concerns, to the extent the changes do not conflict with the overall goal of achieving taxpayer savings and promoting public access. But I do not think that a sweeping overhaul of the laws is necessary or even desirable to achieve this. I think many changes can be accommodated within the current general structure of the law.

To those who want to privatize, I would remind them that GPO has moved between 75 and 80 percent of all printing we handle out the door to private contractors without changing the law. If Congress decides it wants the Congressional Record and other legislative documents turned over to the private sector, the law would not have to be changed, although Congress would have to be prepared to pay more and to relinquish a considerable amount of control.

While some changes to reflect the impact of electronic information technology are needed, I also think technology poses less of a challenge to Title 44 than the conventional wisdom suggests. Within the structure of the current law, we completely converted GPO's prepress processes to electronics, increasing productivity and lowering costs. That conversion made a vast amount of downsizing possible. Twenty-two years ago there were more than 8,600 employees at GPO, with more than 1,000 in our composition division; today there are about 3,800, with fewer than 500 involved with prepress work. With the move to direct-to-plate technology, there will be even fewer involved with prepress. The conversion also made it possible for GPO to move directly into CD-ROM premastering and online dissemination without massive development costs.

The nature of printing itself has changed dramatically over the past two decades. GPO has not only kept up with those changes but helped to lead many of them. We are continuing to change the nature of our operations, as our fiber optic links with Capitol Hill, our assistance to agency publishers and printers, and our training programs for agency personnel all demonstrate. This progress has been and can continue to be accomplished within the structure of the current law.

The single most important change to Title 44 we have needed was statutory clarification of our role in electronic information dissemination, and that was achieved, through the leadership of this Committee, with the passage of the GPO Electronic Information Access Enhancement Act of 1993 (P.L. 103-40). That law requires us to provide online versions of the Congressional Record and the Federal Register, and to work with Federal agencies to provide other information online. I think this was a wise way to structure GPO's electronic information dissemination authority, and experience is proving that point. Our implementation of the GPO Access service, which now is used by more than 2 million people every month, has shown that GPO can successfully participate in the electronic dissemination arena. We make a large volume of congressional information available, and an increasing number of executive agencies are approaching us to offer their products and to mount their Government Information Locator Service (GILS) records on the Access service.

As the recently concluded "Study to Identify Measures Necessary For a Successful Transition to a More Electronic Federal Depository Library Program" found, the Access service can now be used as the platform to move the FDLP into the electronic future, representing a major step forward in Federal information dissemination policy. Some adjustments to chapter 19 of Title 44 are needed to facilitate the transition, but I do not think that electronic technology itself necessarily spells the end of Title 44. As the FDLP study concluded, there is an important role for a centralized entity in a distributed system to assist the public in standardizing information, providing locator and pathway services, and coordinating permanent access.

There are other changes that could be made to the law to address specific requirements. I think the transfer of certain JCP authorities to the Public Printer would obviate the Justice Department's constitutional concerns. A change in GPO's funding mechanism to allow us to finance general administrative expenses from appropriations rather than through cost-recovery from our customers would put us on an equal footing with the way other agencies are financed and reduce printing charges to Congress and agencies. Limited changes to our sales authority, such as increasing our discount authority, would allow us to manage the sales program more effectively. GPO's name should be changed to recognize our involvement with a broad range of information dissemination strategies, and outdated or archaic provisions of the law--such as references to stereotype plates--should be dispensed with.

But the principal structure of Title 44--a centrally managed printing, printing procurement, and documents dissemination capability, providing localized service through a nationwide network of procurement offices, bookstores, and other operations--should remain in place. It simply saves too much money for the taxpayers and performs too valuable a service for public access to be done away with. In none of the proposals for change that have been offered to date are there the kind of benefits that Title 44 and GPO currently provide.

In our production plant, Congress has direct control over a cost-effective system for obtaining printing that is essential to the orderly conduct of legislative business. In our Printing Procurement Program, which works directly with the printing industry, the agencies have immediate access to one of the Government's most successful public sector/private sector partnerships for obtaining their printing needs at highly competitive prices. In GPO's Superintendent of Documents programs, the public has a single, highly visible system to provide comprehensive and equitable access to the vast wealth of information produced by the Government in both print and electronic format through depository libraries and over the Internet as well as through our sales program. Under the proposals I have seen to date, many of these benefits would be reduced if not lost.

With the growing use of electronics, there is the temptation to say that the Government no longer needs a printing capability. I think this temptation should be resisted. Last year, we produced over $800 million in printing services (and that is only about half of all the printing the Government produces), sold some $80 million worth of documents, and distributed tens of thousands of titles to depository libraries. We measured the printing of congressional publications alone at approximately 2 billion units, or copy-pages. Tax forms, press releases, passports, informational pamphlets and books, regulations, statutes, statistical data, and more--in printed form these documents represent a major avenue of communication between the Government and the public.

Moreover, printing remains an effective safeguard for ensuring that those without access to computers--and there are many--can still use Government information, and for guaranteeing permanence. The transition to full electronics is certainly coming, but it is a long way off. We need to manage that transition effectively, and maintaining a cost-effective printing and dissemination capability for the foreseeable future gives us an important management tool.

A major issue confronting Title 44 is a failure on the part of executive branch agencies to comply with the law. This Committee has already heard testimony on the consequences of increased agency control over their publications from GPO's Superintendent of Documents. Another problem is the growth of underutilized capacity in agency plants, and the natural tendency of agencies to try to keep that capacity occupied with work which otherwise should be procured through GPO. To ensure taxpayer savings and equitable public access requires vigilance by Congress over the performance of agency printing and dissemination capabilities. This hearing is a good example of what I am talking about.

There are other things Congress can do. It can require a review of executive agency printing plants. The review should not necessarily be designed to close down all agency printing capacity, as the printing industry has suggested. We know that there is a need in the agencies for a capability to produce quick turnaround printing to support administrative operations, and we do not have a problem with that. We also do not have a problem with certain agencies, such as the national security agencies, having the authority to produce their own printing inplant when doing so is clearly in the national or public interest.

The problem comes when capability established for simple internal administrative purposes is expanded to produce printing needs that are more cost-effectively performed through GPO's procurement program, and when agencies fail or refuse to provide publications produced internally for public distribution through GPO's programs. The objective of a review--and it could be conducted by OMB as well as Congress--should be to ensure that agencies are equipped with only the capacity necessary to fulfill either their own limited administrative needs or, as in the case of the General Services Administration (GSA), to support the limited internal administrative needs of multiple agencies grouped in close physical proximity through the central administrative support unit (CASU) or similar concept. GPO can assist in this review. Congress should also review the way some agencies have expanded their limited authorities, gained either through statute or by waivers from the JCP, into significant printing and dissemination operations that undermine the letter and spirit of Title 44.

In summary, I urge this Committee to critically examine the various proposals to amend Title 44 that have been advanced to date and to balance them against the public benefits that Title 44 currently achieves. Congress took a positive step forward when it enacted the "panoply of reforms," as one observer has put it, that constitute the current public printing and documents statutes. There are changes that can be made within the structure of Title 44 that would be responsive to many of the concerns that have been raised, but care needs to be taken that they do not cancel out the benefits achieved by the current system.

Privatization Proposals

Earlier, on February 2, 1995, in testimony before the House Subcommittee on Legislative Appropriations, former Public Printer Thomas McCormick, appointed by President Nixon, and former Public Printer John Boyle, appointed by President Carter, both disagreed that the concept of privatizing the printing GPO performs for Congress would generate savings.

Procurement of Congressional Printing. GPO fully supports the procurement of printing that is commercially procurable, e.g., purchasable from the private sector on the same terms (timeliness, quality, and control) at less cost. However, based on long experience with Government printing, we have serious concerns about whether any savings can be achieved by procuring congressional printing. We also have concerns about the level of control that could be retained over printing that is essential to the legislative process if it is commercially procured.

It has yet to be established that Congress can procure items such as the Congressional Record and related products from the private sector on the same terms as it now receives them, at reduced cost. The historical record, although dated, shows the opposite: that when the production of congressional work was removed from the private sector and turned over to the newly created GPO, Congress realized immediate and lasting savings on printing production.

The House recently voted to require GPO to study privatizing the Congressional Record and the Federal Register. The study, however, will not be able to fully disclose what a private sector contractor's costs would likely be. Without a clear decision by Congress to procure congressional work and the consequent expectation by potential contractors that they will be held financially and legally responsible for performing the work under the estimates they have submitted, any such cost estimate should be regarded as speculative.

The volume of congressional printing would preclude all but a few of the very largest printing firms in the United States from competing for this work, since the average firm employs just 20 people, according to printing industry estimates. Widespread competition for Government printing jobs is a key factor to achieving any savings in printing procurement. Without such competition, the potential to achieve savings from the procurement of congressional work is not likely, and in fact the costs of this printing would most likely rise above current levels.

If the sense is to break up congressional printing jobs and parse out this work to a wide variety of printers, economies made achievable by the consolidated operation of GPO's electronic prepress systems are likely to be overtaken by the increased administrative burden of managing multiple procured jobs among multiple contractors. The control of congressional printing would also be a problem under such conditions. Some observers have suggested that by changing the delivery requirements of congressional printing, it could be procurable. That is probably true, but changing those requirements would also substantially lower GPO's production costs as well.

Comparison of GPO and Private Sector Costs. GPO performs congressional printing at cost, while a contractor would charge cost plus a percentage for profit. There are allegations that GPO's costs are too high for the services performed, but these allegations are usually made within the context of comparing average printers' wages drawn from around the Nation to the wages paid to GPO employees in the Washington, DC, area.

These comparisons often reveal a great deal less than they purport to. GPO is not comparable to any other printing firm in the United States, based first on the types of work GPO is required to do (book and job work, newspaper printing, and a wide range of other work). Most printers specialize in one kind of work; GPO must maintain expertise in many. As a Government agency, GPO is required by law and regulation to maintain many expensive programs and operations (Inspector General, EEO, etc.) that private sector companies are not required to have. GPO's employees are entitled by law to organize and have union representation. Most private sector companies, because of their size, are unorganized.

Compared to other organized workplaces in the Washington, DC, metropolitan area (one of the highest cost-of-living areas in the Nation), GPO's wages are not wholly out of line. Compositors' wages at the Washington Post for both day and night shifts, for example, when averaged against GPO's are in fact slightly higher; this fact needs to be weighed against data indicating that an average issue of the Congressional Record contains 4 to 6 times the amount of type as in an average metropolitan daily newspaper. Wage comparisons for press and bindery work with other area printing wage contracts also indicate much smaller disparities than has been alleged. The charge that GPO's costs are wholly out of line with costs in the printing industry in general needs to be balanced against these factors.

Comparison of the Congressional Record to a Newspaper. There is little realistic comparability between the production of the Congressional Record and the average newspaper, regardless of the newspaper's delivery system. A large metropolitan daily newspaper will run approximately 60 pages each day, and may vary from 48 to 72 pages, a much more tightly defined and consistent variation from the median than the Record. With the exception of front pages held open for late breaking news, a daily newspaper follows a schedule of rigid and consistent deadlines. Most ad space, which takes up a vast amount of newspaper space, is preset, and journalists are frequently assigned column inches for their stories. As noted above, the average Record contains 4 to 6 times the amount of type as the average metropolitan daily newspaper, which contains a high percentage of camera copy ads, photos, and other work.

A variety of unique and complex requirements drive the production of the Congressional Record. GPO must print everything that happens on the floor of Congress each day in the Record, regardless of whether the printing totals 16 pages or 400 pages. We cannot assign column inches to Members; if a Member speaks at any length on a subject, if a Senator conducts a filibuster, GPO must print every word spoken. The printing cannot occur until Members have had a chance to "revise and extend" their remarks, as is their privilege. This frequently adds new material to be printed, much of which has not been captured electronically, such as newspaper articles and clippings. The Record must contain additional materials, such as the Daily Digest, that can be compiled only when the day's proceedings are ended. Print production cannot take place until Congress has adjourned for the day or GPO makes a decision to cut the proceedings so that at least some portion of the Record will be available when the Members reconvene. In all of these respects, the production of the Record is significantly different from the production of a newspaper. Indeed, apart from the fact that both products are printed on newsprint, there is little similarity at all.

Desirability of Procuring Congressional Printing. The overriding question that needs to be addressed is the desirability of contracting out printing needs which are essential to the conduct of the Nation's legislative business. Within the context of history, this represents less a movement toward reform than it does a revisitation of the past, when congressional printing was the exclusive province of private printers.

As history discloses, contractors frequently failed to perform on time or even to complete the work, and generated enormous costs (for their time). Congress remedied these problems with the creation of the GPO in close physical proximity and under close congressional scrutiny. Government performance of this critical work also parallels the performance by many other governments of their legislative printing both among the 50 States and around the world.

GPO has taken a variety of measures, including equipment back-ups, personnel contingency plans, and paper storage, to ensure that congressional work is delivered on time when it is needed. GPO has inclement weather emergency plans to ensure the plant is always staffed and operating when Congress is in session, and a variety of security measures are in place to prevent the unauthorized disclosure of highly important congressional work. In November 1995, at the specific request of the Senate Majority Leader and the Speaker of the House of Representatives, GPO kept its production staff on duty during the Government-wide budget shutdown to provide essential printing services. As will be recalled, most Government contractors were forced to cease operations during the budget shutdown due to lack of funding authority by the contracting agencies to continue operations. These factors must be weighed in any decision to remove the responsibility for this critical work from GPO.

House Document Management Plan. Recently, the House Oversight Committee approved a House Document Management Plan. The specific objective of this plan is to "decrease our reliance on GPO." In the resolution approving this plan, House officials are directed to work with GPO to develop a document management system, and we will provide all possible assistance. We are concerned, however, that this plan undermines the capability already resident in GPO to support all congressional printing.

The premise of the plan to provide the House with an electronic document management capability appears to overlook the initiatives GPO is undertaking to provide increased electronic document capabilities for all of Congress, particularly through the conversion of all congressional documents to Standard Generalized Markup Language (SGML). The plan apparently would rely on congressional staff for document inputting, formatting, and management, despite recent reductions in congressional staff. Also, the high turnover rate for congressional staff would tend to work against the effectiveness of training efforts to operate the document management system. It is unclear from the plan what would happen to existing electronic interfaces with the Appropriations Committee, the House Office of Legislative Counsel, House Information Resources, and the Office of Law Revision Counsel.

Where public dissemination is concerned, the document management system apparently would utilize the Library of Congress's THOMAS information system, leaving in question the provision of access to key House documents, including the House portion of the Congressional Record, through GPO's Access service, which is required by law to provide public online access to the Record.

National Performance Review

Earlier, during the 103d Congress, the House considered legislation that essentially would have implemented the recommendations of the NPR with respect to GPO. The recommendations were amended substantially before they were passed by the House, however, and the Senate did not take them up.

"Monopoly" Characterization is Inaccurate. Proposals for decentralizing Government printing in the executive branch are targeted at ending GPO's so-called "monopoly" on Government printing. However, a monopoly confers on an organization the means to exercise exclusive control over the provision of a specific commodity. GPO cannot be accurately characterized as exercising this kind of control over Government printing.

GPO is not the only organization providing Government printing. Only half of all Government printing, as measured by OMB in the U.S. Budget, is produced by or through GPO. There are at least 145 printing plants operated by other Federal agencies under the authority of waivers originally issued by the JCP pursuant to Title 44. In addition, agencies most likely also operate a number of other printing and duplicating facilities, as OMB has found in the past. Title 44 (and to a limited extent other legislation in section 207 of the Legislative Branch Appropriations Act) contain provisions for agencies to contract for their own printing, and there are also a number of agencies, such as the national security agencies, that are authorized by law to perform their own printing. The claim that GPO is a "monopoly," therefore, has no basis in fact.

It also has no basis because exclusive control over printing prices is denied to GPO in the area of printing procurement. The prices of the work performed by thousands of commercial printers for GPO are determined by the competitive forces of the marketplace, not by GPO, which adds only a marginal surcharge to cover the costs of procurement services. Most executive branch printing sent to GPO is purchased from the private sector, subjecting the vast majority of all Government printing to intensely competitive economic forces rather than any kind of monopolistic control.

Finally, the way we conduct business for executive branch customers undermines the claim that GPO is a "monopoly." GPO's mission is to fulfill the customer's printing needs. To do so we provide a capability to produce over 100 different products and services ("books" alone constitute only one such product category), and we maintain an extensive equipment line, a sophisticated range of graphic design services, a highly trained production and procurement workforce, a master bid list of a vastly diversified range of printing providers in the private sector, and specialists in marketing, sales, and other Government information mechanisms to help meet agency printing and dissemination requirements. The capability to provide such a diversified range of products and services--regardless of how uneconomical some of these capabilities may occasionally be to maintain--is not the hallmark of a monopoly, which typically bends customer demand to suit its needs.

Comparison of GPO and GSA as Procurement Models. The question has been raised whether agencies might not be better off procuring their own printing, just as they are able to procure items that they formerly were required to obtain through the GSA. However, the use of GSA as a model for comparison to GPO's procurement operation has a number of limitations.

GSA uses the Federal Acquisition Regulation (FAR) in conducting procurements. GPO utilizes its own Printing Procurement Regulation (PPR), which ensures a greater degree of flexibility and timeliness in printing procurements than the FAR. GSA buys "off-the-shelf"items, some of which are relatively easy to obtain through other channels. GPO, by contrast, is involved only in the procurement of printing, which is essentially a custom product since the specifications of each job vary widely, and it must be timely to have value.

Buying printing is not like buying paper clips. A knowledge of printing requirements and processes is essential to ensuring the best possible value. GPO printing specifications are developed by knowledgeable printing experts. There are cases in which agencies have ended up paying exorbitant prices for printing they have procured themselves. For example, the national news media recently reported that the Department of Labor incurred a $30,000 printing bill for copying services that would have cost approximately one-tenth that amount if procured with the same requirements through GPO, and approximately $500 if procured through GPO utilizing GPO-recommended cost-saving measures. Thus, comparing GSA's operation, which buys many things, with GPO's, which buys only printing, does not fully address the possible impacts of decentralizing printing procurement authority back to Federal agencies.

The decentralization of printing procurement authority is likely to significantly increase the costs of Federal printing. With such authority, agencies are likely to choose to produce much of their printing in-house, which several studies have shown is more expensive than procuring printing from the private sector. For those agencies that choose to procure printing, increased costs are also likely. It would be extremely costly for each agency to maintain the range of procurement services that GPO provides. GPO maintains a significant universe of competitors (10,000 private sector printers) that would be difficult for each agency to maintain. Without it, competition would decrease and prices would be likely to rise. In an environment of reduced competition, there are also likely to be increased opportunities for favoritism and corruption in Federal printing, twin evils the GPO was originally established to prevent.

GPO offers "one-stop-shopping" to printing contractors. Without it, the contractors would be compelled to increase their sales forces to search for contracting opportunities among multiple agencies, which would increase printing costs. They would also confront the difficulties imposed by having to deal with a multitude of solicitation formats, a problem that is overcome by GPO's standardized procurement solicitation packages. GPO offers a package of contracting services. We review requisitions and offer suggestions for economizing; develop specifications; compete, award, and administer contracts; perform press inspections and other on-site reviews to assure quality; perform quality control reviews utilizing a unique program that quantifies quality ranking factors that has become widely recognized throughout the industry; provide voucher examination and payment services; provide legal advice on contracting; and make available a dispute resolution service through GPO's Board of Contract Appeals. For every agency to provide a similar scope of services would result in increased printing costs.

In a centralized procurement system, GPO can ensure that Federal printing procurements are conducted in concert with pertinent statutory requirements, such as requirements for the use of recycled paper and vegetable inks, and can ensure that printed products are placed in the appropriate dissemination programs, such as the FDLP. For these reasons, a decentralized printing procurement program has several distinct disadvantages.

Justice Department Opinion. I do not concur with the view of the Department of Justice's (DOJ) Office of Legal Counsel that GPO has an "extensive control" over executive branch printing that renders our operations unconstitutional. We handle executive work in a purely ministerial capacity. We cannot refuse to fulfill an executive branch requisition for printing. We have no control over the editorial or informational content of executive branch publications. Our job is only to ensure that printing requisitions are fulfilled in the most cost-effective and timely manner possible, and to provide for the comprehensive and equitable dissemination of Government information to the public through copies of publications purchased with GPO funds.

These functions are performed under the authority of the Public Printer who, like members of the Cabinet, heads of executive branch agencies, and other high level Federal officials, is appointed by the President and confirmed by the Senate and serves solely at the pleasure of the President.

Under the law, the Public Printer maintains no veto authority over executive branch work. This view was expressed in a September 13, 1993, memorandum of the DOJ Office of Legal Counsel. At that time, the Office of Legal Counsel said that the requirement contained in section 207 of the Legislative Branch Appropriations Act [which was re-enacted and is currently at P.L. 103-283] for agencies to procure printing by or through GPO:

does not violate the separation of powers by delegating executive authority to the GPO...It does not give the GPO authority to refuse to print any materials, but merely requires that printing be procured "by and through" the GPO. Moreover, because 44 U.S.C. §1101 provides that "[the] Public Printer shall execute such printing and binding for the President as he may order and make requisition for," the executive branch retains its ability to ensure that materials are printed.

A March 11, 1994, memorandum of the Office of Legal Counsel reaffirmed this position. Since there have been no procedural changes or statutory modifications of any substance in this area of the law since that time, GPO's general counsel has requested the DOJ Office of Legal Counsel to reconsider this matter. We have also referred the Justice Department's memorandum to the general counsel of the GAO, which in a September 30, 1994, letter to then-JCP Chairman Ford held that contractors performing printing work directly for Federal agencies outside the authority of Title 44 and section 207 may not be paid from appropriated funds.

If the Justice Department's memorandum were followed, it would have strongly negative consequences for Government printing. Approximately 90 percent of all printing work performed by GPO is for the executive branch. Approximately two-thirds of all publications in the FDLP are executive branch publications, as are about 90 percent of all sales program publications. Assuming that executive branch printing policy would be characterized by decentralization as recommended by the NPR, the removal of executive branch printing from GPO would increase the cost of Government printing significantly. It would also increase the cost and reduce both the comprehensiveness and equity of current publications distribution policies, such as the FDLP, undercutting the congressional appropriations made for that purpose.

In July 1994, when President Clinton signed the requirement into law that the procurement of all printing, including duplicating, must with certain exceptions be performed through GPO, he expressed an intention to work toward Government printing reform. Subsequently, congressional leaders and representatives of the Administration worked out an agreement to maintain the status quo under Title 44 for the ensuing year while they worked together to produce a plan for Title 44 reform.

Under this agreement, they stated that agencies would: (1) use procurement as the preferred method of fulfilling printing requirements; (2) conduct all printing procurement through GPO to the extent required by law; (3) continue normal operation of existing agency inplant facilities without expansion; (4) continue to carry out plans downsize inplant capacity; and (5) ensure that all Government publications as defined in Title 44 are made available to the FDLP. The agreement was communicated to the heads of all executive branch departments and agencies in a memorandum dated September 19, 1994, by the Director of OMB, who stated that "agency compliance with these policies, and cooperation with congressional oversight, is essential to the ultimate success of a comprehensive initiative to reform Government printing."

On April 11, 1996, the White House Chief of Staff reiterated this agreement and its specifics in another memorandum to the heads of departments and agencies. This approach provides a far more constructive basis for reforming Title 44, one that emphasizes taxpayer savings and ensures public access, than the approach taken by the Justice Department.

Electronic Technology

GPO in fact already plays a significant and successful role in the electronic dissemination of Government information. For several years we have been a leading Government producer of CD ROM. In 1992, we received the annual award of the Special Interest Group for CD-ROM Applications and Technology (SIGCAT), the largest CD-ROM user group in the world. In 1993, the GAO cited our CD-ROM program as one of the most cost-effective in the Government, specifically noting that GPO's CD-ROM products are among the least expensive for users. That same year, Congress enacted the GPO Electronic Information Access Enhancement Act (P.L. 103 40). The act mandated GPO to provide online versions of the Congressional Record and Federal Register and authorized the Superintendent of Documents to work with Congress and Federal agencies to put additional information online.

The implementation of P.L. 103-40 has been enormously successful. Since we made the GPO Access service free to all users last December, we have averaged more than 2 million document downloads per month, and the number is growing; in June 1996, it reached 2.5 million. The service, which is now accessible through GPO's home page on the World Wide Web (http://www.access.gpo.gov.), currently features 58 databases, including a variety of agency GILS records, and it continues to grow. Some of the databases we prepare for GPO Access are also utilized by the Library of Congress for its THOMAS information service and by House Information Resources.

P.L. 103-40 statutorily recognized an electronic information role for GPO that we had been seeking for some time. The results of its implementation confirm that GPO can participate effectively in the electronic information dissemination arena. We just concluded a major study of how to convert the FDLP to a predominately electronic basis involving a broad range of participants from both the Government and the library and information communities. Among other things, the study found strong support for having a single entity in the Superintendent of Documents to coordinate library-related information dissemination activities in the electronic era.

Electronic Technology in Support of the Printing Process. GPO is often mistakenly characterized as an outmoded printer, when in fact the technologies used in the central office plant are state-of-the-art. GPO pioneered electronic prepress technologies in the Federal Government, and indeed throughout most of the printing industry, beginning in the 1960's. By the late 1970's, the production of most congressional publications had been converted to electronics, and in the early 1980's GPO concluded its phase-out of hot metal typesetting. Since then, GPO has gone through successive generations of automated composition technologies, each more powerful and less expensive than its predecessor.

Today, GPO's electronic prepress systems are characterized by a complex of direct electronic fiber optic linkages to a variety of congressional offices on Capitol Hill for data interchange. Since 1985, GPO has been connected to Congress with a 24-line fiber optic network. Only 6 of the lines are in use, but GPO deliberately oversized the network to handle Congress's expanding needs for electronic interchange.

Using this network, GPO currently receives a portion of Senate proceedings for the Congressional Record directly via fiber optic transmission, and we have begun receiving input from the House. Drafts of legislation destined to be introduced are received electronically from the Senate and House Offices of Legislative Counsel; the input is stored at GPO and can be accessed directly back on Capitol Hill for redrafting. A significant number of congressional committees are also linked directly with GPO's prepress systems for the preparation of hearings, committee prints, and other documents.

GPO's electronic systems make it possible to manipulate data and convey it from one publication to another, a capability that saves time, effort, and money. With this capability, bills, reports, and other documents can be conveyed for production in the Congressional Record, as required by congressional rules, without the need to rekey the data. Once inputted, the data from GPO's composition systems can also be converted for production in multiple medias, including print, CD-ROM, and online access.

GPO's electronic data capabilities are unique because of format standardization and the volume of its database storage capacity. Currently, GPO has 21 gigabytes of online storage, most of which is protected by RAID 5 technology, and an additional 120 gigabytes of near-line optical disk storage in the form of an 88-disk juke box. A second juke box ensures data protection. GPO's WAIS databases reside on a computer with 27.5 gigabytes of online storage. All databases are backed up regularly to ensure reliability.

With the increasing productivity of electronic information technologies supporting its printing systems, GPO has made significant strides forward in reducing the costs of congressional printing. In the past two decades, GPO's workforce has declined by more than 50 percent, virtually all of the reduction due to technology.

At the same time, costs have come down. Discounted for inflation, the cost per page of producing the Congressional Record is today only two thirds of what it was before the Record was converted to electronic prepress production, and we just recently reduced the Congressional Record page rate by 7 percent, primarily due to technology improvements. Similar savings have been achieved on other congressional products. As a result, for the past decade GPO has been able to hold a relatively flat line on appropriations for congressional printing. Continued technology upgrades, in addition to new technology approaches where appropriate, will continue to provide Congress with the best information services possible.

The Senate's print-on-demand capability is a good example of how electronic technologies have been used by GPO to satisfy a customer requirement for localized service. Two years ago, a joint effort of the Secretary of the Senate, the Senate Sergeant at Arms, this Committee, the JCP, and GPO was launched to provide a print-on-demand capability for the Senate Document Room. An electronic printing system was provided by the Senate and linked by fiber optic cables to GPO's automated prepress systems in the central office plant. GPO provides staff to operate the system for the Senate.

The print-on-demand capability now permits copies of bills, resolutions, and other official documents to be communicated electronically directly from databases resident at GPO to the Documents Room, where the exact number of copies needed can be produced on the spot. It also makes possible a reduction in the volume of copies that are originally printed at GPO, reducing printing costs, and it also reduces the paper storage and inventory requirements for the Documents Room.

Electronic Support for the Procurement Process. Electronic technology is also being used to improve the printing procurement process. Electronic versions of printing procurement bid solicitations are now accessible from the Internet via GPO's World Wide Web home page. We are currently posting solicitations from the central office and from a regional office in Hampton, VA, and we want to expand these postings to all regional offices. We estimate that less than $1,500 in programming was spent developing the system capability for making these solicitations available on GPO's home page. These costs were absorbed by current staff and no out-of-pocket expenses were incurred.

GPO did not need to undertake a large-scale effort to develop a system for electronically posting copies of bid solicitations. We developed our World Wide Web page originally as part of our mission to electronically disseminate Federal documents mandated by P.L. 103-40. To post bid solicitations, we simply added another electronic link to a printing procurement web page containing a list of currently available bid solicitations. A user need only click on the name of the desired document to initiate a download of the full text of the document. The technical expertise, as well as the computer hardware and software needed to accomplish this, was already in place.

The electronic posting of bid solicitations benefits the Government in several ways. First, it allows the GPO to reduce the cost of making this information available to the public. The estimated cost of preparing a typical multipaged bid solicitation for the web page is less than $1.00 per document. This cost is expected to drop further, because once a GPO term contract or other solicitation has been tagged it will not need to be tagged again when it is renewed. Approximately 90 percent of all term contracts are renewed. Making paper copies of solicitations available in response to walk-in and telephone requests is more expensive, ranging from $1.13 to $4.66.

Posting bid solicitations electronically also has the benefit of enabling more contractors to bid on Government printing jobs, thereby increasing competition and lowering procurement costs. Printed versions of bid solicitations are available to potential contractors and the public free of charge at any GPO bid room. However, only a limited number of copies are printed and, as a practical matter, access is limited to those printers located near GPO facilities. Other potential bidders therefore cannot gain ready access to bid opportunities except through reliance on proprietary bid services. Thus, for the vast majority of GPO contractors, most of whom are small businesses, the opportunities to review GPO bid solicitations are limited.

By posting electronic versions of these documents on the Internet, all potential bidders, even remotely located small businesses, will have immediate access to additional bid opportunities. We have found that increasing the dissemination of formal bid solicitations results in more contractors submitting bids. The increase in competition also results in a decrease in contract prices, lowering the overall cost to the Government for printing.

GPO's electronic posting initiative has generated considerable interest and enthusiasm in the printing industry. It has been praised by the Printing Industries of America, Inc., and by the GPO Contractors Coalition. In addition to the trade groups, individual printers have contacted GPO, both in writing and electronically, to voice their support for the program.

It is clear that the Federal Government will begin to rely more heavily on electronic commerce to meet its procurement needs. In October 1993, Executive Order No. 12,864 directed the Federal Government to explore and use electronic commerce in Federal procurement. This was followed by the Federal Acquisition Streamlining Act of 1994 (P.L. 103-355), which mandated, among other things, that the Federal Government begin using a computer network to disseminate solicitations in order to make Federal procurement more flexible, responsive, and timely. GPO's electronic posting of procurement solicitations is one way that we are using electronic commerce measures, consistent with the intent of both Administration and congressional efforts to streamline the Federal procurement process.

We are also using electronic funds transfer to pay those contractors who can receive payments electronically, and we are currently developing a pilot project with the Defense Printing Service to accept print orders filed electronically. In the future, we envision posting small purchases on the Internet and receiving price quotations electronically.

These and related developments can only be accomplished through office automation improvements in GPO's Printing Procurement Department. We currently have a request before the JCP to equip our procurement offices with computers, servers, and software. This equipment would end the Department's reliance on an obsolete mainframe-based system, reduce computer and telecommunications costs, and improve communications Department-wide and with customer agencies and contractors. The capabilities of this system could be used to implement additional electronic commerce initiatives that will increase competition, improve service, and lower the costs of the Printing Procurement Program.

OMB Draft Legislation. Recently, the Director of OMB transmitted to several Members of Congress proposed legislation titled "The Electronic Depository Library Act of 1996." We were provided with a copy of the proposal and analyzed its provisions. Copies of our analysis were transmitted to the Chairman and ranking Member of this Committee as well as other Members of Congress.

Several key provisions of OMB's proposal contrast sharply with the legislative direction suggested in the recently concluded study on the electronic transition of the FDLP. Although the current FDLP includes information from all three branches of Government, the OMB proposal would apply to the executive branch only.

In addition, the OMB proposal would potentially circumscribe the type of information that would be made available to depository libraries. It would transfer policy oversight for electronic dissemination to depositories from the Superintendent of Documents to agency Chief Information Officers (CIO's), who are accountable by law to OMB. It would allow agencies to make individual arrangements with depository libraries to pay for fee-based information services operated by agencies, and would set no limit on the fees, signaling a dramatic departure from the traditional principle of making Government information available to depositories without charge. It would effectively transform the National Technical Information Service (NTIS) of the Department of Commerce--which has no depository distribution authority and which could provide no guarantee of assuring equitable public access to any information it disseminates to depositories- into the executive branch's depository of final resort. It would also establish NTIS as the centralized point of access for executive branch information, in contrast to OMB's objection to the "centralized management model" currently operational in the FDLP. While the OMB proposal is purportedly designed to improve depository access to electronic information, the enforcement mechanism it provides would be weak and ineffectual.

The most prominent shortcoming of OMB's proposal lies in the provisions regarding the dissemination of online services. Agency CIO's would make direct, individual arrangements with depository libraries for access to online services "in some suitable electronic format." Online services would be required to be listed in each agency's information locator service so that libraries could find them.

This part of the bill seriously miscalculates the value and utility to the library community of having a single entity in the Superintendent of Documents to coordinate library-related information dissemination activities. Currently, GPO deals with hundreds of departments, agencies, bureaus, and offices, represented by approximately 4,500 billing address codes. To deal with a potentially vast number of online publishing entities in an increasingly distributed electronic environment, each of approximately 1,400 depository libraries would assume an enormous administrative burden without the presence of a central entity to coordinate services, provide notice of information availability, promote standards, develop guidelines on timeliness, and suggest user-friendly interfaces.

Providing equitable dissemination to all libraries, regardless of differing equipment capabilities in the library community, would be equally burdensome to agencies. The GPO Electronic Information Access Enhancement Act of 1993 was enacted in part to relieve agencies of this burden by establishing a single online service that would standardize service provision to all depository libraries. Since it is unlikely that many agencies would want or be able to provide differing levels of services, the result would be that only those libraries with the most comprehensive electronic capabilities would be able to access most Federal online systems.

To support its legislative proposal OMB stated their opinion that "the centralized management model of chapter 19 [of Title 44, U.S.C.] is not appropriate for distributed electronic information products and services." However, this is not the conclusion that was reached by the broad spectrum of participants in the FDLP study. The study confirmed the value of a central entity to assist libraries and the public in accessing electronic Government information in a distributed environment. Statutory improvements identified by the study could be attained through legislative revisions that would be less sweeping than those proposed by OMB, less burdensome on agencies, depository libraries, information users, and the taxpayers, and less damaging to the comprehensive and equitable system of access to Government information provided through the depository library program today.

Proposals for Change to Title 44

Also in need of change are provisions that would improve our capability to manage our sales program effectively. For example, we are not permitted to offer volume discounts of more than 25 percent, although the industry standard is 40 percent. Greater flexibility in managing this program could reduce costs and prices and improve dissemination to the public. However, we do not seek authority to "publish" Government information, as some observers have suggested.

Reform of Funding Mechanism. Another possible change is reform of GPO's funding mechanism. Outside of the small annual appropriation we receive for the Superintendent of Document's depository library, cataloging and indexing, by-law distribution, and international exchange programs, we are required by section 309 to recover all of our costs from the prices we set for our work, including costs for a number of functions that do not contribute to production and distribution but which every Federal agency is required to perform. Our printing rates must recover costs for our Inspector General's office, budgeting and accounting, personnel, equal employment opportunity, health and safety, security, data systems, and so on. Recovering these expenses increases the prices of GPO's products and services. Most agencies receive an annual salaries and expenses appropriation to cover these costs. With an appropriation to GPO for general administrative expenses, the costs charged to Congress and agencies for printing and related products would decrease.

Changes to Facilitate FDLP Transition. The recent study on the transition to a more electronic FDLP contained several suggestions for change to chapter 19 of Title 44 to facilitate that transition. The suggestions, which are currently under review by the library community, would provide explicit statutory recognition of GPO's FDLP role and responsibilities in the Information Age.

Transfer of JCP Authorities. Transferring certain JCP authorities to the Public Printer would address constitutional concerns about the role of the JCP. However, I also recognize that the JCP performs a valuable function for GPO as a referee over the printing priorities of the two Houses of Congress, and in providing policy oversight for printing and information dissemination issues, and I would not recommend its abolition altogether.

Oversight of Agency Inplant Capabilities. Congress needs to review the operation of printing plants in Federal agencies. While some of the work done in these plants is justified (e.g., work for the Central Intelligence Agency, Defense Intelligence Agency, and the National Security Agency, and short run work for limited internal administrative purposes only, which was the reason the plants were originally established), much of it could be procured far less expensively. Over the past 15 years, there have been a number of studies that compared the cost of performing procurable printing in agency plants with the cost of purchasing it through GPO. Most have concluded that it is far less expensive to procure through GPO. Also, printing procured through GPO is automatically eligible for dissemination to the public through the FDLP and sales programs. A review of agency plants that results in ensuring that they are limited to the performance of internal administrative printing that is outside the requirements for public distribution contained in the FDLP provisions of Title 44 would reduce the cost of Federal printing and improve public access to Government information through established dissemination programs.

Congressional Review of Other Authorities. As part of the effort to change Title 44, Congress needs to examine the authorities of other agencies to perform printing and dissemination functions that are being used to conflict with those that GPO is required to perform.

During the conduct of the recent study concerning the electronic transition of the FDLP, the Congressional Research Study compiled the various statutes authorizing the dissemination of Government information to the public. Approximately 400 statutes were identified, although only a handful are of real significance to GPO in terms of conflicting authorities. Most appear to be requirements to produce and/or disseminate information related to specific program authorities, a feature of law that became common in the 1960's and 1970's with the unwritten assumption that statutory information functions would be carried out in accordance with Title 44. The statutes that are being used to conflict with GPO's responsibilities are those governing the operations of the National Technical Information Service (NTIS) of the Department of Commerce, those that authorize the Federal prisons to perform Government printing (Federal Prison Industries, or UNICOR), and laws that authorize printing for other Federal agencies by the General Services Administration (GSA).

In each of these cases, statutes were originally enacted for sound but limited public policy purposes--to provide for the dissemination of scientific and technical information resulting from Federal research to American business (NTIS), the use of Federal prisoners to perform labor intensive work to keep them occupied (UNICOR), and the performance of administrative printing as a centrally-located service for Federal agencies physically located together in the field (GSA). However, in recent years operations conducted under these authorities have been expanded beyond their original intent to conflict directly with GPO's statutory mission.

NTIS is using its authority to establish printing and dissemination arrangements for a wide range of information products, resulting in copyright or copyright-like restrictions on the information that was originally generated at taxpayer expense. UNICOR has invested in long-run printing equipment as a revenue-generating measure, equipment that paradoxically leads to idle rather than busy prison labor. GSA has established cross-servicing arrangements for its printing and duplicating plants to assist agencies in some cases with a wide variety of printing, expanding beyond its original central administrative service unit (CASU) authority.

These operations remove work from the GPO procurement stream and frequently result in decreased public access when the publications are not made available to the Superintendent of Documents for dissemination. Congress needs to review the operation of these authorities to ensure that they are staying within the original legislative intent. That would be a major contribution to reform of the printing statutes.

Financial Condition. Apart from the potential revision of GPO's funding mechanism, a major revision of the law is not necessary to address the financial situation at GPO. As you know, GPO has experienced a series of year-end financial losses since 1991, due principally to declining revenues. The losses were funded out of retained earnings in GPO's revolving fund and did not require additional funding from Congress. Revenue declines have been the result of less work coming in, due primarily to budget cuts in the agencies but also to the increasing use of electronics and the production of work by agency plants. They were also the result of a printing rate structure at GPO that remained unchanged from January 1990 until just recently, when we worked with the JCP to establish a pricing structure that fully recovers costs.

In response to this situation, GPO has moved to reduce expenses. Since I took office in 1993, we have reduced personnel levels by more than 1,000, generating an estimated savings of $50 million, and total employment at GPO is now lower than at any time in the post World War II period. In fiscal 1995, we reduced the loss by year end to $3 million on $852.6 million in revenues, for a margin of less than one-half of one percent. However, due to continuing decreases in workload, resulting in part from two Government shutdowns and several snow days, combined losses were $17 million through May 1996.

We will continue to downsize and adjust operations to meet workload levels. The House of Representatives has directed us to reduce to 3,600 full-time equivalent (FTE) work years by the end of fiscal year 1997, and we will achieve that goal. Other cost-saving and downsizing actions can also be taken within the context of current law. Apart from a revision that would permit us to fund general administrative expenses from appropriated funds, Title 44 provides GPO with sufficient financial tools to meet the current situation.

* * *

Beyond these measures, I think Title 44 is a good set of laws for ensuring economy in printing and for providing comprehensive and equitable public access to Government information. With some changes, it would be broad enough and flexible enough to accommodate the transition to new information technologies, and it currently is sufficient to permit adjustment to reduced printing workloads. Focussed adjustments to Title 44, in combination with congressional review of other authorities and continued downsizing and technological transition at GPO, would provide a realistic and achievable program for improving operations without endangering the ability to generate economies in Government printing and ensure public access.

Questions or comments regarding this service?
Contact wwwadmin@gpo.gov

Page #public-affairs/rules.html March 19, 1997