of the importance of computer operations to the mission of an organization, maintenance contracts often provide for maintenance personnel to be physically located on-site. Maintenance by Government personnel is not normally undertaken unless service is not readily available from a contractor or the need for maintenance occurs at remote or combat installations. Users prefer on-site maintenance because of the importance of computer operations to mission accomplishment. On-site maintenance can be inefficient and expensive because when the computer is operating, maintenance personnel are underutilized. Conversely, when the computer is down, computer operators are idle. A more cost-effective approach might be the combined use of maintenance and operating personnel. This could be accomplished by contract or by use of Government employees. Maintenance personnel would be on-site at all times. When they were not doing maintenance, they would operate the computer or perform related functions. Even when the equipment is rented, combined operator-maintenance should be considered. Software Comptroller General Report B-115369 17 presents a broad review of the present policies and practices in the acquisition of computer software by individual agencies. It makes broad recommendations regarding the formulation of a master plan for the acquisition and use of software in addition to the structure needed to implement that plan. The findings of GAO relative to repeated acquisition of the same software package without benefit of quantity discounts are significant. GAO noted that private industry generally employs a single purchaser for the total corporate entity. This is feasible because in private industry each corporation generally is oriented toward a single mission. However, in the Government the complete centralization of procurement of computer software would 17 U.S. Comptroller General, Report B-115369, Acquisition and Use of Software Products for Automatic Data Processing Systems in the Federal Government, June 30, 1971. be difficult to achieve because of significant differences in agency missions. The benefit of substantial discounts may be available through improved procurement management within an agency and in interagency coordination. As GAO noted, the Marine Corps centralized its procurement of software and acquired two standard software packages in multiple copies at a substantial discount. Although software standardization is increasing, many noncompatible systems are still on the market. The delay required to ascertain compatibility can be expected to take at least 20 working days; this is now the time required to process an agency's request for delegated authority to procure software. It could be 130 calendar days; this is now the time required for advance notice for the release of equipment in Government inventory. In either case, the delay of one to three months can be a highly restrictive factor in software procurement. Personnel and associated costs involved in establishing and maintaining a central catalog of software, or of other products, also must be considered. In addition, manufacturers would develop new pricing structures on programs previously sold as single or limited system-use products. For example, if GSA were to advise a manufacturer that it would purchase a single restrictive or licensed software product which would then be made available to all Government users, it would be necessary for the manufacturer to raise his price in order to recover all development costs on the single sale. It is more reasonable to envision an extension of the present multi-award contracts for computer software with a provision for the Government to receive the benefits of quantity discounts based on actual purchases of the product concerned. ADP Supplies In defining the authority of GSA in implementing FPASA, the Comptroller General ruled: exclusive authority to GSA to procure all general purpose ADPE and related sup plies and equipment for use by other Federal agencies. 18 (italics added) Strictly interpreted, this decision requires agencies to obtain a waiver in order to procure ADP supplies. GPO has operated under a continuing waiver from GSA to contract for all marginally-punched continuous forms. Paper supplies in the form of tabulating cards, punch-paper tape, and continuous forms are unique in the data processing field because they are standard. Since the early days of tabulating cards, there has been a single standard size for punched cards; and early in the development of high-speed printers, manufacturers standardized the sizes of printed forms. This has been done to such an extent that today only forms designed for particular optical-character or optical-mark reading applications are machine-dependent. Work is now underway at NBS to correct this. The distinction between ADP paper supplies and general paper supplies is rapidly disappearing. Agencies currently use manually prepared checks, which are later processed through a data processing operation, as well as plain typewritten pages that both communicate with people and are machinereadable. Limiting authority to GSA for the procurement of ADP supplies unduly complicates their acquisition and appears to be unwarranted. National Policy GSA, NBS, and OMB all have leadership responsibilities for the Government-wide acquisition of ADPE. A Federal ADP users task force found that: .. the legislative branch and central executive agencies have generally confined their principal concerns with ADPE hardware and software to maximize economy in its acquisition and efficiency in clock time utilization, with only a residual concern with more effective use of the ADP technology in terms of executive branch missions and functions. 18 U.S. Comptroller General Decision B-151204/B-157587, Jan. 10, 1969, appendix D. These three agencies seem to have construed their authority and responsibilities for ADP quite narrowly, being concerned almost exclusively with property management . . . Telecommunications and data processing are treated separately from ADPE equipment almost everywhere including the central agencies. The technology and new applications in many agencies clearly demonstrate the emerging interdependence of the two, and the need to consider common management of both in the future. [The Government is not motivated, equipped, or structured well for development of the most effective use of ADP; and few of the presently constituted Government-wide ADP entities have demonstrated either the necessary capability or intentions to lead the charge in this area, which represents about 70 percent of total ADP costs. Executive agencies seem to have been so diverted in purpose or constrained by externally imposed concern with direct economy of ADP acquisitions and operations that they do not feel free, and therefore fail to pursue the best use of ADP technology.] 19 It should be recognized that substantial savings in ADPE procurement must be realized to offset the administrative costs that are now associated with it. Elaborate controls and approval processes that have been built up are stifling the use of computers in new applications and have stretched the computer acquisition cycle to well over two years. A frequent result is that the Government buys bigger computers than it needs because the computer is obtained to handle a sophisticated workload, which may be only a small part of the total workload. The technology exists to give users access to sophisticated computers, but smaller computers can be used for routine tasks. Economy is measured only in terms of the speed of computers, rather than the efficiency of the total data processing task. The large hardware costs are tied to the large computers, with about 10 percent of the Government computers representing more than 19 U.S. Interagency Committee on Automatic Data Processing Equipment, Report of a Task Force on Long-Range Plans for Automatic Data Processing Equipment in the Federal Government, May 1971. 50 percent of the total purchase costs. (See Appendix E.) The scope and complexity of Government acquisition and operation of ADP systems warrant national attention regarding standards for ADP, visibility and effective use of total resources, and coordination of acquisition of new or additional resources. Current legislation provides an effective mechanism for assuring necessary management attention. Implementation of this legislation, while alleviating some of the problems existing prior to enactment of the current version of FPASA, appears to have administratively stifled the overall program by preoccupation with procedures directed toward improving acquisition price. those used in the acquisition of other equipment and systems could not be determined, nor could it be justified. There are many problems associated with Government-wide procurement of computer software. The extension of Federal Supply Schedule contracts for software with provisions for quantity discounts should be explored. Vesting Government-wide purchase authority in GSA for all ADP supplies is not nec essary. Adequate consideration has not been given to the increased costs and delays arising from complex procurement procedures. Conclusions National management of ADP resources by GSA is a special responsibility. It does not require that GSA directly procure computers for use by a single agency. The present rules for delegating ADPE procurement authority to agencies are overly restrictive. Clear and practical procedures are essential for effective planning and optimum use of manpower. Agencies that have the necessary procurement expertise should be authorized to acquire their own systems except when centralized procurement by GSA is clearly more cost-effective. The ADPE Fund has not been capitalized adequately to take advantage of "best buy" situations. Multi-year leasing would permit the Government to benefit from potential cost savings and could relieve the problem of maintaining and disposing of obsolete ADPE. The policies and procedures for evaluating alternatives in fulfilling an agency's ADPE requirements are generally effective, but complexities in the approval procedures may account for the present lack of adequate consideration of services as an alternative. The current lack of visibility regarding available commercial and interagency services reduces effective use of services as an alternative. The basis for use of late proposal provisions for the acquisition of ADPE that differ from FOOD In fiscal 1971 the production, processing, distribution, and preparation of food in the United States was a $140 billion a year business and represented about 14 percent of our gross national product (fig. 2). Statutes designed to help or to protect the public consumer involve the Government in many aspects of the entire business. No other commercial product is so inherently and necessarily subject to the degree of production, packaging, and quality controls by all levels of Government. No other industry in recent years has been involved in as many congressional hearings, consumer protection activities, and controversies regarding the effectiveness of industry-related Federal programs. Recommendation 16. Assign responsibility for consistent and equitable implementation of legislative policy concerning food acquisition to the Office of Federal Procurement Policy or to an agency designated by the President. Recommendation 17. Establish by legislation a central coordinator to identify and assign individual agency responsibilities for management of the Federal food quality assurance program. The Commission's review of Government food acquisition was undertaken in response to congressional, industrial, and user interest. Many studies, investigations, congressional Federal Food Market The Government spends about $4.6 billion annually in the food market. As noted earlier, the Government also regulates the industry through measures designed to prevent disease, improve sanitation, establish quality standards, and raise quality levels. Federal regulatory provisions have evolved over more than 60 years. State and local laws have been enacted to accomplish similar objectives. Until recently the growth of these programs was not coordinated, but in the last decade increasing attention has been given to cooperative Federal-State efforts. These efforts recognize the State's ability to handle regulatory and inspection functions when State laws establish standards equal to, or stricter than, those established at the Federal level. Using Federal funding to supplement the State funds, the State's inspection activities can be improved, with the potential for eliminating duplicate effort. As there is no standard procedure for Federal agencies to report purchases by commodity, the exact amount the Government spends on food is indeterminable. The figures in table 1 of food procurement by Federal agencies are program managers' estimates and contain some duplication where interagency support is used for procurement or distribution. TABLE 1. FEDERAL ACQUISITION OF FOOD, FISCAL 1971 hearings, and writings by industry, agencies, and GAO preceded our review. Therefore, the Commission decided to limit its effort to an analysis of data readily available, to make field visits to become familiar with the industry, and to perform cost studies to compare operational concepts. Results of the cost studies are in Chapter 6. A review of basic legislation is essential to an understanding of the interrelationships among Federal agencies relating to specifications, standards, quality assurance, distribution policies, and procurement techniques. Food legislation is the result of efforts by many congressional committees to resolve specific problems as they arose, and there is no coordinated legislative policy for uniform application by executive agencies. Because DOD and USDA are the two major food purchasers, they can adversely affect each other's programs. Timing of the purchases of similar items can affect the market price of all Government procurements. Commodities procured by one agency can result in a shortage for another. Attempts have been made by the two agencies to coordinate the accomplishment of their respective missions. To date, coordination consists of informal liaison between commodity specialists of the two agencies; this arrangement was recognized in writing in 1966.20 DOD specialists receive notice of USDA market actions three to four weeks in advance of formal announcements. The Defense Personnel Support Center (DPSC) is pleased with this notification and regards it as a definite asset to its operations. DSA headquarters 21 cited continuing efforts to achieve further coordination. It recognizes the potential role of DOD in stabilizing prices in the domestic food market, and on the advice of USDA buys items in plentiful supply and does not buy items in short supply. However, lack of acquisition system flexibility and responsiveness reduces the effectiveness of this coordination. DOD stock purchases of food are based on a 120-day leadtime from requisitions developed to satisfy menu requirements. Except for purchases through the commercial distribution system, this long leadtime reduces opportunities to use large DOD purchases for market stabilization purposes. Efforts should 20 Memorandum from the U.S. Department of Agriculture, Office of the Secretary, to the Defense Personnel Support Center, Feb. 11, 1968. 21 Briefing on "Foods" presented to Study Group 13A by the Defense Supply Agency, May 25, 1971. continue toward this goal between the two largest Federal food purchasing agencies. Distribution Systems The food industry probably operates the largest and most widespread commodity distribution system. The size, scope, and commodity commonality of the commercial system have made it one of the most competitive. As a result, the options available to provide food at the point of need are almost unlimited. In this environment, every Federal activity that buys food for its own use or for use outside the Government should evaluate commercial distribution alternatives in order to optimize procurement economy and efficiency. The distribution systems currently used by the various agencies do not adequately consider total economic cost. DOD is the largest user of Federally procured foods and uses a combination of systems to fill requirements that range from troop feeding to commissary resale. Most DOD food procurement is for resale items bought by commercial description. With the exception of some local call contracts for perishables, troop issue acquisition systems historically have considered volume purchases as the primary principle, with quality controlled by detailed specifications. This concept requires elaborate research laboratories, specification libraries, and Government distribution systems. Since the troop issue portion of the total food market is very small, competition in this special market is reduced to those processors willing to comply with DOD procedures and conditions. Apart from the problems with specifications and contract terms discussed elsewhere in this report, some problems arise because of specific statutory requirements. Federal contracts over $10,000 require adherence to the Walsh-Healey Public Contracts Act,22 which imposes wage and labor standards on contractors. Food that is not produced using these labor standards cannot be sold to the Government, although less expensive but equally wholesome food produced under other Federal statutory standards 22 41 U.S.C. 35-45 (1970). |