Order Code 94-261 CRS Report for Congress Received through the CRS Web Intelligence Spending: Public Disclosure Issues Updated September 25, 2006 Richard A. Best, Jr. Specialist in National Defense Foreign Affairs, Defense, and Trade Division Elizabeth B. Bazan Legislative Attorney American Law Division Congressional Research Service ˜ The Library of Congress Intelligence Spending: Public Disclosure Issues Summary Although the United States Intelligence Community encompasses large Federal agencies — the Central Intelligence Agency (CIA), the Defense Intelligence Agency (DIA), the National Reconnaissance Office, the National Geospatial-Intelligence Agency (NGA), and the National Security Agency (NSA) — among others — neither Congress nor the executive branch has regularly made public the total extent of intelligence spending. Rather, intelligence programs and personnel are largely contained, but not identified, within the capacious budget of the Department of Defense (DOD). This practice has long been criticized by proponents of open government and many argue that the end of the Cold War has long since removed any justification for secret budgets. In 2004, the 9/11 Commission recommended that “the overall amounts of money being appropriated for national intelligence and to its component agencies should no longer be kept secret.” The Constitution mandates regular statements and accounts of expenditures, but the courts have regarded the Congress as having the power to define the meaning of the clause. From the creation of the modern U.S. Intelligence Community in the late 1940s, Congress and the executive branch shared a determination to keep intelligence spending secret. Proponents of this practice have argued that disclosures of major changes in intelligence spending from one year to the next would provide hostile parties with information on new program or cutbacks that could be exploited to U.S. disadvantage. Secondly, they believe that it would be practically impossible to limit disclosure to total figures and that explanations of what is included or excluded would lead to damaging revelations. On the other hand, some Members dispute these arguments, stressing the positive effects of open government and the distortions of budget information that occur when the budgets of large agencies are classified. Legislation has been twice enacted expressing the “sense of the Congress” that total intelligence spending figures should be made public, but on several separate occasions both the House and the Senate have voted against making such information public. The Clinton Administration released total appropriations figures for intelligence and intelligence-related activities for fiscal years 1997 and 1998, but subsequently such numbers have not been made public. Legal efforts to force release of intelligence spending figures have been unsuccessful. Central to consideration of the issue is the composition of the “intelligence budget.” Intelligence authorization bills have included not just the “National Intelligence Program”–the budgets for CIA, DIA, NSA et al, but also a wide variety of other intelligence and intelligence-related efforts conducted by the Defense Department. Shifts of tactical programs into or out of the total intelligence budgets have hitherto been important only to budget analysts; disclosing total intelligence budgets could make such transfers matters of concern to a far larger audience. Legislation reported by the Senate Intelligence Committee in May 2006 (S. 3237) would require that funding for the National Intelligence Program be made public but it does not address other intelligence activities. Earlier versions of this Report were entitled Intelligence Spending: Should Total Amounts Be Made Public? This report will be updated as circumstances change. Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 What Constitutes the Intelligence Budget? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Past Budgetary Practice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 Authorization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Appropriations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 The Question of Disclosure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Policy Arguments, Pro and Con . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Constitutional Questions Related to Disclosure of Aggregate Intelligence Budget Figure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 History of the Constitutional Language . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 Judicial Interpretation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Conclusions Regarding Statement and Account Clause . . . . . . . . . . . . . . . 31 Post-Cold War Developments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Recommendations by the 9/11 Commission and Subsequent Legislation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Intelligence Spending: Should Total Amounts Be Made Public? Introduction Since the creation of the modern U.S. intelligence community after World War II, neither Congress nor the executive branch has made public the total extent of intelligence spending except for two fiscal years in the 1990s. Rather, intelligence programs and personnel have largely been contained, but not identified, within the capacious expanse of the budget of the Department of Defense (DOD). This practice has long been criticized by proponents of open government. The intelligence reform effort of the mid-1970s that led to greater involvement of Congress in the oversight of the Intelligence Community also generated a number of proposals to make public the amounts spent on intelligence activities. Many observers subsequently argued that the end of the Cold War further reduced the need to keep secret the aggregate amount of intelligence spending. According to this view, with the dissolution of the Soviet Union, there are few foreign countries that can take advantage of information about trends in U.S. intelligence spending to develop effective countermeasures. Terrorist organizations, it is argued, lack the capability of exploiting total intelligence spending data. In recent years, proposals for making public overall totals of intelligence spending have come under renewed consideration. In 1991 and 1992 legislation was enacted that stated the “sense of the Congress” that “the aggregate amount requested and authorized for, and spent on, intelligence and intelligence-related activities should be disclosed to the public in an appropriate manner.” Nevertheless, both the House and the Senate voted in subsequent years not to require a release of intelligence spending data. During the Clinton Administration, Director of Central Intelligence (DCI) George Tenet twice took the initiative to release total figures for appropriations for intelligence and intelligence-related activities. Despite the release of data for fiscal years 1997 and 1998, however, no subsequent appropriations levels have been made public. The issue has not, however, died. The 9/11 Commission, in its final report, recommended that “the overall amounts [or the “top line”] of money being appropriated for national intelligence and to its component agencies should no longer be kept secret. Congress should pass a separate appropriations act for intelligence, defending the broad allocation of how these tens of billions of dollars have been assigned among the varieties of intelligence work.”1 A number of proposals for Intelligence reform legislation in 2004 included provisions for making the budget 1 U.S., National Commission on Terrorist Attacks Upon the United States, The 9/11 Commission Report (Washington: Government Printing Office, 2004), p. 416. CRS-2 public, but the legislation ultimately enacted as the Intelligence Reform and Terrorism Prevention Act of 2004 (P.L. 108-458) [hereafter referred to as the Intelligence Reform Act] did not include provisions for making budget numbers public. More recently, the FY2007 Intelligence Authorization legislation (S. 3237) reported in the Senate in May 2006 would require publication of budget totals for national, but not tactical, intelligence programs. This report describes the constituent parts of the intelligence budget, past practice in handling intelligence authorizations and appropriations, the arguments that have been advanced for and against making intelligence spending totals public, a legal analysis of these issues, and a review of the implications of post-Cold War developments on the question. It also describes past congressional interest in keeping intelligence spending totals secret. What Constitutes the Intelligence Budget? The meaning of the term “intelligence budget” is not easily described. Although some may assume it is equivalent to the budget of the Central Intelligence Agency, in actuality it encompasses a wide variety of agencies and functions in various parts of the Federal Government that are involved in intelligence collection, analysis, and dissemination. At the same time, some important information collection efforts (such as reporting by U.S. embassies to the State Department) are not considered as intelligence activities and their funding is not included in the intelligence budget. A further complication, to be addressed below, is the separate category of intelligencerelated activities undertaken in DOD that are included in overall intelligence spending categories. For some purposes, it is sufficient to describe intelligence and intelligence-related activities as those authorized by annual intelligence authorization acts. In the context of annual budget reviews, both the executive branch and Congress have sought a comprehensive overview of all intelligence collection systems and activities. Thus, there emerged the concept of an intelligence community, not a monolithic organization but a grouping of governmental entities ranging in size from the CIA and NSA down to the small intelligence offices of the Treasury and Energy Departments. Except for the CIA, this community consists of components that are integral parts of agencies that are not themselves part of the Intelligence Community and their budgets are subject to separate authorization processes. Thus, for instance, the State Department’s Bureau of Intelligence and Research is both part of the Intelligence Community and an organizational component of the Department of State. Its budget is considered as part of the overall intelligence budget and as a component of the State Department budget. Similar situations apply, on a much larger and expensive scale, in the Defense Department. Since these intelligence components are closely tied to their parent departments and share facilities and administrative structure with them, it is not always possible to desegregate intelligence and non-intelligence costs with precision. For the purposes of this discussion, the U.S. “intelligence budget” is considered to consist of those activities authorized by the annual intelligence authorization acts, CRS-3 viz. the intelligence and intelligence-related activities of the following elements of the United States government: (1) the Central Intelligence Agency (CIA); (2) the National Security Agency (NSA); (3) the Defense Intelligence Agency (DIA); (4) the National Geospatial-Intelligence Agency (NGA) (formerly the National Imagery and Mapping Agency (NIMA)); (5) the National Reconnaissance Office (NRO); (6) the intelligence elements of the Army, Navy, Air Force, and the Marine Corps (7) the State Department’s Bureau of Intelligence and Research (INR); (8) the Federal Bureau of Investigation (FBI); (9) the Department of Homeland Security (DHS); (10) the Coast Guard; (11) the Department of the Treasury; (12) the Department of Energy; (13) the Drug Enforcement Administration (DEA). The parameters of the intelligence budget are, to some extent, arbitrary. Lines between intelligence and other types of information-gathering efforts can be fine. As noted earlier, reporting by the State Department’s Foreign Service Officers is an invaluable adjunct to intelligence collection, but is not considered an intelligence activity. Similarly, some reconnaissance and surveillance activities, mostly conducted in DOD, are very closely akin to intelligence, but for administrative or historical reasons have never been considered as being intelligence or intelligencerelated activities per se. The intelligence budget as authorized by Congress is now divided into two parts, the National Intelligence Program (NIP) and the Military Intelligence Program (MIP). NIP programs (formerly categorized as the National Foreign Intelligence Program (NFIP)) are those undertaken in support of national-level decision making and are conducted by the CIA, DIA, NSA, the NRO, NGA, and other Washington-area agencies. MIP programs are those undertaken by DOD agencies in support of defense policymaking and of military commanders throughout the world. Until September, 2005, there were two sets of programs within DOD–the Joint Military Intelligence Program (JMIP) and Tactical Intelligence and Related Activities (TIARA). JMIP programs, established as a separate category in 1994, supported DOD-wide activities. TIARA programs were defined as “a diverse array of reconnaissance and target acquisition programs which are a functional part of the basic military force structure and provide direct support to military operations.”2 In recent years the overlap among intelligence and intelligence-related activities has grown — satellite photography, for instance, can now be made immediately available to tactical commanders and intelligence acquired at the tactical level is frequently 2 U.S. Congress, House of Representatives, 103d Congress, 1st session, Permanent Select Committee on Intelligence, Intelligence Authorization Act for Fiscal Year 1994, Report 103162, Part 1, June 29, 1993, p. 12. CRS-4 transmitted to national-level agencies. As a result, JMIP and TIARA were combined by the Defense Department into the MIP in September 2005. Within the MIP are programs that formerly constituted the JMIP that support DOD-wide intelligence efforts as well as programs directly supporting military operations that were formerly categorized as TIARA. The relationship of intelligence-related programs to regular intelligence programs is a complex one that is not likely to be understood by many public observers. In 1994, then-DCI R. James Woolsey described them as a “loose amalgamation of activities that may vary from year to year, depending on how the various military services decide what constitutes tactical intelligence.”3 Intelligence-related programs, which may constitute somewhere around a third of total intelligence spending, are integral parts of defense programs; in many cases they are also supported by non-intelligence personnel and facilities. (The administrative expenses, for instance, of a military base that has intelligence-related missions as well as non-intelligence functions would probably not be included in intelligence accounts.) The role of intelligence-related programs is sometimes misinterpreted in public discussions of the multi-billion dollar intelligence effort. With the passage of the Intelligence Reform Act in 2004, the Director of National Intelligence (DNI) has extensive statutory authorities for developing and determining the NIP and for presenting it to the President for approval.4 The President in turn forwards the NIP to Congress as part of the annual budget submission in January or February of each year. The Office of the DNI (ODNI) serves as the DNI’s staff for annual budget preparation and submission. The DNI participates in the development of the MIP by the Secretary of Defense. The Under Secretary of Defense for Intelligence (USD(I)) has the responsibility to “oversee all Defense intelligence budgetary matters to ensure compliance with the budget policies issues by the DNI for the NIP.”5 The USD(I) also serves as Program Executive for the MIP and supervises coordination during the programming, budgeting, and execution cycles. Thus, in the development of both the NIP and the MIP essential roles are played by the Office of the DNI and the office of the USD(I). The two offices have overlapping responsibilities and close coordination is required. Past Budgetary Practice Budgeting for secret intelligence efforts has long presented difficult challenges to the Congress. Realizing the need for some direction over the intelligence effort that had been disbanded in the immediate aftermath of World War II, President Truman established, in a directive of January 22, 1946, a coordinative element for intelligence activities, the Central Intelligence Group (CIG), headed by a Director of 3 of the Director of Central Intelligence before the Permanent Select Committee on Intelligence, U.S. House of Representatives, “Public Disclosure of the Intelligence Budget,” February 22, 1994, p. 10. 4 Pursuant to 50 USC 403-1. Many of these functions were previously the responsibility of the DCI. 5 Department of Defense Directive 5143.01. CRS-5 Central Intelligence, and consisting of representatives from the State, War, and Navy Departments. This was not the creation of a new agency, but a coordinative group; personnel and facilities were to be provided “within the limits of available appropriations.”6 This arrangement was questioned, however, because of concern that specific authorization by Congress would be legally required to make funds available to any agency in existence more than a year. Thus, it might have been illegal for the CIG to expend funds after January 22, 1947.7 Shortly after taking office in June, 1946, the second DCI, General Hoyt S. Vandenberg, arranged for the creation of a “working fund” consisting of allotments from the Departments of State, War, and the Navy, under the supervision of the Comptroller General, to cover the costs of the relatively small CIG.8 It cannot be readily determined if funds were transferred from all three departments; the larger budgets of the War and Navy Departments may have made them more likely contributors than the State Department.9 Vandenberg, realizing the administrative weakness of this situation, began an effort to obtain congressional approval of an independent intelligence agency with its own budget. The National Security Act of 1947, which created the unified National Defense Establishment, included provisions for a Central Intelligence Agency, headed by a Director of Central Intelligence. It also authorized the transfer of “personnel, property, and records” of the CIG to the new CIA; it did not, however, provide additional statutory language regarding the administration of the CIA. With the creation of the CIA by the National Security Act of 1947, arrangements were made for the continuation of previous funding mechanisms; “[t]he Agency was to conform as nearly as possible to normal procedures until further legislation by Congress should make exceptions fitting the special needs of the Agency.”10 6 Directive on Coordination of Foreign Intelligence Activities, January 22, 1946, Public Papers of the Presidents, Harry S. Truman, 1946, (Washington: Government Printing Office, 1962), p. 88. 7 See the discussion in an official CIA history prepared in 1952-1953 and published commercially in 1990, Arthur B. Darling, The Central Intelligence Agency: An Instrument of Government to 1950 (University Park, PA: Pennsylvania State University Press, 1990), pp. 104-105. 8 Ibid., pp. 114-115, 166-192. 9 There was, in addition, opposition in the House Appropriations Committee to the State Department’s intelligence effort. See Hanson W. Baldwin, “Intelligence Arm Vital,” New York Times, April 24, 1946, p. 4, cited in Wesley K. Wark, “‘Great Investigations:’ The Public Debate on Intelligence in the US after 1945,” Defense Analysis, June 1987, p. 123. 10 Darling, Central Intelligence Agency, p. 189. Darling comments on the use of unvouchered funds, i.e., funds provided to the DCI for unspecified purposes, “They must be kept secret; even the provision for them by Congress should not be known.” Ibid. Darling continues: the Comptroller General “was willing that unvouchered funds which the National Security Council approved should be exempt from the normal restrictions upon expenditure. But the Bureau of the Budget held that such approval in advance was more properly the function of the Director of the Budget. To this the Comptroller agreed and the proposal went to Congress. The Senate’s committee, however, thought otherwise and exempted the Agency from any control by the Bureau of the Budget over the amount of the expenditures CRS-6 It was recognized that follow-on enabling legislation would be required. After some delays, Congress passed the Central Intelligence Act of 1949 (P.L. 81-110) to provide a firmer statutory base for the CIA and to establish procedures for regular appropriations. This legislation, reported by the two armed services committees, provided authority for the CIA “to transfer to and receive from other Government agencies such sums as may be approved by the Bureau of the Budget [predecessor of today’s Office of Management and Budget]. . . .” The 1949 Act also provided that “sums transferred to the [CIA]... may be expended for the purposes and under the authority of this Act without regard to limitations of appropriations from which transferred . . . .” Representative Carl Vinson, speaking on the floor of the House shortly after passage of the 1949 Act, stated that the legislation contained: the authority to transfer and receive from other Government agencies such sums as may be approved by the Bureau of the Budget for the performance of any of the agency functions. This is how the Central Intelligence Agency gets its money. It has been going on since the agency was created, and this simply legalizes that important function which is the only means by which the amount of money required to operate an efficient intelligence service can be concealed.11 In practice, the CIA Act of 1949 provides funding for CIA through the defense authorization and appropriation process.12 Funding for other intelligence activities undertaken by DOD agencies was logically included in defense bills. For many years, authorizations and appropriations for CIA were handled by a relatively small number of Members and staff of the two appropriations committees with consultation with members of the two armed services committees. According to available sources, senior Members of the Appropriations Committees insisted on maintaining the secrecy of the contents of the CIA’s budget requests and congressional actions in response.13 In 1956, subcommittees were created in the Armed Services and Appropriations Committees of each House to oversee the CIA. Many assessments of the practice of congressional oversight of intelligence activities during the Truman, Eisenhower, Kennedy, and Johnson Administrations have concluded that the congressional role was in large measure supportive and perfunctory. This view has, however, come under serious challenge and there is considerable evidence that Congress took close interest in intelligence spending, which should be unvouchered.” Page 190. This passage reflects the concern that existed in Congress in 1947 for the secrecy of intelligence expenditures. 11 Congressional Record, March 7, 1949, p. 1949. 12 At one point some funds for CIA were included in the State Department budget, but reductions in the overall State Department budget (resulting from the unpopularity of State among some Members at that time) also resulted in cuts in CIA spending; accordingly one Member suggested in 1951 that CIA spending be included in DOD accounts. See David M. Barrett, The CIA and Congress: the Untold Story from Truman to Kennedy (Lawrence, KS: University Press of Kansas, 2005), p. 120. 13 Barrett, CIA and Congress, pp. 118-119, quoting Lyman Kirkpatrick, The Real CIA (New York: MacMillan, 1968), pp. 116-117. CRS-7 especially in regard to major surveillance systems and the construction of headquarters buildings.14 The small handful of Members responsible for intelligence oversight had a close working relationship with the CIA. For a number of years, beginning in the Eisenhower Administration, Senator Richard Russell served both as chairman of the Armed Services Committee and of the Subcommittee on Defense Appropriations and had an especially important influence on intelligence spending.15 During these Cold War years, intelligence budgets grew considerably in significant part because of efforts to determine the extent of Soviet nuclear capabilities through overhead surveillance by manned aircraft such as the U-2s and reconnaissance satellites, and through a worldwide signals intelligence effort. NSA and DIA emerged as major intelligence agencies with large budgets; other agencies were created to launch satellites and interpret overhead photography. These capabilities, which contributed directly to the design of strategic weapons systems and to the negotiation of strategic arms control agreements with the Soviet Union, cost many billions of dollars. These programs were initiated, funded by Congress, and administered in secrecy and involved a number of intelligence agencies and components of DOD. President Lyndon Johnson said on March 16, 1967: I wouldn’t want to be quoted on this but we’ve spent 35 or 40 billion dollars on the space program. And if nothing else had come out of it except the knowledge we’ve gained from space photography, it would be worth 10 times what the whole program has cost. Because tonight we know how many missiles the enemy has and, it turned out, our guesses were way off. We were doing things we didn’t need to do. We were building things we didn’t need to build. We were harboring fears we didn’t need to harbor. Because of satellites, I know how many missiles the enemy has.16 During the Ford Administration, E.O. 11905 of February 18, 1975, consolidated the budget for all intelligence agencies and provided for a comprehensive review of the National Foreign Intelligence Program by the DCI and senior DOD and NSC officials.17 Subsequent executive orders (most recently E.O. 12333 of December 4, 1981) and the Intelligence Authorization Act for FY1993 (P.L. 102-496)18 clarified and strengthened the DCI’s role. The Intelligence Reform and Terrorism Prevention Act of 2004 (P.L. 108-458) gave the newly established position of Director of National Intelligence (DNI) authority to coordinate intelligence activities across the government and to manage the NIP. The DNI has specific responsibilities for developing and determining the annual consolidated NIP budget. The DNI also 14 See Barrett, CIA and Congress, especially pp. 118-124; 215-222. 15 See Gilbert C. Fite, Richard B. Russell, Jr., Senator from Georgia (Chapel Hill, NC: University of North Carolina Press, 1991), especially pp. 368-369. 16 Quoted in Dwayne A. Day, John M. Logsdon, and Brian Latell, eds., Eye in the Sky: the Story of the Corona Spy Satellites (Washington: Smithsonian Institution Press, 1998), p. 1. 17 18 By Executive Order 11905 of February 18, 1975. Section 104(b) of this legislation modified the National Security Act of 1947 to provide that “The Director of Central Intelligence shall provide guidance to elements of the intelligence community for the preparation of their annual budgets and shall approve such budgets before their incorporation in the National Foreign Intelligence Program.” CRS-8 participates in the development of the MIP which is the responsibility of the Secretary of Defense. A key factor encouraging consolidated review of the intelligence budget has been increasingly detailed oversight by Congress. Efforts in the 1950s and 1960s to establish intelligence committees or to involve a larger number of Members in intelligence oversight were rebuffed, with oversight remaining in the hands of a small number of senior members. This situation was altered in the aftermath of the Vietnam War. In reaction to a series of revelations about allegedly illegal and improper activities by intelligence agencies in 1975, Congress created two (temporary) select committees to investigate the CIA and other intelligence agencies.19 The Church and Pike Committees investigated a wide range of intelligence issues and conducted well-publicized hearings. Although budgetary issues were not at the heart of the investigations, there emerged a consensus that congressional oversight of intelligence agencies needed to be strengthened and formalized and permanent intelligence committees established. There was also widespread sentiment expressed that more information regarding intelligence agencies and activities should be made public.20 Following the work of the Church and Pike Committees, Congress moved to revamp oversight of intelligence agencies. The Senate Select Committee on Intelligence (SSCI) was established in 1976, the House Permanent Select Committee on Intelligence (HPSCI) in 1977. Each of these committees was granted oversight of the CIA as well as other intelligence agencies and charged to prevent the types of abuses that the Church and Pike Committees had criticized. In conjunction with their oversight duties, HPSCI and SSCI were responsible for authorizing funds for intelligence activities undertaken by the CIA and other agencies throughout the government. There is, however, a crucial difference between the charters of the two committees. Although HPSCI has oversight of NIP and shares (with the Armed Services Committee) oversight of the MIP, the SSCI has oversight only over the NIP. In the Senate, oversight of the MIP is conducted by the Armed Services Committee 19 The Senate Select Committee to Study Governmental Operation with Respect to Intelligence Activities, known as the Church Committee, and the House Select Committee on Intelligence, known as the Pike Committee. Two years earlier, the Special Senate Committee to Study Questions Related to Secret and Confidential Documents recommended (S.Res. 466, 93rd Congress) that appropriations committees include line items in defense appropriations bills for each of the major intelligence agencies and for the intelligence programs of the armed services. This recommendation was not adopted. See U.S. Congress, Senate, 102d Congress, 1st session, Select Committee on Intelligence, Authorizing Appropriations for Fiscal Year 1992 for the Intelligence Activities of the U.S. Government, the Intelligence Community Staff, the Central Intelligence Agency Retirement and Disability System, and for Other Purposes, S.Rept. 102-117, July 24, 1991, pp. 9-10. 20 The Church Committee concluded: “Although there is a question concerning the extent to which the Constitution requires publication of intelligence expenditures information, the Committee finds that the Constitution at least requires public disclosure and public authorization of an annual aggregate figure for United States national intelligence activities.” U.S. Congress, 94th Congress, 2d session, Senate, Select Committee to Study Governmental Operations with Respect to Intelligence Activities, Final Report, Book I, Foreign and Military Intelligence, S.Rept. 94-755, April 26, 1976, p. 425. CRS-9 (with informal consultation with the intelligence committee).21 Both SSCI and the Senate Armed Services Committee are represented in conferences on intelligence authorization bills; the final bill, as reported by the conference committee, authorizes both intelligence activities and intelligence-related activities. The two intelligence committees are not the sole organs of congressional oversight. The armed services committees often issue sequential reports on intelligence authorization bills. Annual defense authorization acts include the large national intelligence agencies in DOD as well as the intelligence efforts of the four services. Intelligence activities of agencies outside of CIA and DOD are authorized in other legislation although some departments have standing authorizations rather than annual authorization acts. Authorization As is the case with other congressional committees, intelligence oversight has entailed reviewing annual budget proposals for the Intelligence Community submitted by the administration, conducting hearings, preparing an annual authorization bill, and managing it for the respective chamber. The two committees publish reports to accompany the annual intelligence authorization bills, with dollar amounts for various intelligence agencies and activities included in classified annexes.22 The classified annexes are available to all Members, but only within Intelligence Committee offices and sanctions exist for any unauthorized release of classified data. The intelligence committees, however, do not have exclusive jurisdiction over expenditures for intelligence programs. National defense authorization acts also contain authorizing legislation for intelligence activities funded within their purview. There are various parts of defense authorization bills that are classified; some cover what are known as special access or “black” programs.23 These include not only some intelligence programs but also procurement of new weapons systems such as stealth aircraft. Members can obtain information about classified parts of defense authorization bills from the Armed Services Committees. Other authorization bills cover some intelligence activities providing a form of shared oversight. Budgets for INR, DEA and the FBI are funded through the appropriation bills that cover the Departments of Commerce, Justice, and State and similar procedures are used for Treasury and Energy Department intelligence entities in the Treasury, Postal Service, and General Government and Energy and Water 21 See U.S. Congress, Senate, 104th Congress, 2d session, Select Committee on Intelligence, Authorizing Appropriations for Fiscal Year 1997 for the Intelligence Activities of the United States Government and the Central Intelligence Agency Retirement and Disability System, S.Rept. 104-258, pp. 2-3. 22 As noted below, a relatively small portion of the authorization, having to do the CIA Retirement and Disability Fund and the Intelligence Community Management Staff Account, is included in the unclassified reports. 23 Legislative provisions regarding the reporting of budgetary data for special access programs were enacted in the FY1988 Defense Authorization Act (P.L. 100-180). CRS-10 Development appropriations bills. All of these combined, however, represent a small percentage of total intelligence spending–for instance, the FY2007 budget request for INR totaled only $51 million and other agencies are considerably smaller. There has been some controversy regarding the nature of authorizing legislation required. Section 504(a) of the National Security Act provides that appropriated funds may be obligated or expended for an intelligence or intelligence-related activity only if ... those funds were specifically authorized by the Congress for use for such activities... .” The nature of specific authorization had not, however, been defined. On November 30, 1990, President George H.W. Bush refused to sign (“pocket vetoed”) the FY1991 Intelligence Authorization bill when it was presented to him (after the 101st Congress had adjourned) and for over eight months intelligence activities were continued without an intelligence authorization act.24 Although some believed that authorizations contained within the National Defense Authorization Act for FY1991 (P.L. 101-510) were sufficiently specific to meet the requirements of the statute, the House Intelligence Committee subsequently stated that, “It is the view of the congressional intelligence committees that only an intelligence authorization bill provides the degree of specificity necessary to comply with the meaning and intent of Section 504(a).”25 In 1993, language was included in the House report accompanying the FY1994 Defense Authorization Act that the Armed Services Committee “does not intend that the inclusion of ... authorization [of NFIP programs] be considered a specific authorization, as required by section [504] of the National Security Act of 1947... .”26 (This statement indicated that, whereas NFIP programs were not specifically authorized in defense authorization bills, TIARA programs were.) In addition, section 309 of the FY1994 Intelligence Authorization Act for FY1994 (P.L. 103-178) amended the National Security Act of 1947 to make it explicit in law that the general authorization included in the 1947 legislation27 does not satisfy the requirement for specific authorization of intelligence and intelligencerelated activities. 24 In a letter of December 4, 1990, the chairmen of the two Intelligence Committees wrote to the President advising him of their view that only authorizations in the annual intelligence authorization bills satisfied the requirement of Section 504(a) of the National Security Act of 1947 (as amended) for a specific authorization for the funding of intelligence or intelligence-related activities. “While recognizing a need for important intelligence activities and programs to proceed in the interim, the chairmen’s letter underscored the committees’ expectation that intelligence agencies would comply with all of the limitations and conditions on the expenditure of funds which were contained in the vetoed bill.” U.S. Congress, 102d Congress, 1st session, House of Representatives, Permanent Select Committee on Intelligence, Intelligence Authorization Act, Fiscal Year 1991, H.Rept. 10237, April 22, 1991, p. 3. 25 U.S. Congress, House of Representatives, 102d Congress, 2d session, Permanent Select Committee on Intelligence, Pursuant to Clause 1(d) of Rule XI of the Rules of the House of Representatives, H.Rept. 102-1082, December 30, 1992, p. 11. 26 H.Rept. 103-200, p. 464. The Report refers to Section 502 of the National Security Act, but the context makes it clear that a reference to Section 504 was intended. 27 Sec. 307. There are hereby authorized to be appropriated such sums as may be necessary and appropriate to carry out the provisions and purposes of this Act. CRS-11 In some years when appropriations have been passed prior to final action on authorization bills, the appropriations acts have included a provision similar to section 8092 of the FY2006 Defense Appropriations Act (P.L. 109-148): Funds appropriated by this Act, or made available by the transfer of funds in this Act, for intelligence activities are deemed to be specifically authorized by the Congress for purposes of section 504 of the National Security Act of 1947 (50 U.S.C. 414) during fiscal year 2006 until the enactment of the Intelligence Authorization Act for fiscal year 2006. No FY2006 intelligence authorization bill has been passed and, as a result, this brief clause in the appropriations bill has served as the requisite authorization during FY2006. It is considered likely that a defense appropriations bill will be passed prior to floor consideration of the FY2007 intelligence authorization bill (S. 3237) and a similar clause has been included in the defense appropriations bill (H.R. 5631, section 8086). Although these provisions meet the statutory requirement for a “specific authorization,” significantly less congressional guidance is provided for intelligence programs. Appropriations As is the case with all government activities, the appropriations committees have a central role in intelligence programs. Even during the Cold War period when congressional oversight of intelligence activities received little public attention, annual appropriations were required and extensive hearing were held. In recent years, appropriations committees have had an increasingly significant influence on the conduct of intelligence activities. In 1998 a supplemental appropriation act (P.L. 105-277) added substantial funds for intelligence efforts not included in the annual authorization bill, and in the post-9/11 period the practice of relying on supplemental appropriations for funding the regular operations of intelligence agencies has limited the extent of congressional guidance in regard to the intelligence budget. The reliance on supplemental appropriations has been widely criticized; the House Intelligence Committee in 2003 noted that while supplemental appropriations had reflected crisis in the aftermath of terrorist attacks, “The repeated reliance on supplemental appropriations has an erosive negative effect on planning, and impedes long-term, strategic planning. The Committee hopes that the IC has finally reached a plateau of resources and capabilities on which long-term strategic planning can now begin.”28 28 U.S. Congress, 108th Congress, 1st session, House of Representatives, Permanent Select Committee on Intelligence, Intelligence Authorization Act for Fiscal Year 2004, H.Rept. 108-163, June 18, 2003, p. 22. The following year nine members of the House Intelligence Committee in a minority report to the FY2005 intelligence authorization bill argued even more forcefully against funding through supplemental appropriations acts:”Members on both sides of the aisle have roundly criticized this growing practice of funding the Intelligence Community in bits and pieces, rather than for a full fiscal year, the Congress is supposed to do it. Senior intelligence officials have told the Committee that this practice makes it impossible to plan, forcing them to ‘rob Peter to pay Paul’ until the additional funds arrive–potentially jeopardizing key counterterrorism operations.” Minority views of CRS-12 In addition to use of supplemental appropriations to fund intelligence activities, as noted above the required “specific authorization” of intelligence programs required by the section 504 of the National Security Act has in FY2006 been supplied by one paragraph (section 8092) of the FY2006 defense appropriations act (P.L. 109148). The reliance on appropriations measures to authorize intelligence programs may change the contours of intelligence oversight in Congress by emphasizing the role of the two appropriations committees. The defense subcommittees of the two appropriations committees review intelligence budget requests and approve funding levels for intelligence agencies that are part of DOD or whose budgets are contained (but not publicly identified) in defense appropriations acts, i.e., CIA as well as NSA, DIA, the NRO, and NGA. There is a difference between appropriations for the CIA and the ODNI which, although included in defense appropriations acts, are transferred by the Office of Management and Budget (OMB) directly to the DNI and the CIA Director without the involvement of DOD. The Secretary of Defense is, however, heavily involved in the budgets and activities of intelligence agencies in DOD. The CIA and the defense agencies account for the vast bulk of all intelligence spending. Much smaller amounts are funded in appropriations measures for other departments that contain elements of the Intelligence Community. The role of the appropriations committees can be significant. For instance, in 1992, the Defense Appropriation Act for FY1993 (P.L. 102-396) reportedly reduced intelligence spending to a level significantly lower than authorized by the Intelligence Authorization Act (P.L. 102-496).29 In 1990-1991, the Senate Appropriations Committee and the SSCI worked closely together to sponsor a facilities consolidation plan for some CIA activities without the active involvement of the HPSCI. Substantial changes have been made to intelligence programs by appropriations measures and in FY2006 no intelligence authorization act exists and thus agencies rely solely on appropriations legislation. The Question of Disclosure Since the creation of the modern Intelligence Community in the aftermath of World War II, intelligence budgets have not been made public.30 At the conclusion of hostilities in August 1945, intelligence activities were transferred from the Office of Strategic Services (OSS) to the Army, Navy, and State Departments, which Representatives Harman, Hastings, Reyes, Boswell, Peterson, Cramer, Eshoo, Holt, and Ruppersberger, U.S. Congress, 108th Congress, 2d session, House of Representatives, Permanent Select Committee on Intelligence, Intelligence Authorization Act for Fiscal Year 2005, H.Rept. 108-558, June 21, 2004, p. 69. 29 See statements by Representative Combest, Congressional Record, August 3, 1993, p. H5678; Representative Pelosi, Congressional Record, August 3, 1993, p. H5697. See also George Lardner Jr. and Walter Pincus, “Congress May Seek Review of All Intelligence Spending,” Washington Post, January 10, 1993, p. A4. 30 During World War II, spending for the Office of Strategic Services (OSS), the predecessor of the CIA, was openly included in National War Agencies Appropriation Acts. CRS-13 assumed responsibility for their funding. Meeting the expenses of the CIG, created in 1946, required the establishment of a “working fund,” as noted above, which received allocations from the three departments. This pattern was continued when the CIA was established the following year (although there may have been relatively few, if any, transfers from the State Department). The transfer of appropriated funds was done secretly, reportedly at the insistence of Members of Congress.31 There are several parts of the intelligence budget that are made public. The costs of the Intelligence Community Management Account (CMA) are specified in annual intelligence authorization acts as are the costs of the CIA Retirement and Disability System (CIARDS). The CMA includes staff support to the DNI role and the CIARDS covers retirement costs of CIA personnel not eligible for participation in the government-wide retirement system. For FY2005, $310.4 million was authorized for 310 full-time CMA personnel and $239.4 million was authorized for CIARDS. In addition, the budget for the State Department’s Bureau of Intelligence and Research is made public and some, but not all, tactical intelligence programs are identified in unclassified DOD budget submissions. Careful scrutiny of officiallypublished data on intelligence expenditures would not, however, provide a valid sense of the size and content of the intelligence budget. The Church and Pike committees both called for public disclosure of the total amounts of each annual intelligence budget.32 The then DCI, George H.W. Bush, and President Ford both appealed to the Senate not to proceed with disclosure and the question was referred to the newly created SSCI. After conducting hearings,33 SSCI recommended (by a one vote margin) in May 1977 (S.Res. 207, 95th Congress) that aggregate amounts appropriated for national foreign intelligence activities for 31 Walter Pforzheimer, the first Legislative Counsel to the CIG and the CIA, testified in 1992 to the SSCI, that “... from the very beginning ... [the intelligence budget] has always been secret, and it was not at our initial request, although we supported it. It was the Congress who kept it secret....” U.S. Congress, Senate, 102d Congress, 2d session, Select Committee on Intelligence, S. 2198 and S. 421 to Reorganize the United States Intelligence Community, S. Hearings 894, February 20, March 4,12,19, 1992, p. 151. 32 The Church Committee recommended that the planned congressional “intelligence oversight committee[s] should authorize on an annual basis a ‘National Intelligence Budget,’ the total amount of which would be made public.” The Church Committee further recommended that the intelligence committees “consider whether it is necessary, given the Constitutional requirements and the national security demands, to publish more detailed budgets.” U.S. Congress, Senate, 94th Congress, 2d session, Select Committee to Study Governmental Operations with Respect to Intelligence Activities, Foreign and Military Intelligence, Book I, April 26, 1976, p. 470. Similarly, the Pike Committee recommended that “all intelligence related items be included as intelligence expenditures in the President’s budget, and that there be disclosure of the total single sum budgeted for each agency involved in intelligence, or if such an item is a part or portion of the budget of another agency or department that it be separately identified as a single item.” U.S. Congress, House of Representatives, 94th Congress, 2d session, Select Committee on Intelligence, Recommendations of the Final Report, H.Rept. 94-833, February 11, 1976, p. 3. 33 U.S. Congress, Senate, 95th Congress, 1st session, Select Committee on Intelligence, Whether Disclosure of Funds Authorized for Intelligence Activities is in the Public Interest, Hearings, April 27-28, 1977. CRS-14 FY1978 be disclosed.34 recommendation. The full Senate did not, however, act on this HPSCI, established by House Rule XLVIII after the termination of the Pike Committee, made an extensive study of the disclosure question. After conducting hearings in 197835 (and despite the willingness of then DCI Stansfield Turner to accept disclosure of “a single inclusive budget figure”) the House Committee concluded unanimously that it could find “no persuasive reason why disclosure of any or all amounts of the funds authorized for the intelligence and intelligence-related activities of the government would be in the public interest.”36 With the failure of either chamber to take action, the disclosure question receded into the background as efforts (ultimately unsuccessful) were underway during the Carter Administration to draft a legislative charter for the entire Intelligence Community. The Reagan Administration showed markedly less interest in such questions as it launched a major expansion of intelligence activities. The issue would return during the Clinton Administration after the end of the Cold War and again in the recommendations of the 9/11 Commission as noted below. It should be understood that with the establishment of the two intelligence committees in the 1970s, Members have been able to review budget figures contained in the classified annexes accompanying reports intelligence authorization bills, although rules of both chambers prevent the divulging of classified information. Policy Arguments, Pro and Con Since the 1970s, arguments for and against the public disclosure of intelligence spending levels have turned on essentially the same issues, viz. the constitutional issue regarding the requirement for full reports of government expenditures (discussed below) and the broader question of the value of open political discourse, the dangers of revealing useful information to actual or potential enemies, and the difficulty of providing and debating aggregate numbers without being drawn into providing details.37 Advocates of disclosure argue that greater public discussion of intelligence spending made possible by the disclosure of spending levels would ultimately lead 34 U.S. Congress, Senate, 95th Congress, 1st session, Select Committee on Intelligence, Whether Disclosure of Funds for the Intelligence Activities of the United States is in the Public Interest, S.Rept. 95-274, June 16, 1977, p. 9. The appropriated amount, rather than the authorized amount, was to be disclosed because it represented final congressional action embodied in legislation. Ibid., p. 4, 10. 35 U.S. Congress, House of Representatives, 95th Congress, 2d session, Permanent Select Committee on Intelligence, Disclosure of Funds for Intelligence Activities, Hearings, January 24,25, 1978. 36 U.S. Congress, House of Representatives, 95th Congress, 2d session, Permanent Select Committee on Intelligence, Annual Report Pursuant to Section 3 of House Resolution 658, 95th Congress, 1st session , Report No. 95-1795, October 14, 1978, pp. 15-16. 37 Much of the discussion in this section is taken directly from Alfred B. Prados, Intelligence Budgets: Contents and Releasability, CRS Report 89-465, pp. 35-43. CRS-15 to a stronger intelligence effort. They maintain that no organization, even one with superior management and personnel, is immune to waste and inefficiency and that wider appreciation of the costs and benefits of intelligence could contribute in the long run toward improvements in the organization and functioning of intelligence.38 Senator William Proxmire put the case as follows: ... people not only have a right to know, but you are going to have a much more efficient government when they do know. We only make improvements when we get criticized, and you can only criticize when you know what you are talking about, when you have some information. If you know that there is a certain amount being spent on intelligence, then you are in a much stronger position to criticize what you are getting for that expenditure.39 Also, in terms of efficiency, publication of an aggregate figure for intelligence spending would result in a cleaner, more accurate defense budget. As presently handled, the defense budget includes significant unspecified national intelligence expenditures (e.g., the greater part of the CIA budget) that in many cases are not actually part of defense spending per se. Such expenditures make the defense budget and various components of it seem larger than is the case. Identification of those intelligence expenditures that are extraneous to defense could give the public a more accurate perception of defense costs.40 Those holding this position argue, in addition, that publication of limited intelligence spending totals would provide no useful information to a present or future adversary. Even during the height of the Cold War, Soviet authorities, they maintain, undoubtedly had a reasonably accurate knowledge of the extent of the U.S. intelligence budget and, in any event, were more concerned with the nature of our 38 In making a recommendation in favor of disclosure of aggregate figures appropriated for national intelligence and for its various component agencies, in mid-2004 the 9/11 Commission stated: The specifics of the intelligence appropriation would remain classified, as they are today. Opponents of declassification argue that America’s enemies could learn about intelligence capabilities by tracking the top-line appropriations figure. Yet the top-line figure by itself provides little insight into U.S. intelligence sources and methods. The U.S. government readily provides copious information about spending on its military forces, including military intelligence. The intelligence community should not be subjected to that much disclosure. But when even aggregate categorical numbers remain hidden, it is hard to judge priorities and foster accountability. 9/11 Commission Report, p. 416. 39 U.S. Congress, Senate, 95th Congress, 1st session, Select Committee on Intelligence, Whether Disclosure of Funds Authorized for Intelligence Activities Is in the Public Interest, Hearings, April 27-28, 1977, p. 41. 40 See Louis Fisher, “Confidential Spending and Governmental Accountability,” George Washington Law Review, vol. 47, January 1979. CRS-16 activities rather than the size of expenditures. Noting the demise of the Soviet Union, Representative Dan Glickman, then the Chairman of the House Intelligence Committee, stated in 1994 that “Unless a justification on national security grounds exists, keeping the budget totals secret serves only one purpose, and that is to prevent the American taxpayer from knowing how much money is spent on intelligence.”41 Opposition to public release has been based on the conviction that intelligence by its very nature stands apart from other activities of the government and the publication of general budgetary information, potentially exploitable by an adversary attempting to discern U.S. intelligence capabilities and operations, could compromise the nation’s intelligence capabilities. This concept perceives intelligence to be an exceptional activity that cannot be handled according to normal procedures of an open society. This is particularly true of those operations that involve the collection of intelligence information. Sophisticated reconnaissance devices, electronic technology, and human resources operating at significant risk are particularly vulnerable to human error or hostile penetration; consequently, they require extraordinary protective measures. In 1983, HPSCI described the unique vulnerabilities of intelligence systems as follows: Intelligence activities and capabilities are inherently fragile. Unlike weapons systems, which can be countered only by the development of even more sophisticated systems developed over a long period, intelligence systems are subject to immediate compromise. Often they can be countered or frustrated rapidly simply on the basis of knowledge of their existence. Thus budget disclosure might well mean more to this country’s adversaries than to any of its citizens. Further, this information could then be used to frustrate United States intelligence missions.42 At the end of the Cold War along with the downsizing of the defense budget it was argued that intelligence spending should be significantly reduced. Some advocates of reduction anticipated that publication of spending totals would lead to a perception by the public that such levels of intelligence spending were unjustified and could be lowered. This potential for public opposition to existing levels of spending was also recognized by many who defended intelligence spending levels and probably reinforced their opposition to making the budget public. Although such perspectives may have been widely shared in the early 1990s, later in the decade the emergence of international terrorism and other transnational threats lead to concerns that intelligence spending should not be further reduced. The 9/11 attacks altered the climate regarding intelligence spending; even though there was widespread criticism of the performance of intelligence agencies, there was a pervasive determination to spend whatever was necessary on intelligence as part of the global war on terrorism. In recent years the argument for making intelligence spending levels public has not in general been a proxy argument for reducing 41 42 Opening Statement, Chairman Dan Glickman, February 22, 1994, p. 3. U.S. Congress, House of Representatives, 98th Congress, 2d session, Permanent Select Committee on Intelligence, Intelligence Authorization Act for Fiscal Year 1984, Report to Accompany H.R. 2968, H.Rept. 98-189, Part I, May 16, 1983, p. 2. CRS-17 intelligence spending inasmuch as few would argue that less intelligence is needed given the realistic potential for more Al Qaeda attacks. Other opponents of disclosure have argued that making public a few numbers indicating total spending levels (whether budget requests, authorizations, or appropriations) will be meaningless to the public debate. Explanations will be immediately required to show that these figures are divided among several functions, threats, and agencies, cover national and tactical programs, may or may not include administrative and logistical support, etc. Pressures will in a politically adversarial context mount to publish these sub-totals as well as an aggregated figure. It is further argued that these explanations would likely result in a degree of transparency for U.S. intelligence activities that would allow adversaries to take effective countermeasures. There is also a contrary argument that intelligence spending, even within the NIP, is in large measure related to defense programs and could be usefully expressed as a percentage of overall defense spending. Admiral Bobby Ray Inman, who served as Deputy Director of Central Intelligence in the early Reagan Administration, testified in 1991 that, “I am certainly prepared to make unclassified the total amount, and defend to the public why 10% of our total defense efforts spent for both national and tactical intelligence is not a bad goal at all. Just as I don’t think that 11 or 12% of the budget for research and development is a bad goal at all for the country.”43 Some opponents of greater disclosure point out that large fluctuations in intelligence spending might also reveal major new programs under development (the example of the U-2s and satellites is sometimes mentioned). Premature exposure of such new capabilities could severely limit their ability to acquire valuable information before adversaries become aware of U.S. capabilities. On the other hand, according to a 1991 Senate report, DCI Stansfield Turner “testified in 1977 that there had been no ‘conspicuous bumps’ in the intelligence budget for the preceding decade. The [Senate] Select Committee’s experience is similarly that no secrets would have been lost by publishing the annual aggregate budget total since then.”44 Unconvinced defense analysts insist that revealing the fact of significant changes in U.S. intelligence budgets from year to year will alert unfriendly governments or groups to new efforts against them (or to a slackened effort by the U.S. that can be exploited). Public discussion of the question of making intelligence budgets public has usually turned on the question of the constitutionality or the propriety of keeping intelligence spending figures classified. Beyond these issues, however, lies the lessdiscussed issue of the nature of intelligence and intelligence-related spending. The 43 U.S. Congress, Senate, 102d Congress, 1st session, Select Committee on Intelligence, Review of Intelligence Organization, S. Hearing 102-91, March 21, 1991, p. 24. 44 U.S. Congress, Senate, 102d Congress, 1st session, Select Committee on Intelligence, Authorizing Appropriations for Fiscal Year 1992 for the Intelligence Activities of the U.S. Government, the Intelligence Community Staff, the Central Intelligence Agency Retirement and Disability System, and for Other Purposes, S.Rept. 102-117, July 24, 1991, p. 12. It is not clear if this conclusion included covert actions (officially secret but widely debated) that have been funded in the intelligence budget process. CRS-18 existence of the NIP and the MIP has been publicly acknowledged in many Executive and Legislative Branch publications. However, the respective roles of the separate programs are not well known outside of a relatively narrow circle of intelligence specialists. The role of tactical programs in particular is rarely considered in the context of discussions of making intelligence spending levels public. Observers express concern that characterizing some projects related to information support for targeting as a tactical intelligence program could be characterized in some cases as arbitrary inasmuch as similar projects may be included in other parts of the Defense budget. Reportedly, inclusion of some projects in the MIP program is not consistent from year to year and thus could lead to confusion in tracking intelligence spending. Some consideration has been given to making public only the budget for the NIP which contains funding for the CIA, the National Reconnaissance Office (NRO), the National Geospatial-Intelligence Agency (formerly the National Imagery and Mapping Agency (NIMA)), and the National Security Agency (NSA). When making total NIP spending public, some observers would consolidate responsibility for authorizing NIP in the two intelligence committees, leaving the armed services to deal with the MIP. It is likely that jurisdiction of the Armed Services committees will continue inasmuch as the NIP includes the budgets of major defense agencies that report to the Secretary of Defense and to which are assigned many thousands of military personnel. Some argue that the close ties between the NRO, NGA, and NSA and other Defense agencies also require that their budgets be prepared in the same Department. In 2002 the position of Under Secretary of Defense for Intelligence (USD(I)) was established by section 901 of the FY2003 National Defense Authorization Act (P.L. 107-314). The incumbent of this position, currently Stephen Cambone, is charged with overseeing the budgets of DOD’s intelligence agencies, including the portions that fall within the NIP and those are contained in the MIP. The USD(I) is the key point of contact between DOD and the Office of the DNI and the two offices collaborate in the preparation of annual budget submissions to Congress along with those of other intelligence agencies. Constitutional Questions Related to Disclosure of Aggregate Intelligence Budget Figure An issue that arises in considering whether or not to disclose an aggregate intelligence budget figure is whether the Statement and Account Clause of the United States Constitution requires such disclosure. The pertinent constitutional language is contained in Article I, Section 9, Clause 7, which states: No money shall be drawn from the Treasury, but in Consequence of Appropriations made by Law; and a regular Statement and Account of the Receipts and Expenditures of all public Money shall be published from time to time. [Emphasis added.] CRS-19 A brief examination of the history of this language and of the scant case law interpreting the Statement and Account Clause may be of assistance in placing the disclosure issue in context. History of the Constitutional Language During the Constitutional Convention in Philadelphia, the first language on the subject of statements and accounts was offered on September 14, 1787, by George Mason. The debate on the matter, as reflected in Madison’s “Notes of Debates,” was as follows: Col. Mason moved a clause requiring “that an Account of the public expenditures should be annually published” Mr. Gerry 2ded. the motion Mr Govr. Morris urged that this wd. be impossible in many cases. Mr. King remarked, that the term expenditures went to every minute shilling. This would be impracticable. Congs. might indeed make a monthly publication, but it would be in such general Statements as would afford no satisfactory information. Mr. Madison proposed to strike out “annually” from the motion & insert “from time to time” which would enjoin the duty of frequent publications and leave enough to the discretion of the Legislature. Require too much and the difficulty will be get a habit of doing nothing. The articles of Confederation require halfyearly publications on this subject — A punctual compliance being often impossible, the practice has ceased altogether — Mr Wilson 2ded. & supported the motion — Many operations of finance cannot be properly published at certain times. Mr, Pinckney was in favor of the motion. Mr. Fitzimmons — It is absolutely impossible to publish expenditures in the full extent of the term. Mr. Sherman thought “from time to time” the best rule to be given. “Annual” was struck out — & those words — inserted nem: con: The motion of Col. Mason so amended was then agreed to nem: con: and added after — “appropriations by law as follows — “And a regular statement and account of the receipts & expenditures of all public money shall be published from time to time.”45 45 2 M. Farrand, THE RECORDS OF THE FEDERAL CONVENTION OF 1787 618-19 (1937) (hereinafter Farrand); I W. Benton, 1787 DRAFTING THE U.S. CONSTITUTION 1004-05 (1986). There is some non-substantive variation between these two sources as to use of abbreviations, and occasionally as to punctuation or spelling. The Farrand version is quoted directly above. No corrections of punctuation, spelling or capitalization have been made, so that the quotation would as closely parallel the original as possible. The term nem. con. stands for nemine contradicente, the Latin term meaning “no one contradicting”. The CRS-20 During the Virginia ratifying convention, the Statement and Account Clause occasioned comment on at least two occasions. On June 12, 1788, James Madison observed: The congressional proceedings are to be occasionally published, including all receipts and expenditures of public money, of which no part can be used, but in consequence of appropriations made by law. This is a security which we do not enjoy under the existing system. That part which authorizes the government to withhold from the public knowledge what in their judgment may require secrecy, is imitated from the confederation — that very system which the gentleman advocates.46 On the 17th of June, 1788,47 George Mason raised a question as to the “from time to time” language, and the following debate ensued: Mr. GEORGE MASON apprehended the loose expression of “publication from time to time” was applicable to any time. It was equally applicable to monthly and septennial periods. It might be extended ever so much. The reason urged in favor of this ambiguous expression was, that there might be some matters which require secrecy. In matters relative to military operations and foreign negotiations, secrecy was necessary sometimes; but he did not conceive that the receipts and expenditures of the public money ought ever to be concealed. The people, he affirmed, had a right to know the expenditures of their money; but that this expression was so loose, it might be concealed forever from them, and might afford opportunities of misapplying the public money, and sheltering those who did it. He concluded it to be as exceptionable as any clause, in so few words, could be. Mr. LEE (of Westmoreland) thought such trivial argument as that just used by the honorable gentleman would have no weight with the committee. He conceived the expression to be sufficiently explicit and satisfactory. It must be supposed to mean, in the common acceptation of language, short, convenient periods. It was as well as if it had said one year, or a shorter term. Those who would neglect this provision would disobey the most pointed directions. As the Assembly was to meet next week, he hoped gentlemen would confine themselves to the investigation of the principal parts of the Constitution. Journal of the Convention, published in Boston in 1819, is quite cryptic, shedding no light on the debate on this clause. However, it does indicate, at pp. 377-78, on September 14, 1787: “Add at the end of the sixth clause of the ninth section, first article, `and a regular statement and account of the receipts and expenditures of all public money shall be published from time to time.’” See 2 Farrand at 610 n.2. Farrand also notes that the last paragraph of the quotation included in the text above may have been a later insertion, and if so he opines that it was taken from this notation in the Journal. 2 Farrand at 619 n. 17. 46 3 M. Farrand, THE RECORDS OF THE FEDERAL CONVENTION OF 1787 311 (1911), citing Robertson, DEBATES OF THE CONVENTION OF VIRGINIA, 1788 236 (2d ed. 1805). 47 Elliot lists this date as the 15th of June 1788, but he lists both the immediately preceding Saturday and Monday as being the 14th of June. In fact, the 14th of June was on a Saturday in 1788, so the correct date for the Tuesday of that week would be the 17th of June, 1788, as reflected in Farrand. CRS-21 Mr. MASON begged to be permitted to use that mode of arguing to which he had been accustomed. However desirous he was of pleasing that worthy gentleman, his duty would not give way to that pleasure. Mr. GEORGE NICHOLAS said it was a better direction and security than was in the state government. No appropriation shall be made of the public money but by law. There could not be any misapplication of it. Therefore, he thought, instead of censure it merited applause; being a cautious provision, which few constitutions, or none, had ever adopted. Mr. CORBIN concurred in the sentiments of Mr. Nicholas on this subject. Mr. MADISON thought it much better than if it had mentioned any specified period; because, if the accounts of the public receipts and expenditures were to be published at short, stated periods, they would not be so full and connected as would be necessary for a thorough comprehension of them, and detection of any errors. But by giving them an opportunity of publishing them from time to time, as might be found easy and convenient, they would be more full and satisfactory to the public, and would be sufficiently frequent. He thought, after all, that this provision went farther than the constitution of any state in the Union, or perhaps in the world. Mr. MASON replied, that, in the Confederation, the public proceedings were to be published monthly, which was infinitely better than depending on men’s virtue to publish them or not, as they might please. If there was no such provision in the Constitution of Virginia, gentlemen ought to consider the difference between such a full representation, dispersed and mingled with every part of the community, as the state representation was, and such an inadequate representation as this was. One might be safely trusted, but not the other. Mr. MADISON replied, that the inconveniences which had been experienced from the Confederation, in that respect, had their weight in him in recommending this in preference to it; for that it was impossible, in such short intervals, to adjust the public accounts in any satisfactory manner. Mr. HENRY. Mr Chairman, we have now come to the 9th section, and I consider myself at liberty to take a short view of the whole. I wish to do it very briefly. Give me leave to remark that there is a bill of rights in that government. There are express restrictions, which re in the shape of a bill of rights; but they bear the name of the 9th section. The design of the negative expressions in this section is to prescribe limits beyond which the powers of Congress shall not go. These are the sole bounds intended by the American government. Whereabouts do we stand with respect to a bill of rights? Examine it, and compare it to the idea manifested by the Virginian bill of rights, or that of the other states. The restraints in this congressional bill of rights are so feeble and few, that it would have been infinitely better to have said nothing about it. The fair implication is, that they can do every thing they are not forbidden to do. What will be the result if Congress, in the course of their legislation, should do a thing not restrained by this 9th section? It will fall as an incidental power to Congress, not being prohibited expressly in the Constitution.... If the government of Virginia passes a law in contradiction to our bill of rights, it is nugatory. By that paper the national wealth is to be disposed of under the CRS-22 veil of secrecy; for the publication from time to time will amount to nothing, and they may conceal what they may think requires secrecy. How different it is in your own government! Have not the people seen the journals of our legislature every day during every session? Is not the lobby full of people every day? Yet gentlemen say that the publication from time to time is a security unknown in our state government! Such a regulation would be nugatory and vain, or at least needless, as the people see the journals of our legislature, and hear their debates, every day. If this be not more secure than what is in that paper, I will give up that I have totally misconceived the principles of the government. You are told that your rights are secured in this new government. They are guarded in no other part but this 9th section. The few restrictions in that section are your only safeguards. They may control your actions, and your very words, without being repugnant to that paper. The existence of your dearest privileges will depend upon the consent of Congress, for they are not within the restrictions of the 9th section.... 48 Some attention to this clause was also given in the New York ratifying convention and in the Maryland House of Delegates. The pertinent portion of the New York debates took place on June 27, 1788. During those debates, Mr. Chancellor Livingston, in expounding upon concerns raised with regard to the power to tax, stated in pertinent part: ... You will give up to your state legislatures every thing dear and valuable; but you will give no power to Congress, because it may be abused; you will give them no revenue, because the public treasures may be squandered. But do you not see here a capital check? Congress are to publish, from time to time, an account of their receipts and expenditures. These may be compared together; and if the former, year after year, exceed the latter, the corruption will be detected, and the people may use the constitutional mode of redress.... ... I beg the committee to keep in mind, as an important idea, that the accounts of the general government are, “from time to time,” to be submitted to the public inspection. Hon. Mr. SMITH remarked, that “from time to time’ might mean from century to century, or any period of twenty or thirty years. The CHANCELLOR asked if the public were more anxious about any thing under heaven than the expenditure of money. Will not the representatives, said he, consider it essential to their popularity, to gratify their constituents with full and frequent statements of the public accounts? There can be no doubt of it.49 48 III J. Elliot, THE DEBATES OF THE SEVERAL STATE CONVENTIONS ON THE ADOPTION OF THE FEDERAL CONSTITUTION AS RECOMMENDED BY THE GENERAL CONVENTION AT PHILADELPHIA IN 1787 459-62 (1888) (hereinafter Elliot); see also, 2 Farrand at 326-27 (spelling, capitalization, and punctuation as in the original) (Farrand contains ellipses in place of part of this day’s debates, including comments of Mr. Lee, Mr. Mason, Mr. Corbin, Mr. Nicholas, and Mr. Henry). 49 II Elliot at 345-47. CRS-23 On November 29, 1787, the Delegates to the Constitutional Convention were called before the Maryland House of Delegates to explain the Principles, upon which the proposed Constitution was founded.50 James McHenry, in his explanation of Section 9, stated in part: ... When the Public Money is lodged in its Treasury there can be no regulation more consistent with the Spirit of Economy and free Government that it shall only be drawn forth under appropriation by Law and this part of the proposed Constitution could meet with no opposition as the People who give their Money ought to know in what manner it is expended.51 Thus, the history of this provision sheds some light upon the range of views with regard to anticipated benefits and intended sweep of this language, but does not give great attention to the possibility of secret funding for intelligence activities. Rather, the debate focused principally upon the general need for such a provision, the timing of the statements and accounts, and the practical impact of such a requirement. Nevertheless, there were a few indications that some of the delegates considered the possibility of secrecy attached to some of those statements and accounts. For example, one might compare Mr. Wilson’s observations during the Constitutional Convention with those of Mr. Mason at the Virginia ratifying convention. Mr. Wilson noted that some financial operations could not be published at certain times. Mr. Mason recognized that at times necessity might dictate that some secrecy would attach to military operations or foreign negotiations, but rejected the notion that receipts and expenditures of public money should ever be concealed.52 The most explicit mention of receipts and expenditures shrouded in secrecy is contained in the remarks of Mr. McHenry. He regarded the clause’s requirement of publication from time to time as so broad as to permit the Congress to dispose of the public wealth in secrecy or to conceal what they determine requires secrecy. The history of the clause leaves it uncertain whether or to what extent his views were shared by others. It appears clear that the concern over how public funds would be spent was the motivating force behind the inclusion of the Statement and Account Clause. The clause seems to impose an affirmative duty to disclose information with regard to public receipts and expenditures. These general outlines do not appear to provide unequivocal guidance as to the scope and frequency of these disclosures, however, and there are some indications that at least delay in releasing some information and possibly secrecy of some information was anticipated, whether with approbation or alarm, by some of those at the Constitutional Convention and the ratifying conventions. 50 This document, included in 3 Farrand at 144-150, was among the manuscripts of John Leeds Bozman, and was identified by Mr. Bernard C. Steiner of the Enoch Pratt Free Library, Baltimore, as part of the legislative records of Maryland, in the handwriting of one of the clerks of the legislature. 3 Farrand at 144 n. 2. 51 52 3 Farrand at 149-150. For a discussion of some historical instances of confidential spending, see Fisher, “Confidential Spending and Governmental Accountability,” 47 G.W. L. REV. 347 (1979); see also, Halperin v. CIA, 629 F.2d 144, 157-60 (D.C. Cir. 1980). CRS-24 Judicial Interpretation Further insight may be drawn from an examination of judicial interpretation of the clause in the intelligence budget context. Several cases appear to be of significance in this regard. In 1974, the United States Supreme Court decided United States v. Richardson, 418 U.S. 166 (1974). There a federal taxpayer challenged the constitutionality of provisions of the Central Intelligence Agency Act of 1949 concerning public reporting of expenditures on the ground that they violated the Statement and Account Clause. The provisions at issue permitted the CIA to account for its expenditures solely on the certificate of the Director, 50 U.S.C. § 403j(b). Richardson had made several attempts to obtain detailed information regarding the CIA’s expenditures from the Government Printing Office and the Fiscal Service of the Bureau of Accounts of the Treasury Department, but found the information he received unsatisfactory. He questioned the constitutionality of the provision and requested that the Treasury Department seek an opinion from the Attorney General on this question. The Treasury Department declined to do so, and Richardson then filed suit. The district court dismissed for lack of standing and on the ground that the subject matter raised political questions not amenable to judicial determination. Richardson’s request for a three-judge court to try the matter was also rejected by the District Court. The Court of Appeals for the Third Circuit, sitting en banc, reversed, and remanded for consideration by a three-judge court. The Supreme Court granted certiorari and reversed. The issue before the Court was whether the respondent had standing to sue. The Court found that he did not, without reaching the merits of the constitutional question.53 In so doing, the Court noted: It can be argued that if respondent is not permitted to litigate this issue, no one can do so. In a very real sense, the absence of any particular individual or class to litigate these claims gives support to the argument that the subject matter is committed to the surveillance of the Congress, and ultimately to the political process....54 In footnote 11, 418 U.S. at 178, the Court also observed: Although we need not reach or decide precisely what is meant by “a regular Statement and Account,” it is clear that Congress has plenary power to exact any 53 But see Justice Douglas’ dissenting opinion, 418 U.S. at 197-202. Justice Douglas reviewed the history of the Statement and Account Clause, concluding that it was inserted into the Constitution to give the public knowledge of the way public funds are spent. He concluded that to permit Congress to determine to withhold a regular statement and account with regard to an agency is to reduce the clause to a nullity. Further, he asserted that if the solution to the failure of the Congress to provide such a statement and account is the electoral process, then the public must have “a basic knowledge of at least the generality of the accounts under every head of government” if the franchise is to be exercised intelligently. Id., at 201. Justice Douglas would have affirmed the Court of Appeals holding that the taxpayer had standing to sue. 54 418 U.S. at 179. CRS-25 reporting and accounting it considers appropriate in the public interest. It is therefore open to serious question whether the Framers of the Constitution ever imagined that general directives to the Congress or the Executive would be subject to enforcement by an individual citizen. While the available evidence is neither qualitatively nor quantitatively conclusive, historical analysis of the genesis of cl. 7 suggests that it was intended to permit some degree of secrecy of governmental operations. The ultimate weapon of enforcement available to the Congress would, of course, be the “power of the purse.” Independent of the statute here challenged by respondent, Congress could grant standing to taxpayers or citizens, or both, limited, of course, by the “cases” and “controversies” provision of Art. III. Not controlling, but surely not unimportant, are nearly two centuries of acceptance of a reading of cl. 7 as vesting in Congress plenary power to spell out the details of precisely when and with what specificity Executive agencies must report the expenditure of appropriated funds and to exempt certain secret activities from comprehensive public reporting. See 2 M. Farrand, The Records of the Federal Convention of 1787, pp. 618-619 (1911); 3 id., at 326-327; 3 J. Elliot, Debates on the Federal Constitution 462 (1836); D. Miller, Secret Statutes of the United States 10 (1918). Several lower court decisions are also instructive here. In Harrington v. Bush, 553 F.2d 190 (D.C. Cir. 1977), a Member of Congress sought declaratory and injunctive relief to foreclose the CIA from using the funding and reporting provisions of the 1949 Central Intelligence Act in connection with allegedly illegal activities. The United States Court of Appeals for the District of Columbia Circuit dismissed the suit for lack of standing. Plaintiff did not challenge the constitutional sufficiency of the funding and reporting provisions.55 In outlining the statutory and constitutional framework to set the case in context, the court noted that the funding and reporting requirements of the CIA Act ... represent an exception to the general method for appropriating and reporting the expenditure of federal funds. Article I, section 9, clause 7 of the U.S. Constitution ... is not self-defining and Congress has plenary power to give meaning to the provision. The Congressionally chosen method of implementing the requirements of Article I, section 9, clause 7 is to be found in various statutory provisions.... With respect to the reporting of expenditures, the key statutory provision of general application is 31 U.S.C. § 1029 which imposes a duty on the Secretary of the Treasury to provide Congress on an annual basis with “... an accurate, combined statement of the receipts and expenditures ... of all public moneys....” Since Congressional power is plenary with respect to the definition of the appropriations process and reporting requirements, the legislature is free to establish exceptions to this general framework, as has been done with respect to the CIA....56 In Halperin v. CIA, 629 F.2d 144 (D.C. Cir. 1980), a private citizen sought access to CIA documents regarding legal bills and fee arrangements of private 55 553 F.2d at 194, 196. 56 553 F.2d at 194-95, relying in part on United States v. Richardson, supra. CRS-26 attorneys retained by the Agency through the Freedom of Information Act, 5 U.S.C. § 552. The documents were held to be exempt from disclosure under FOIA, exception 3, which addressed documents specifically exempted by statute. Judge Gasch found both that the documents were exempted under the protection from unauthorized disclosure afforded intelligence sources and methods, 50 U.S.C. § 403(d)(3) (1976), and that the information sought was specifically exempted by Section 6 of the Central Intelligence Act, 50 U.S.C. § 403g (1976).57 The plaintiff argued that application of these statutes under the FOIA exemption was violative of the Statement and Account Clause. The appellate court, relying upon United States v. Richardson, supra, rejected his argument, holding that he lacked standing to challenge the constitutionality of secret appropriations and expenditures for the CIA. The court found that the nature of the injury alleged by the plaintiff under FOIA was undifferentiated and common to all members of the Public and therefore, like the taxpayer in Richardson, the plaintiff had not shown the “`particular concrete injury’ required for standing.”58 In determining the constitutionality and justiciability of statutory secrecy for CIA expenditures, the Halperin court reviewed the history of the Statement and Account Clause. As to the debates in the Virginia ratifying convention in June of 1788, the court opined: Mason’s statement clarifies several points concerning the Framers’ intent. First it appears that Madison’s comment on governmental discretion to maintain the secrecy of some expenditures, far from being an isolated statement, was representative of his fellow proponents of the “from time to time” provision. Second, as to what items might legitimately require secrecy, the debates contain prominent mention of military operations and foreign negotiations, both areas closely related to the matters over which the CIA today exercises responsibility. Finally, we learn that opponents of the “from time to time” provision, exemplified by Mason, favored secrecy only for the operations and negotiations themselves, not for receipts and expenditures of public money connected with them. But the Statement and Account Clause, as adopted and ratified, incorporates the view not of Mason, but rather of his opponents, who desired discretionary secrecy for the expenditures as well as the related operations.... The court regarded Patrick Henry’s concern59 over the “time to time” language and the potential for expenditures being concealed by Congress as confirmation for the court’s interpretation of the Madison-Mason debate. It observed further: Viewed as a whole, the debates in the Constitutional Convention and the Virginia ratifying convention convey a very strong impression that the Framers of the Statement and Account Clause intended it to allow discretion to Congress and the 57 629 F.2d at 146. Section 403g provided an exemption from provisions of any other law requiring disclosure or publication of the organization, functions, names, official titles, salaries, or numbers of personnel employed by the CIA. 58 59 629 F.2d at 152. As the court relied upon Elliot as its source for Patrick Henry’s words, it references a date of June 15th rather than June 17th of 1788. As noted in fn. 3, above, the correct date appears to be June 17th. CRS-27 President to preserve secrecy for expenditures related to military operations and foreign negotiations. Opponents of the “from time to time” provision, it is clear, spoke of precisely this effect from its enactment. We have no record of any statements from supporters of the Statement and Account Clause indicating an intent to require disclosure of such expenditures.60 The Halperin court also found confirmation for its conclusion that the Statement and Account Clause did not require disclosure of the expenditures at issue from the historical evidence of government practices with regard to disclosure and secrecy before and after the advent of the Constitution. The Committee of Secret Correspondence of the Continental Congress was created on November 29, 1775, Congress resolving to provide for expenses incurred by the Committee in sending out its “agents”.61 When the Committee received information from Arthur Lee, one of its agents, regarding French plans to send arms and ammunition to the Continental Army, it determined to maintain strict secrecy, even from Congress, because of the nature and importance of this information.62 The court notes that the Congress appears to have exerted greater direct control over the Committee after the Declaration of Independence. The camouflaging of the actual recipient and intended use of intelligence funds also appears to have had early usage under George Washington, commander-in-chief of the colonial armies, as reflected in a letter to him from Robert Morris, a member of the Committee of Secret Correspondence, from January 21, 1783. The letter reflects both the provision of a cash account in anticipation of needs which might arise for contingencies and secret service. Drafts drawn from that account appear to have been drawn in favor of member’s of Washington’s family on account of secret services, seemingly a means of concealing the identity of the actual recipients.63 The court also noted a series of statutes creating contingent funds or secret service funds giving the President a means of providing secret funding for foreign intelligence activities.64 For example, in the Act of July 1, 1790, 1 Stat. 128 (10), the Congress created such a fund, appropriating such monies for “persons to serve the United States in foreign parts.” In this act, the President was required to provide a regular statement and account of his expenditures from the fund, but permitted him to not disclose “such expenditures as he may think it advisable not to specify.”65By the Act of February 9, 1793, 1 Stat. 299, 300 (1793), Congress re-enacted the 1790 statute, but modified its language to allow the President to make secret expenditures without specification by making a certificate or by directing the Secretary of State to 60 629 F.2d at 156. 61 3 JOURNALS OF THE CONTINENTAL CONGRESS 392 (1905), cited at 629 F.2d at 157. 62 629 F.2d at 157, citing statement of Committee members Benjamin Franklin and Robert Morris, concurred in by Richard Henry Lee and William Hooper, from II American Archives, Fifth Series, 818-19 (P.Force, ed. 1851). 63 629 F.2d at 157-58, citing 6 U.S. Department of State, DIPLOMATIC CORRESPONDENCE OF 428 (F. Wharton, ed. 1889). THE AMERICAN REVOLUTION 64 629 F.2d at 158-60. 65 1 Stat. at 129. CRS-28 make a certificate for the amount.66 It might be noted that although the specific expenditures from these funds do not appear to have been expected to be disclosed, the statutes did include aggregate numbers for the appropriations for the funds created.67 66 For a more detailed discussion of the statutes and historical precedents upon which the Halperin court relied, see 629 F.2d at 157-60. 67 While they have not provided additional constitutional analysis, other FOIA cases have also involved plaintiffs who have sought disclosure of the executive budget request for intelligence and intelligence-related activities. For example, in Aftergood v. Central Intelligence Agency, 1999 U.S. Dist. LEXIS 18135 (D.D.C. 1999) (Aftergood I), Steven Aftergood, on behalf of the Federation of American Scientists, sought disclosure of the Administration’s total budget request for FY1999 for all intelligence and intelligence-related activities. The CIA denied this request under exemption 1 (on the grounds that the information was properly classified in the interest of national defense or foreign policy under E.O. 12958) and under exemption 3 (on the basis that release of the aggregate figure would tend to reveal intelligence sources and methods which are expressly exempted from disclosure by statute). In its motion for summary judgment, the CIA relied upon statements by from DCI Tenet, one filed as an unclassified exhibit attached to the motion, and two classified statements filed under seal and ex parte for in camera review by the district court . In order to be satisfy the exemption 1 requirements, an agency must show “that the records at issue logically fall within the exemption, i.e., that an Executive Order authorizes that the particular information sought be kept secret in the interest of national defense or foreign policy” and “that [the agency] followed the proper procedures in classifying the information.” Id. at *3-*4. The court found both of these criteria to be satisfied. In so doing, the court rejected the Plaintiff’s argument that the DCI’s determination differed from the President’s and was therefore invalid, based upon a statement three years earlier by a presidential spokesman that, “as a general matter, the President believed “that disclosure of the annual amount appropriated for intelligence purposes will not, in itself, harm intelligence activities.” Id. at *5-*6. The court acknowledged that the President, should he choose to do so, had the authority to disclose the information sought, but noted that he had not done so, nor had he ever addressed the impact of disclosure of the 1999 aggregate intelligence budget request or the amount appropriated for these purposes in FY1999. Similarly, the court found the fact that the President had permitted release of similar information in other years unpersuasive. In the absence of the President’s order to release the information or his withdrawal of the DCI’s authority to make classification decisions, where there is no indication that the DCI has acted in bad faith in his refusal to release the information sought, the court found the DCI authorized to make the classification decision at issue and found that his determination was done properly. Id. at *6. In its de novo determination as to whether the information was properly classified, the court applied a deferential standard: “Thus, summary judgment for the government in an Exemption 1 FOIA action should be granted on the basis of agency affidavits if they simply contain ‘reasonable specificity’ and if they are not called into question by contradictory evidence in the record or by evidence of agency bad faith.” Id. at *8. The court found that DCI Tenet’s declarations satisfied this standard: ... Essentially, DCI Tenet explains that disclosure of the budget request reasonably could be expected to cause damage to national security in several ways ...: (1) disclosure “reasonably could be expected to provide foreign governments with the United States’ own assessment of its intelligence CRS-29 In Aftergood v. Central Intelligence Agency, 355 F. Supp. 2d 557 (D.D.C. capabilities and weaknesses,” Tenet Declaration P 14; (2) disclosure “reasonably could be expected to assist foreign governments in correlating specific spending figures with particular intelligence programs,” Tenet Declaration P 16; and (3) official disclosure could be expected to free foreign governments’ limited collection and analysis resources for other efforts targeted against the United States, Tenet Declaration, p. 18. Id. at *9. The court did not require the DCI to demonstrate certainty as to the damage that disclosure of the requested information would cause to national security. “In the area of intelligence sources and methods, the D.C. Circuit has ruled that substantial deference is due to an agency’s determination regarding threats to national security interests because this is “necessarily a region for forecasts in which the CIA’s informed judgment as to potential future harm should be respected.” Id. at *10. The investigative zeal of foreign intelligence agencies was deemed a matter the CIA appropriately could assume. In concluding that the plaintiff had offered no contrary evidence which undercut the DCI’s “highly fact-dependent determination,” the Aftergood I court found the 1996 nonbinding recommendations of a congressionally-chartered commission of private citizens without classification authority (the Brown Commission) made to the Congress and the President in favor of disclosure did not compel disclosure by the court. In so finding, the court noted that neither Congress nor the President had acted upon those recommendations. The court also noted that the Brown Commission did not consider whether it would recommend disclosure of the 1999 figures under the circumstances which the DCI described in his unclassified declaration. The court found the fact that the DCI had disclosed aggregate intelligence budget figures in other years indicative of his careful, case by case assessment of the impact of each disclosure. “Therefore, the Court must defer to DCI Tenet’s decision that release of a third consecutive year, amidst the information already publicly available, provides too much trend information and too great a basis for comparison and analysis for our adversaries.” Id. at *10-*12. As to the applicability of FOIA exemption 3 to the requested disclosure, the court applied a 2 step analysis, looking at whether the statute relied upon was a statute which fell within the exemption, and whether the withheld material satisfied the criteria of the exemption statute involved. Id., at *12. See CIA v. Sims, 471 U.S. 159, 167 (1985); Fitzgibbon v. CIA, 911 F.2d 755, 761 (D.C. Cir. 1990). The court, relying upon Sims, found it well-settled that the statute relied upon, the 1947 National Security Act’s requirement that the DCI “protect intelligence sources and methods from unauthorized disclosure,” 50 U.S.C. § 403-3(c)(6) (formerly 403-3(c)(d)), was an exemption 3 statute. Id. at *13. The court, again applying a deferential standard, concluded that the DCI had demonstrated that the information sought related to intelligence sources and methods. The necessary connection was found in the “special appropriations process used for intelligence activities.” See Military Audit Project v. Casey, 656 F.2d 724, 745 (D.C. Cir. 1981). The Aftergood court relied upon the determination by DCI Tenet that release of the total budget request would “tend to reveal secret budgeting mechanisms constituting ‘intelligence methods’” to hold that the disclosure of the aggregate intelligence budget request was exempt from FOIA disclosure under exemption 3. Aftergood I, supra, at *14-*16. See also, Center for National Security Studies v. Central Intelligence Agency, 711 F.2d 409, 410-411 (D.C. Cir. 1983) (holding that the court lacked jurisdiction over an interlocutory appeal of a district court order granting the CIA’s summary judgment motion on plaintiff’s FOIA request for the CIA’s 1979 budget for the National Foreign Intelligence Program, holding that it was exempted under exemption 1). CRS-30 2005), the plaintiff sought historical intelligence budget information for the years 1947 through 1970, as well as subsidiary agency budget totals, from the Central Intelligence Agency (CIA) under FOIA. The CIA responded that the by asserting that the information sought was exempt from disclosure under exemption 3, 5 U.S.C. § 552(b)(3), based upon 50 U.S.C. § 403-3(c)(7), which provided that the Director of Central Intelligence shall “protect intelligence sources and methods.” Both parties filed for summary judgment.68 The court granted the CIA’s motion and denied Mr. Aftergood’s motion. The plaintiff argued, in part, that the Statement and Account Clause required publication of the information he requested. Based upon the “unequivocal[]” holding of the U.S. Court of Appeals for the D.C. Circuit in Halperin, which, in turn, relied on Richardson, Judge Urbina rejected plaintiff Aftergood’s contention and held that “a FOIA plaintiff does not have standing under the Statement and Account [C]lause to challenge the constitutionality of CIA budget secrecy.”69 68 In reaching its 2005 decision, the court relies on an earlier case involving the same parties, Aftergood v. CIA, 2004 U.S. Dist. LEXIS 27035 at *1, No. 02-1146, slip op. at 4-5 (D.D.C. Sept. 29, 2004), in which the plaintiff sought disclosure, under FOIA, of the FY2002 aggregate intelligence budget. The court in its 2005 decision states that, in the 2004 decision, it held that 50 U.S.C. § 403-3(c)(7) qualified as a basis for an exemption under exemption 3, and that intelligence budget information “‘relate[d] to intelligence methods, namely the allocation, transfer and funding of intelligence programs.’” 355 F. Supp. 2d at 562 (this purports to quote 2004 U.S. Dist. LEXIS 27035, *4). However, that phrase does not appear in the cited decision.) In the 2005 case, the court framed the issue before it as whether the requested intelligence budget information related to intelligence sources that the DCI had an obligation to protect. 355 F. Supp. 2d at 562. The court relied upon its 2004 holding to conclude that the intelligence budget information sought related to intelligence sources and methods. In so doing, Judge Urbina also cited the Acting Director of Central Intelligence’s declaration that “aggregate intelligence budgets are not identified ‘to protect the classified intelligence methods used to transfer to and between intelligence agencies’, and that “the methods of clandestinely providing money to the CIA and the Intelligence Community for the purpose of carrying out the classified intelligence activities of the United States are themselves congressionally enabled intelligence methods.” Id. 69 355 F. Supp. 2d at 562-63: . . . Specifically the court [in Halperin v. CIA, 629 F.2d 144, 152 (D.C. Cir. 1980),] concluded that “the injury alleged by the plaintiff [is] undifferentiated and common to all members of the public” and therefore, the plaintiff “has not shown the particular concrete injury required for standing.” Id. (internal quotations omitted). The plaintiff laments that the Circuit’s holding in Halperin implies that the CIA never has to report its intelligence expenditures . . . What the plaintiff ignores is that fact that within the same opinion, the court explains that “the absence of any particular individual or class to litigate these claims gives support to the argument that the subject matter is committed to the surveillance of Congress and ultimately to the political process.” Halperin, 629 F.2d at 152 (quoting United States v. Richardson, 418 U.S. 166, 179 . . . (1974)). CRS-31 Conclusions Regarding Statement and Account Clause The Statement and Account Clause appears to impose an affirmative duty upon the Congress to periodically make a statement and account of its disposition of the public funds.70 The questions that arose during the debates upon this clause at the Constitutional Convention and the ratifying conventions went largely to the timing and scope rather than the fact of that obligation. The debates suggest that at least some of the delegates to the Constitutional Convention and the participants in the debates on ratification anticipated that some secrecy might be expected or needed in dealing with military and foreign affairs, and that the language of the clause might be broad enough to permit the Congress to determine what expenditures should be kept secret. Historically, both before and after the Constitution’s advent, some provision in practice or statute appears to have been made to keep the substance of some intelligence information or activities closely-held, as well as the nature and recipients of funds for intelligence activities. The early statutes creating funds for contingent expenses or secret service do seem to include aggregate figures as to the money appropriated, but permit circumspection as to the documentation of expenses from the funds so created. The judicial interpretation of the statement and account clause appears to lay the power to define the sweep of the language in the hands of the Congress. The courts have been consistent in denying standing to those who have sought to challenge the constitutionality of the funding structure of the Central Intelligence Agency Act of 1949 under the Statement and Account Clause to try to access information not disclosed because of the strictures of the 1949 Act. The Richardson Court and its progeny have indicated that the Congress possesses plenary authority to give substance to the language of the Clause and to require such reporting of expenditures as it deems in the public interest. The vehicle by which Congress gives substance to the Clause’s obligations is by statutory mandate. The courts seem to suggest that secrecy as to some expenditures particularly in the area of foreign or military affairs appears to have been anticipated in the crafting of the clause and reflected in contemporaneous practice. Since the early years of the nation, Congress has from time to time, by statute, created funds for expenditures for foreign intelligence activities, and has permitted expenditures from those funds to be made by certificate. Many of the statutes do specify aggregate amounts to be appropriated for the contingent or secret funds in question, but do not require detailed reporting on the nature of the expenditures therefrom. The Central Intelligence Agency Act of 1949 permits transfer of funds 70 As noted in III J. Story, COMMENTARIES ON THE CONSTITUTION OF THE UNITED STATES § 1342, 213-14 (1970), in his discussion of the purpose of Article I, Section 9, Clause 7: The object is ... to secure regularity, punctuality, and fidelity, in the disbursements of the public money.... Congress is made the guardian of this treasure; and to make their responsibility complete and perfect, a regular account of the receipts and expenditures is required to be published, that the people may know, what money is expended, for what purposes, and by what authority. CRS-32 for intelligence purposes from funds appropriated for other agencies, thereby facilitating concealment of the actual intelligence funding levels. It appears that there was some uncertainty among the Framers of the Constitution as to the scope of the obligation the clause imposed upon the Congress. From our review of the constitutional language, its history, and the sparse judicial interpretation of its import, it seems that the courts regard the Congress as having the power to define the meaning of the clause. The courts have not had occasion to address the issue on the merits, and, indeed, might refuse to do so on political question grounds if the issue were presented; however, the judicial interpretation of the Statement and Account Clause to date suggests that a court would be unlikely to find the disclosure of the aggregate intelligence budget constitutionally compelled. Post-Cold War Developments The end of the Cold War had a significant effect on intelligence budgets. Since the country does not face the relentless challenge of an enemy superpower with its own hostile intelligence services, significant reductions in intelligence spending were enacted and criticisms of the continued need for budgetary secrecy were raised anew. Senator Robert Kerrey stated in November 1993: “Openness is the order of the day, and unless a threat as formidable and as lethal as the old Soviet Union comes along, our society and Government will steadily become more open. Our task is to make intelligence more useful to more Americans, not hoard it.”71 Some also maintained that the alleged failures of intelligence agencies to appreciate the essential fragility of the Soviet system or to collect intelligence on the Iraqi nuclear capabilities, warrant a significant overhaul and downsizing of collection and analytical efforts.72 Reductions in defense spending across the board affected intelligence spending in two ways. First, it was assumed that a smaller military force structure reduced requirements for intelligence infrastructure; fewer forces would likely require fewer intelligence support personnel. This argument was countered by some who argued that leaner force structures actually required stronger intelligence support to ensure their most effective and efficient use. Secondly, as the bulk of intelligence funding continued to be “hidden” within the DOD budget, reductions in overall defense spending required either proportional reductions in intelligence programs or disproportionate reductions in non-intelligence programs to compensate for maintaining intelligence spending at existing levels. The latter alternative engendered strong resistance among defense planners already hard pressed to 71 72 Congressional Record, November 10, 1993, p. S15569. Senator DeConcini noted that the intelligence community had failed to analyze the condition of the Soviet Union: “Maybe they had too much money. Maybe there was a wrong direction coming from the executive branch. ...” Therefore, “What does Congress do when some agency does something like that? You start cutting away at their budget, and rightfully so. That is what we have done.” Congressional Record, November 10, 1993, p. S15565. CRS-33 maintain other priority programs.73 In short, as defense spending contracted, it became more difficult to launch new intelligence efforts or even to maintain intelligence programs. This debate over future requirements for intelligence programs was related to (albeit not identical with) the continuing controversy over the desirability of public disclosure of intelligence spending levels. Some opposed to existing or higher levels of intelligence spending consider that public knowledge of the high costs of intelligence spending would lead to demands that they be drastically reduced.74 Efforts to reduce funding levels in intelligence authorization bills are complicated by the question of shared oversight. In 1991, there was concern that reductions in FY1992 intelligence programs were reallocated to other defense programs rather than being used to reduce the deficit.75 With the end of the Cold War, the question of the desirability of making public the extent of the intelligence budget re-emerged in congressional debates and floor votes for the first time since 1975.76 In the consideration of the FY1992 intelligence authorization bill, the SSCI reported a bill (S. 1539) that would have mandated disclosure of three different versions of the total intelligence and intelligence-related budget figure: the aggregate amount requested by the President; the aggregate amount authorized to be appropriated by the conference committee on the Intelligence Authorization Act; and the aggregate amount actually obligated by the executive branch.77 (The SSCI eschewed publication of the amount appropriated because it doubted “that such a figure could be tallied ... by the time a conference committee issued its report, due to the large number of line-items in which the intelligence 73 See Bruce D. Berkowitz and Allan E. Goodman, Strategic Intelligence for American National Security (Princeton, NJ: Princeton University Press, 1989), p. 142. 74 Senator Metzenbaum noted on November 10, 1993, “The argument that disclosure of the intelligence budget total would lead to cuts in that budget is . . . interesting. I have to admit that I think it would do just that. I think the American people would object to spending so much on intelligence. If the budget figure is more than the American people want spent on intelligence, then why should we be spending it?” Congressional Record, November 10, 1993, p. S15555. Representative Sanders, on the other hand, argued for reductions in intelligence spending without reference to budgetary data, stating during the House debate on August 3, 1993: “My job is not to go through the intelligence budget. I have not even looked at it. What I am here to tell you is that you have to tell us that your spy satellites are more important than feeding the hungry children, taking care of people sleeping out in the streets, not rebuilding our educational system, not rebuilding our infrastructure.” Congressional Record, August 3, 1993, p. H5692. 75 The question was debated on the Senate floor on August 2, 1991, Congressional Record, pp. S11971-11977. 76 On October 1, 1975 the House voted down an amendment to the FY1976 Defense Appropriation bill to require disclosure of funds appropriated to the CIA. 77 In a floor statement in November 1993, Senator Metzenbaum indicated that he had introduced the provision in the SSCI markup, with the support of both the then-chairman, Senator Boren, and the Republican vice chairman, Senator Murkowski. Congressional Record, November 10, 1993, p. S15554. CRS-34 appropriation is found.)”78 The Senate Armed Services Committee, which received the bill by sequential referral, noted that these requirements represent major departures from past practices of both Congress and the executive branch that have “profound implications for the conduct of United States intelligence activities and the formulation of intelligence policy which have not been considered in detail by all of the committees of jurisdiction”. The Armed Services Committee proposed that the effective date of these provisions be postponed until FY1993 to allow for detailed consideration.79 The HPSCI version of the bill (H.R. 2038) had no provision relating to public disclosure of the intelligence spending. The conference committee “while agreeing with the objective of the Senate provisions” chose to avoid mandating disclosure by law and stated its hope that the “[Intelligence] Committees, working with the President, will, in 1993, be able to make such information available to the American people, whose tax dollars fund these activities, in a manner that does not jeopardize U.S. national security interests.”80 Section 701 of the final version of the legislation as enacted (P.L. 102-183) stated: It is the sense of Congress that, beginning in 1993, and in each year thereafter, the aggregate amount requested and authorized for, and spent on, intelligence and intelligence-related activities should be disclosed to the public in an appropriate manner. Opposition by the George H.W. Bush Administration may have exerted an important influence on the dropping of mandatory disclosures. Bush, himself a former DCI who had argued against public disclosure in 1977,81 stated upon signing the final version, “Because secrecy is indispensable if intelligence activities are to succeed, the funding levels authorized by the Act are classified and should remain so.”82 This was the Administration position, despite the statement by Robert Gates at his confirmation hearing for the DCI position in September 1991 that “ ... from my personal perspective — and it’s not ultimately my decision, I suppose, but the President’s — I don’t have any problem with releasing the top line number of the 78 S.Rept. No. 102-117, p. 14. 79 U.S. Congress, Senate, 102d Congress, 1st session, Select Committee on Intelligence, Authorizing Appropriations for Fiscal Year 1992 for the Intelligence Activities of the U.S. Government, the Intelligence Community Staff, the Central Intelligence Agency Retirement and Disability System, and for Other Purposes, Report 102-172, October 3, 1991, p. 2. 80 U.S. Congress, House of Representatives, 102d Congress, 1st session, Intelligence Authorization Act for Fiscal Year 1992, Conference Report, H.Rept. 102-327, November 18, 1991, p. 29. House Republican conferees did not support the disclosure provision, but no exception was taken by any Senate conferee. 81 Prepared Statement by George Bush, Former Director of Central Intelligence, printed in Senate, Whether Disclosure of Funds Authorized for Intelligence Activities is in the Public Interest, Hearing, pp. 81-82. 82 U.S. Public Papers of the Presidents. George Bush, 1991, II (Washington: Government Printing Office, 1992), p. 1544. CRS-35 Intelligence Community budget. I think we have to think about some other areas as well. But, as I say, it’s controversial.”83 The following year, the two Intelligence Committees were focused on proposals to reorganize the Intelligence Community and held extensive hearings on the question.84 The Senate version of the FY1993 intelligence authorization bill (S. 2991) included the same “sense of Congress” provision that had previously appeared in the FY1992 legislation. Although the House version (H.R. 5095) again did not contain a similar provision, the conference committee included the Senate provision (as Section 303 in the final version) and there was no dissent among conferees who “reiterate[d] their hope that the intelligence committees, working with the President, will, in 1993, be able to make available to the American people, in a manner that does not jeopardize U.S. national security interests, the total amounts of funding for intelligence and intelligence-related activities.”85 In the midst of the election campaign President Bush signed the legislation (P.L. 102-496) on October 24, 1992, without comment. With the advent of the Clinton Administration in January 1993, some observers believed that the question would be revisited with a different conclusion. Senator Howard Metzenbaum, a member of the SSCI, wrote to the President on February 24, 1993, urging the public disclosure of the intelligence budget. Woolsey, the newly appointed DCI, testified to HPSCI on March 9, 1993 of his concerns regarding making the intelligence budget public: There is no electronic or data fence around the United States or around American citizens. Disclosing that [intelligence spending levels] and the ensuing debate publicly means disclosing it to the people overseas who [sic] we target our intelligence assets on. My real sense of skepticism about this derives principally from the fact that coming forth with a single number communicates really nothing until one knows what goes into the number; and, therefore, proposals either to reduce or to increase that number would require a public debate. In such a debate, it is inconceivable to me that we wouldn’t release information and details as the public and the Congress debated these issues in public and that would be damaging.86 83 U.S. Congress, Senate, Select Committee on Intelligence, 102d Congress, 1st session, Nomination of Robert M. Gates, to be Director of Central Intelligence, S. Hearing 102-799, September 16-20, 1991, Vol. 1, p. 509. 84 See Richard A. Best, Jr., Intelligence Reorganization Proposals, CRS Issue Brief IB92053 (Archived). 85 U.S. Congress, House of Representatives, 102d Congress, 2d session, Intelligence Authorization Act for Fiscal Year 1993, Conference Report, H.Rept. 102-963, October 1, 1992, p. 81. 86 U.S. Congress, House of Representatives, 103rd Congress, 1st session, Permanent Select Committee on Intelligence, Director Woolsey — Future of the Intelligence Community, Hearing, March 9, 1993, p. 13; Woolsey repeated his concerns in a MacNeil/Lehrer Newshour interview on October 19, 1993. CRS-36 Clinton himself responded to Metzenbaum on March 27 asking for the opportunity to “evaluate both the benefits and legitimate concerns which are associated such public disclosure.”87 The House version of the FY1994 Intelligence Authorization bill (H.R. 2330) contained no provision regarding public disclosure, but for the first time in three years the Senate version (S. 1301) also lacked such a provision. According to Senator Arlen Specter, the provision was not included “on the expectation that there would be a stronger resolution compelling disclosure.”88 On August 4, 1993, the House, considering H.R. 2330 under an open rule, debated an amendment offered by Representative Barney Frank mandating disclosure of “the aggregate amounts requested and authorized for, and spent on, intelligence and intelligence-related activities” beginning in 1995. The amendment failed on a vote of 169-264. Many of those who voted for the Frank amendment supported other amendments aimed at reducing the size of the intelligence budget and many observers hoped or feared, depending on their point of view, that making the budget public would lead to public demands for spending cuts. This view was not, however, universal.89 In the Senate an amendment to the FY1994 Defense Appropriation bill (H.R. 3116) was introduced on October 18, 1993 by Senator Daniel P. Moynihan to require “a separate, unclassified statement of the aggregate amount of budget outlays for the prior fiscal year for national and tactical intelligence activities. This figure shall include, without limitation, outlays for activities carried out under the Department of Defense budget to collect, analyze, produce, disseminate or support the collection of intelligence.” Although Senator Moynihan, a critic of the Intelligence Community who had also introduced legislation (S. 1682) to transfer the functions of the CIA to the State Department,90 withdrew the amendment shortly after introducing it, the proposal drew support from Senator Daniel Inouye, then the Chairman of the Appropriations Committee Subcommittee on Defense.91 Three weeks later, on November 10, 1993, the Senate debated an amendment to the FY1994 Intelligence Authorization bill offered by Senator Metzenbaum to include essentially the same “sense of Congress” language as included in the two previously enacted intelligence authorization bills. Although the provision had not been controversial in the Senate on the two earlier occasions, in 1993, the incoming Republican vice chairman of the SSCI, Senator John Warner, spoke out against the proposal. After lengthy debate, the Senate first voted not to table the Metzenbaum 87 Inside the Pentagon, May 6, 1993, p. 15; Senator Metzenbaum provided excerpts from the exchange in the Congressional Record, November 10, 1994, p. S15554. 88 Congressional Record, November 10, 1993, pp. S15557-15558. 89 See, for instance, the remarks of Representative Skaggs, Congressional Record, August 4, 1993, p. H5777. 90 In 1977, however, Senator Moynihan had joined a number of Senators in opposing a move by the SSCI to disclose the intelligence budget. See Minority views of Senators Chafee, Garn, Goldwater, Hathaway, Lugar, Moynihan, Pearson, and Wallop printed in S.Rept. No. 95-274, pp. 13-17. 91 Congressional Record, October 18, 1993, pp. S13805-13808. CRS-37 amendment by a vote of 49-51 and then voted 52-48 to incorporate it into the FY1994 Intelligence Authorization bill (S. 1301). The amendment passed with the support of Senator DeConcini, the new SSCI chairman.92 The Committee of Conference on the two intelligence authorization bills subsequently met, but it did not include the provision regarding public disclosure of the intelligence budget in the final version. The conference report stated: “House conferees were of the view that, in light of the House vote [on the Frank amendment], they could not agree to the inclusion in the conference report of the Senate’s ‘sense of the Congress’ provisions and therefore voted to insist on the House position.”93 Thus, the FY1994 Intelligence Authorization Act (P.L. 103-178) that was signed by President Clinton on December 3, 1993 did not address the question of public disclosure of the intelligence budget. Along with the strong opposition to public disclosure by Senator Warner, the vice chairman of the SSCI (unlike his predecessor, Senator Frank Murkowski, who supported disclosure), an important factor was opposition from the Clinton Administration. During the November 10, 1993 debate, Senator Warner inserted into the Congressional Record sections of a letter from the Office of Management and Budget, dated October 18, 1993, that stated, “... the Administration opposes any change to S. 1301 [the Senate version of the FY1994 intelligence authorization bill] that would disclose, or require the disclosure of, the aggregate amount of funds authorized for intelligence activities. The current procedure that provides for the authorization of appropriations in a classified annex continues to be appropriate.”94 The issue did not disappear. The conference committee had indicated that both intelligence committees had agreed to hold hearings on the question of disclosure in early 1994 “in preparation for thoroughly evaluating a provision to require disclosure of the aggregate intelligence budget figure which may be considered during preparation of the Intelligence Authorization Act for Fiscal Year 1995.”95 Shortly after final passage of the FY1994 authorization bill on November 20, 1993, a group of senior congressional leaders, including Speaker of the House Foley, Senate Majority Leader Mitchell, and other present and former leaders of committees having intelligence oversight responsibilities, signed a letter to the President urging a change in Administration policy to permit public disclosure of intelligence spending. The Members stated that, “The level of intelligence spending (although not the details) 92 Congressional Record, November 10, 1993, pp. S15553-15570. 93 The conference report continued: “Nevertheless, the House conferees did state their willingness to entertain bill language expressing the ‘sense of the Senate’ (as opposed to the `sense of the Congress’ expressing the views of both Houses) in favor of disclosure of the aggregate budget figure, but Senate conferees opposed to disclosure prevented agreement to such modification of the Senate amendment on an evenly divided vote of the Senate conferees. To resolve the impasse, the Senate conferees ultimately agreed to recede to the position of the House.” U.S. Congress, House of Representatives, 103d Congress, 1st session, Intelligence Authorization Act for Fiscal Year 1994, Conference Report, H.Rept.. 103-377, November 18, 1993, pp. 29-30. 94 Congressional Record, November 18, 1993, p. S15561. 95 H.Rept. 103-377, p. 30. CRS-38 must be open to the public.” Further, “[t]he norms of our democratic system require that the public be informed.”96 The President, replying in a December 27, 1993 letter to Representative Glickman, noted his opposition to the proposal in 1993 “because I believed that the cost of disclosure outweighed the benefits.” He added, however, that he had asked Anthony Lake, the National Security Adviser, in concert with the DCI and others, to “look carefully at our position in light of your arguments and in consultation with Congress.” By January 1994, both the executive and legislative branches were committed to review the advisability of making intelligence spending levels public. Congressional hearings were scheduled for 1994 and an NSC-level review was underway. At the HPSCI hearings conducted on February 22-23, 1994, DCI Woolsey repeated his opposition to budgetary disclosure. He emphasized the difficulty of conducting a debate on intelligence programs and priorities in public and his concern that it would be impossible to avoid moving from one aggregate number to disaggregated details that would educate “the rulers of North Korea, Iran, Iraq, Libya, terrorist groups, and others about our plans and programs.”97 The 1994-1995 debate took place in the context of declining budgets and an intelligence community grappling with a world that, in the oft-quoted phrase used by DCI Woolsey in his confirmation hearings, has seen the slaying of the Soviet dragon, but still contained jungles “filled with a bewildering variety of poisonous snakes.” Nevertheless, on July 19, 1994, the House voted (in the Committee of the Whole) 194-221-24 to reject an amendment to the intelligence authorization bill (H.R. 4299) proposed by Representative Glickman, Chairman of the Permanent Select Committee on Intelligence, to amend the National Security Act of 1947 to require annual reports of amounts expended and amounts requested for intelligence and intelligence-related activities. The Senate version did not address the question of making intelligence spending levels public. A similar scenario unfolded in subsequent years. On September 13, 1995 the House voted (in the Committee of the Whole) 154-271-9 to reject an amendment to the intelligence authorization bill (H.R. 1655) proposed by Representative Frank to disclose aggregate amounts requested and authorized for intelligence and intelligence-related activities. Again, the Senate bill had no provision relating to the question. The Senate did support disclosure in 1996, when it passed its version of the FY1997 intelligence authorization bill (S. 1718) with a provision requiring the President to include with the annual budget submission the aggregate amount appropriated for the current year for intelligence and intelligence-related activities and the amount requested for the next year. On May 22, 1996, however, the House voted (in the Committee of the Whole) 176-248-9 to reject an amendment to the 96 Quoted in Tim Weiner, “Disclosure Urged for Secret Budget,” New York Times, November 25, 1993, p. A20. 97 Statement of the Director, February 22, 1994, p. 8. CRS-39 intelligence authorization bill (H.R. 3259) proposed by Representative Conyers to require the President to submit a separate, unclassified statement of the appropriations and proposed appropriations for national and tactical intelligence activities. The subsequent Conference Committee acceded to the House and dropped the provision. On July 9, 1997 the House voted 192-237-5 (in the Committee of the Whole) to reject an amendment to the FY1998 Intelligence Authorization bill (H.R. 1775) offered by Representatives Conyers that would require the President to submit a separate, unclassified statement of the appropriations and proposed appropriations for the current fiscal year, and the amount of appropriations requested for the fiscal year for which the budget is submitted for national and tactical intelligence activities. The Senate voted shortly thereafter, on June 19, 1997, 43-56-1, to reject an amendment to the FY1998 intelligence authorization bill (S. 858) proposed by Senator Torricelli to require the President to submit annual aggregate figures on amounts requested and amounts appropriated for intelligence and intelligence-related activities. Despite these congressional votes interest in and pressure for public release of intelligence spending levels persisted. The Commission on the Roles and Capabilities of the U.S. Intelligence Community, known as the Aspin-Brown Commission, established pursuant to the FY1995 Intelligence Authorization Act (P.L. 103-359), recommended in 1996: ... that at the beginning of each congressional budget cycle, the President or a designee disclose the total amount of money appropriated for intelligence activities for the current fiscal year (to include NFIP, JMIP, and TIARA) and the total amount being requested for the next fiscal year. Such disclosures could either be made as part of the President’s annual budget submission or, separately, in unclassified letters to the congressional intelligence committees. No further disclosures should be authorized.98 Responding to the Commission’s recommendations, on April 23, 1996 President Clinton authorized Congress to make public the total appropriation for intelligence at the time the appropriations conference report was approved.99 Such action was not, however, taken by the Legislative Branch. In October 1997, DCI Tenet announced that President Clinton had authorized him to release the aggregate amount appropriated for intelligence and intelligencerelated activities for FY1997 ($26.6 billion). His press release indicated that the decision was based on two important points: 98 Commission on the Roles and Capabilities of the United States Intelligence Community, Preparing for the 21st Century: An Appraisal of U.S. Intelligence, March 1, 1996, p. 142. It is noteworthy that the Commission included several Members of Congress, including Representatives Goss and Dicks both of whom served on the House Intelligence Committee. 99 Walter Pincus, “Clinton Approves Disclosure of Intelligence Budget Figure,” Washington Post, April 24, 1996, p. A19. CRS-40 First, disclosure of future aggregate figures will be considered only after determining whether such disclosures could cause harm to the national security by showing trends over time. Second, we will continue to protect from disclosure any and all subsidiary information concerning the intelligence budget: whether the information concerns particular intelligence agencies or particular intelligence programs. In other words, the Administration intends to draw a firm line at this top-line, aggregate figure. Beyond this figure, there will be no other disclosures of currently classified budget information because such disclosures could harm national security.100 The press release took note of the lawsuit filed earlier under the Freedom of Information Act and indicated that the President had preferred to take action concerning the declassification of the intelligence budget “in concert with the Congress,” but “the present circumstances related to this lawsuit do not allow for joint action.”101 The following March, Tenet announced that the aggregate amount appropriated for intelligence and intelligence-related activities for FY1998 was $26.7 billion. In the announcement Tenet stated that the determination that “this release will not harm national security or otherwise harm intelligence sources and methods.”102 The release of the figure for FY1998 was, however, the final such release. After litigants had sought to require the release of the amount requested for intelligence (in addition to the amount appropriated which had been made public), Tenet declined to make public the amount appropriated for FY1999.103 Some observers speculate Tenet may have been reluctant to address the substantial additional intelligence funds that were reportedly incorporated in the Supplemental Appropriation Act (P.L. 105277), enacted on October 21, 1998. In any event, no such releases have been made subsequently. Recommendations by the 9/11 Commission and Subsequent Legislation The attacks of September 11, 2001, had a profound affect on intelligence issues. No longer was there a concern to reduce intelligence spending; the goal was to determine why there had been no tactical warning of the attacks that shattered 100 Central Intelligence Agency, Press Release No. 13-97, Statement of the Director of Central Intelligence Regarding the Disclosure of the Aggregate Intelligence Budget for Fiscal Year 1997, 15 October 1997. 101 Ibid 102 Central Intelligence Agency, Press Release No. 03-98, Statement by the Director of Central Intelligence Regarding the Disclosure of the Aggregate Intelligence Budget for Fiscal Year 1998, March 20, 1998. 103 Vernon Loeb, “CIA Won’t Disclose Total Intelligence Appropriation for Fiscal Year,” Washington Post, December 25, 1998, p. A10. See the discussion earlier in this Report, pp. 28-31. CRS-41 thousands of American lives. A series of investigations was launched to fix the blame and to make recommendations for improved intelligence performance. There was a clear disposition in the Executive Branch and in Congress to increase intelligence spending significantly in support of the counterterrorism effort. Many of the recommendations for intelligence reorganization lie beyond the scope of this Report, but some addressed issues of intelligence acquisition and budgeting. Ultimately, a new position, the Director of National Intelligence (DNI) was established. The DNI has been given statutory authorities for developing and determining the national intelligence budget and for ensuring the effective execution of the budget for intelligence and intelligence-related activities.104 In addition, the National Commission on Terrorist Attacks Upon the United States, known as the 9/11 Commission, recommended that the “overall amounts of money being appropriated for national intelligence and to its component agencies should no longer be kept secret.”105 This would be different from the Clinton Administration’s practice in FY1997 and FY1998 when the total appropriated amount for all intelligence and intelligence-related activities was released. The Senate bill introduced in response to the recommendations of the 9/11 Commission (S. 2845) provided that the NFIP would be renamed the National Intelligence Program (NIP) and that the President disclose for each fiscal year the aggregate amount of appropriations requested for the NIP. Furthermore, Congress would be required to make public the aggregate amounts authorized and appropriated for the NIP. (The House bill dealing with intelligence reorganization (H.R. 10) contained no similar provision.) An amendment to remove this provision in the Senate bill was tabled on October 4, 2004 by a vote of 55 to 37. Ultimately, the legislation that was enacted largely in response to the recommendations of the 9/11 Commission, the Intelligence Reform and Terrorism Prevention Act of 2004, P.L. 108-458, did not include the Senate’s provision to make intelligence spending figures public. The issue resurfaced in 2006 when the Senate Intelligence Committee reported its version of authorization legislation for FY2007, S. 3237. Section 107 of the bill would require that the President disclose to the public the aggregate amount of appropriations requested annually for the NIP. The bill would further require that Congress make public the aggregate amount authorized and appropriated by Congress on an annual basis which would presumably include funds provided by supplemental appropriations bills. The bill further mandates a study by the DNI of the advisability of making such information public for each of the 16 elements of the Intelligence Community. No similar provision exists in the House version of FY2007 intelligence authorization legislation (H.R. 5020). In approaching the provision in S. 3237, Congress will likely weigh a number of factors. Some Members believe that not only the spirit of constitutional provisions but also the interests of democracy have always required that intelligence budgets be 104 50 USC 403-1(c)(1)(B); 50 USC 403-1(c)(4). 105 9/11 Commission, p. 416. CRS-42 identified. Even some of those who believed that Cold War conditions necessitated that intelligence budgets be kept secret now argue that conditions have changed and that current enemies would not be able to make use of information on overall levels of intelligence budgets. This view is opposed by others, especially in the House, who believe that the declassification of the intelligence budget could inevitably lead to the compromise of important information on sources and methods. There are, in addition, other factors that Members may wish to take into consideration. First, making the NIP public might lead to the need for a separate intelligence appropriations bill. This, in turn, could prevent the possibility of easy trade-offs between intelligence and non-intelligence defense programs, arguably to the detriment of the intelligence effort. Second, is the fact that actions taken in regard to national intelligence efforts in supplemental appropriations bills would have to be reflected in accounts of intelligence spending arguably with more public justification than would be desirable in some circumstances. In addition, providing information on the NIP but not the MIP could give a false sense of the dimensions of the intelligence effort. Most observers argue that in operational terms, intelligence and intelligence-related activities are mutually supportive, even intertwined, and that considering them separately does not permit an understanding of intelligence capabilities. This could affect both those who want to reduce intelligence spending across the board as well as those who argue that intelligence spending has not kept up with the growth of the threats facing the country. If the intelligence-related activities were to be included, as was the case when FY1997 and 1998 budget levels were made public by DCI Tenet, there would have to be a recognition of the subtle and porous dividing lines between intelligencerelated activities and other targeting and information-gathering and processing efforts. It would be possible to play “budget games” to demonstrate greater or lesser levels of commitment to intelligence by moving individual programs into or out of intelligence-related categories. Conclusion After decades of debate, the issues surrounding the question of public disclosure of the intelligence budget have not changed. There is a question of the degree to which the Constitution requires such budgetary information to be made public. Another question centers on whether limited budgetary data can be made public without leading to detailed revelations of properly classified programs and whether information might be made available to adversaries who will use it against the U.S. Beyond these questions, there is an issue of how to frame an informed public debate on the extent of intelligence spending. How do you provide a sense of how the complex and disparate U.S. Intelligence Community fits together without revealing the extensive detail that almost all observers would consider unwise. Even sophisticated outside analysts are unlikely to appreciate some of the finer (and, in some cases, arbitrary) distinctions among military and national programs and the relationship of non-intelligence communications and reconnaissance programs to the overall intelligence effort. Funding for intelligence-related activities presents special difficulties; budgetary totals can fluctuate from year to year solely because certain CRS-43 DOD programs are transferred into or out of intelligence accounts. Making public only the figure for national intelligence programs would simplify the task, but would not give the public an accurate understanding of the extent of the whole intelligence effort. There will continue to be philosophical and political disagreements concerning how much, if any, information regarding the intelligence budget should be provided. The disagreements may in some cases mask policy objectives. Some argue for as inclusive a number as possible, pointing to the size of the total as the basis for urging its reduction in order to transfer funds to what they consider more important governmental functions or to reduce the federal deficit. Others will seek to show bare-bones intelligence spending and urge more rather than less intelligence spending to cope with the uncertainties of the current international environment. Ultimately, the fundamental issue is whether adequate resources are being devoted to intelligence given the extent of requirements by policymakers, military commanders, and other government officials. The more immediate issue for Congress is how to ensure that there is enough information available to inform this public debate without placing intelligence sources and methods at risk.