Nuclear Safeguards

David Anderson [1]

Foreign Affairs, Defence and Trade Group

Parliamentary Research Service

EXECUTIVE SUMMARY

The Nuclear Non-Proliferation Regime

For fifty years, a major dilemma facing mankind has been how to exploit nuclear energy safely for human benefit while at the same time building, maintaining and developing a shield against nuclear weapons proliferation. It must be emphasised that, despite its significance as the verification tool of the nuclear non-proliferation regime, the IAEA's nuclear safeguards system forms only one element in the global effort to halt proliferation. The non-proliferation regime includes export controls, security assurances, nuclear-weapon-free zone treaties, and a variety of arms control and disarmament agreements. These components are mutually reinforcing, and all must function effectively if proliferation is to be halted. The recent Iraqi nuclear violations reflected a failure of the non-proliferation regime as a whole, with export controls failing to stem the flow of equipment relevant to nuclear weapon development, national intelligence failing to identify clandestine activity, and safeguards focused too narrowly on declared nuclear activities and neglecting possible undeclared developments.

The Nuclear Safeguards System

The technical objective of safeguards is to deter a state from the diversion of nuclear material from peaceful activities to military purposes by creating the risk of early detection. However, this deterrent role is secondary to the political objective of building confidence between nations by showing that states which have undertaken not to acquire nuclear weapons abide by their undertakings. The Nuclear Non-Proliferation Treaty, which entered into force in 1970, gave increased responsibilities to the IAEA and its nuclear safeguards system, introducing the concept of comprehensive or full-scope safeguards. Full-scope safeguards commit a state to accept safeguards on all nuclear material in all its peaceful nuclear activities.

The 'classical' system of safeguards - that is, the system which operated prior to 1992 - was based very much on accurate quantification. It was essentially an audit system, establishing the quantities of nuclear material present within defined areas of the nuclear plant and the changes that take place within defined periods of time. IAEA inspections begin at the stage of the fuel cycle where uranium metal or oxide or uranium hexafluoride are present. In Australia, this means IAEA inspection activity is confined to Lucas Heights. However, under measures introduced since the Gulf War, the IAEA will have to be informed of production figures and of shipments of nuclear material that have not reached this stage of the fuel cycle. In Australia, this means that the two uranium mines plus shipping companies must supply this information. In other words, at present there are two levels of verification: the first, once nuclear material has reached a certain level of purity, involving IAEA inspectors; the second, covering production, exports and imports of uranium ore, and relying largely on matching the figures provided when each shipment leaves Australia and arrives elsewhere. The second verification system is certainly a less rigorous system, but it can also be argued that the nuclear activities it covers are less immediately crucial from the point of view of diversion to weapons.

The classical system of safeguards, as designed, has been a generally successful system. Its major weakness, as exposed by Iraq's activities, has been its narrow focus, safeguarding only those nuclear activities declared by the state, and paying virtually no attention to the possibility of a country developing a clandestine nuclear weapons program entirely separate from its declared nuclear activities. The uncovering of possible clandestine facilities was seen to be the task of states, e.g. through intelligence activities. Discovery of the Iraqi violations was fortuitous, for it directed attention to the limitations and weaknesses of the non-proliferation regime and of safeguards at the very outset of the post-Cold War period, and provided stimulus and justification for a series of changes.

Iraq's ploy had been to bypass the safeguards system, establishing facilities outside the declared, peaceful activities which were under IAEA safeguards. One reason for the narrowness of the classical system of safeguards which failed to anticipate such violations was the traditional tension within the IAEA between transparency and sovereignty, and the Agency's consequent concern with states' rights. International organisations are not independent political entities but rather creatures of the sovereign states which form their membership, and there is continual need to find a balance between transparency of nuclear activities and the legitimate commercial concerns of the state.

The Strengthening of Safeguards Since the Gulf War

The major challenge for the IAEA in the early 1990s was to develop alternative methods for the location of undeclared facilities. The first reforms could be taken quickly as it was realised that the main safeguards document, INFCIRC/153, grants the IAEA powers it had not utilised. Thus, in February 1992, the IAEA reaffirmed the right to undertake special inspections of undeclared facilities where the Agency considers the information provided by the state is not adequate. A more difficult task was to devise ways for the IAEA to obtain reliable information upon which to base a special inspection. One obvious step was to enhance nuclear transparency within the state by increasing provision of information by the state itself, and several initiatives have been taken to achieve this. The most controversial information-gathering step was the invitation by the Director General for all states to assist with the provision of intelligence on nuclear activities. Some member states were fearful that the United States would become the dominant source of information, turning the IAEA into a tool of Washington.

The initial post-Iraq safeguards reforms represented a move from the traditional non-intrusive verification methods. A second group of reforms (known as 'Programme 93+2') is gradually being introduced, and includes measures such as the following:

• The use of environmental samples and analysis with the aim of detecting minute quantities of nuclear material, the presence of which could indicate undeclared nuclear activities. This technique has been used successfully and has great potential.
• The submission by states of 'expanded declarations', containing much greater detail than present declarations concerning their nuclear activities.
• Extended access for inspectors to areas anywhere in and around a nuclear site (present access is limited to predetermined 'strategic points'), to locations included in expanded declarations, and to other locations in order to carry out environmental sampling.

Some of these measures will require additional legal powers, and the conclusion of a model Protocol giving the necessary authority is expected to be agreed shortly.

Australia and Nuclear Safeguards

Australia ratified the Non-Proliferation Treaty in January 1973 and concluded its safeguards agreement with the IAEA in 1974. The Australian Safeguards Office (ASO) was established within the then Australian Atomic Energy Commission, initially to implement the IAEA safeguards agreement. Subsequently its role was extended to encompass Australia's bilateral safeguards system. In 1987 specific safeguards legislation, the Nuclear Non-Proliferation (Safeguards) Act created the statutory position of Director of Safeguards. In Feburary 1990, the ASO was moved to Canberra, a principal objective being the better integration of the Office into the policy-making and administration process.

In June 1994, the ASO was transferred from the Department of Primary Industries and Energy to the Department of Foreign Affairs and Trade (DFAT). This was a logical move in that DFAT has policy responsibility for Australia's network of bilateral safeguards agreements, and plays a key role in the formulation and implementation of uranium export policy. The move has also been mutually beneficial, with ASO being able to contribute significantly to relevant areas of Government policy. DFAT's diplomatic activity, especially through its missions in Geneva, Vienna and New York, plays an essential part in Australia's pursuit of its non-proliferation policies and objectives. The Ambassador in Vienna is Australia's representative on the IAEA's Board of Governors.

The ASO's safeguards activity can be broken into three main functions:

• The safeguarding of nuclear material within Australia, utilising nuclear materials accounting, and focused mainly on ANSTO's site at Lucas Heights.
• Ensuring peaceful use of Australia's nuclear exports. Although this relies to a large extent on information supplied by the participants - that is, there is no independent check of procedures and statistics - these exports of nuclear material seem secure, with confirmation of figures being received at each stage. Yellowcake, or uranium ore concentrate, which forms the bulk of material exported, is considered to be of little immediate strategic significance. As the yellowcake is processed into more strategic materials, however, it will be subject to IAEA safeguards.
• Contributing to the effectiveness of IAEA safeguards. This is accomplished through activities such as participation in international experts' groups, preparation of technical papers, and through the operation of the Australian Safeguards Assistance Program, which was established in 1989-90 to provide support in research and development. Despite the modest budget provided for this support program, it has enabled Australia to make valuable contributions. Recently the program has focused on supporting the development of a new safeguards system, including collaborating with the IAEA to test new techniques and procedures, demonstrating new technology, and the provision of consultancy services. Australia would do well to provide more resources for research with the aim of contributing to international safeguards, especially given the present emphasis on strengthening the safeguards system.

There cannot be a 100% reliable safeguards system, and there will always be limitations in their application; for example, they cannot physically prevent diversion, and they are dependent on the cooperation of member states. But providing expectations of their role are realistic, and continual efforts are made towards their improvement, IAEA safeguards will remain a central component of the nuclear non-proliferation regime.

Under Section (e) of its terms of reference, the Senate Select Committee on Uranium Mining and Milling is required to inquire into and report upon:

The effectiveness of Australia's bilateral agreements with countries importing Australian uranium in ensuring that Australian-sourced uranium is not used in military nuclear technology or nuclear weapons testing activities.

All Australian Obligated Nuclear Material (i.e. under obligations pursuant to one of Australia's bilateral safeguards agreements) is subject to safeguards operated by the International Atomic Energy Agency (IAEA), and verification of non-diversion is carried out by the IAEA. Australia's safeguards activities are additional to the IAEA's international safeguards system, although they derive much of their effectiveness from that system. This paper sets out to:

(a) outline the significance of the safeguards system to the nuclear non-proliferation regime;

(b) provide brief background about the IAEA;

(c) describe the operations of the international safeguards system, and assess its effectiveness;

(d) describe, and assess the effectiveness of, the safeguards system as it operates in Australia.

The author wishes to acknowledge assistance given by the Australian Safeguards Office and the Department of Foreign Affairs and Trade in the writing of this paper.

THE NUCLEAR NON-PROLIFERATION REGIME

The Nuclear Proliferation Threat

Since Enrico Fermi carried out the first controlled chain reaction in 1942, the major dilemma facing humanity has been how to exploit nuclear energy safely for human benefit while at the same time building, maintaining and developing a shield against proliferation of nuclear weapons. This paper examines one of the most significant components of the anti-proliferation shield, the safeguards system administered by the International Atomic Energy Agency.

It is often forgotten how much longer the list of potential nuclear weapon powers was two or three decades ago, when it included countries like the Federal Republic of Germany, Italy, Switzerland, Sweden, Japan, Saudi Arabia and Australia. Proliferation of nuclear weapons has been far slower than expected at that time, when it was widely believed that by now there would be as many as 25 nuclear weapon states. It was assumed then that the appeal of strategic warheads as a deterrent, and also as a status symbol, would cause them to be widely acquired, and that any 'secrets' involved in fission weapon technology would be quickly disseminated among scientists, assisted by the rapid spread of nuclear energy for power generation.

However, from about 1970, when the Treaty on the Non-Proliferation of Nuclear Weapons (NPT) entered into force, the declaration of a nuclear arsenal generally ceased to be a point of national pride, and by now the majority of states have accepted that nuclear weapons would detract from rather than enhance their security. Security is more to be sought in non-proliferation and disarmament pledges, backed up, for those who can afford it, by the new generation of highly accurate and lethal conventional weaponry. Also, while the elements of nuclear fission weapon design have been widely available for many years, the actual engineering and construction of a deliverable bomb still present formidable (and costly) challenges. Thus the growth of the "nuclear club" beyond the five original members has been restrained by the high cost involved, technology transfer issues and mechanisms put in place under the NPT regime. The extraordinary lengths to which e.g. Pakistan and Iraq have had to go to acquire critical technologies illustrate the point. Thermonuclear (fusion or "hydrogen") weapons pose even more daunting technical problems and no state beyond the five nuclear-weapon powers has ever demonstrated any interest in such devices.

Thus the nuclear proliferation outlook at present may seem to many to be generally favourable. There have been no additions to the five declared nuclear-weapon states, and there are only three 'threshold' nuclear-weapon states, Israel, India and Pakistan, which are assumed either to possess nuclear weapons or have the capacity to assemble them at short notice. The number of 'threshold' states has recently fallen in number, with South Africa renouncing nuclear weapons and joining the NPT. However, as will be seen, there is no room for complacency.

Has the Nuclear Proliferation Threat Diminished Since the Cold War Ended?

After more than 40 years, the Cold War came to an end in 1989-90. Have the years since then seen a reduction in the proliferation threat? The answer to this question is that progress has been mixed, with some of the motivations for acquiring nuclear weapons weaker than before and others stronger. Looking first at the factors which have lessened the danger of proliferation:

• The reduction in Cold War tension has given a stimulus to arms control and disarmament negotiations, with the signing of START I in July 1991 and START II in January 1993. The Treaty on Conventional Armed Forces in Europe was signed in November 1990, the NPT was extended indefinitely in May 1995, and recently, in September 1996, the Comprehensive Test Ban Treaty was opened for signing. The general impact of these international agreements has been to reduce the security concerns of most nations.
• In the six years since the treaty on the unification of Germany was signed in September 1990, the number of parties to the NPT has increased from 137 to 182. More important than the actual number is the status of some of the new members, with the nuclear weapons states France and China, the nuclear 'threshold' state South Africa, uranium exporting states such as Niger, Namibia and Kazakhstan, and countries of proliferation concern such as Argentina among those joining.

Factors that have aggravated the proliferation danger are:

• Although the danger of a nuclear attack generating a Third World War has virtually gone, this very fact has removed one key restraint on a number of lesser nations which, during the Cold War, were to a large extent under the discipline or threat of one or the other superpower blocs. As the superpowers scaled down many of their Third World commitments in recent years, states have been given more room to manoeuvre. One result of this new independence has been eruption of a variety of internal conflicts; another has been the increased interest of a handful of states in developing their own weapons of mass destruction. Motives here can vary; some countries, with North Korea being a possible example, will seek nuclear weapons largely as compensation for the diminished role of the major nuclear states in providing security assurances; others, such as Iraq, may see them as a means to promote expansionist policies.
• Fragmentation of the former USSR into a number of independent sovereign states has given rise to a new set of proliferation concerns. Mitchell Reiss summed up these dangers in a recent book:

A weakening of the nuclear custodial system, the possibility of nuclear terrorism, emigration of talented nuclear weapons design and engineering personnel, a growing sense of personal desperation amid domestic disorder and instability, the expanding reach of the mafia, and the collapse of export controls on nuclear materials and technologies could seriously compromise the nonproliferation regime. [2]

• There has been a further change in the nature of the proliferation problem since the Gulf War. Major concern now centres upon the possibility of clandestine uranium enrichment plants being built specifically for security purposes in NPT parties.

It is clear from the above summary that the organisations and mechanisms which exist to counter nuclear proliferation cannot be relaxed even though the threat of a nuclear holocaust, as generally feared in the four decades following the Second World War, is significantly diminished. A brief examination of the non-proliferation system follows.

What Does the Nuclear Non-proliferation Regime Consist of?

The international nuclear non-proliferation regime has been described as "a series of linked arrangements and mechanisms to influence the demand for, and control the ability to procure, nuclear explosive devices." [3] It has evolved progressively during the last 25 years in response to specific political and technological developments.

The regime involves measures to limit both nuclear demand and supply, with those affecting the demand side - reducing the political and security motivation to obtain nuclear weapons - being the most important. While supply-side constraints, for example, on exports of missiles and nuclear components and technologies, may slow down proliferation, developments in Iraq, Israel, South Africa and North Korea have demonstrated such constraints will not prevent a determined proliferator from achieving its goal.

The main components of the non-proliferation regime are:

• The Nuclear Non-Proliferation Treaty, which entered into force on 5 March 1970, provides the backbone to the regime. The essence of the Treaty is Article II, which prohibits the non-nuclear-weapon states from developing or acquiring nuclear weapons. As inducement, the Treaty promotes the peaceful uses of nuclear energy, under effective international safeguards protecting against misuse. Since the revelations in Iraq and North Korea, the image of nuclear self-denial once seen as an attribute of NPT membership has been somewhat tarnished.
• Under Article III, the NPT assigns to the International Atomic Energy Agency (IAEA) the role of verifying that every non-nuclear-weapon state party to the Treaty meets its obligations not to divert nuclear energy from peaceful uses to nuclear weapons.

Under Article III.2 of the NPT, all parties must ensure that safeguards are applied to any nuclear material (defined as 'source or special fissionable material') they export and to all nuclear material produced by exported hardware. After India's explosion of a nuclear device in 1974, the main nuclear exporters began to introduce additional export controls in the belief that the NPT/IAEA regime alone would not prevent nuclear proliferation from taking place in states outside the regime.

• The first move by the nuclear exporters was an attempt to clarify ambiguities in the NPT by drawing up a 'trigger list' of categories of equipment or material which, if exported, would trigger application of IAEA safeguards. This list was produced in 1974 by the Nuclear Exporters Committee, known as the Zangger Committee.
• In 1977 an enlarged group of exporters (with China remaining significantly absent) promulgated the Nuclear Suppliers' Group guidelines, a more stringent list of requirements. These guidelines were revised in 1993 to include restrictions on the export of dual-use items. Particularly in the 1970s and early 1980s, when there was still widespread optimism about the prospects for nuclear power, some developing countries were critical of these export constraints for depriving the Third World of its rights under Article IV of the NPT. A few still remain critical.

Other elements of the non-proliferation regime include:

• Regional structures, such as the nuclear-weapon-free zone treaties in Latin America and the South Pacific. A key development since 1990 has been growth of proposals for similar regional agreements.
• Security assurances, given by nuclear-weapon states to non-nuclear-weapon states.
• National intelligence activities, especially satellite observations, by which states, especially the nuclear-weapon states, monitor the nuclear programs of other countries.
• A variety of arms control and disarmament agreements, from the Partial Test Ban Treaty of 1963 to the recent Comprehensive Test Ban Treaty, make their contribution to non-proliferation.

What is the Place of Nuclear Safeguards in the Non-proliferation Regime?

It is clear from the above that the IAEA and its safeguards system form only one element in the global effort to halt nuclear proliferation. Nevertheless, it must be emphasised that, as the verification tool of the non-proliferation regime, safeguards play an extremely significant role. Writing in the 1990 SIPRI Yearbook, Dr Harald Muller describes the IAEA as the 'institutional and operational mainstay of the non-proliferation regime'. And according to the IAEA Annual Report for 1995, there was, at the end of 1995, a total of 1112 installations (including power and research reactors, as well as fuel fabrication, reprocessing and enrichment plants) in 69 states, which were under safeguards or contained material under safeguards. [4]

International nuclear safeguards have been described as "an arrangement under which the IAEA verifies a pledge by a state that it will not use its nuclear activities to make a nuclear weapon or any other kind of nuclear explosive." [5] The technical objective of safeguards can be stated as follows: to deter a state from the diversion of nuclear material from peaceful activities to military purposes by creating the risk of early detection. But the deterrent role is secondary to the political objective of building confidence between nations by showing that states which have undertaken not to acquire nuclear weapons abide by their undertakings. Safeguards thus allow a state to demonstrate its adherence to its legal undertakings, and provide assurance to the international community that declarations are being complied with, and offer early warnings if they are not. The resulting transparency and confidence is an essential basis for international nuclear trading.

The effectiveness of the safeguards system will be examined more extensively later. At this stage it must be emphasised that the technical aim is, at most, deterrence, not prevention, of diversion or misuse of nuclear material. Safeguards were never designed to prevent proliferation, so it is incorrect to judge them according to this criterion. To some extent, the name 'safeguards' is misleading; 'alarm bell' provides a more accurate indication of the purpose of safeguards.

Safeguards are not a uniform concept. Different models are designed to deal with a variety of situations. The NPT introduced the concept of comprehensive or full-scope safeguards, meaning that safeguards were to apply to all nuclear material in peaceful use in all non-nuclear-weapon states parties to the Treaty, including large and complex fuel-cycles in industrialised countries. As Lawrence Scheinman, now Deputy Director of the US Arms Control and Disarmament Agency, notes:

Parties were required to declare all nuclear material in all peaceful nuclear activity. If all nuclear material subject to safeguards could be accounted for through IAEA inspection, it could be concluded that there was no reason to suspect any diversion from peaceful to proscribed uses. [6]

Most exporters and importers of nuclear-power-related items have agreed in practice that full-scope safeguards are necessary for nuclear cooperation, and most international nuclear trade now takes place within the framework of non-proliferation controls provided by the NPT. The recent accession to the NPT of some major supplier states has consolidated this trend, and even some states not yet in the NPT [7], such as Brazil, have adopted full-scope safeguards.

IAEA safeguards are not linked with either nuclear safety or the physical security of nuclear material, although the IAEA has an increasing role in these matters, for example, in issuing standards, providing assistance and service and arranging for the exchange of experience.

THE INTERNATIONAL ATOMIC ENERGY AGENCY (IAEA)

The IAEA: Background

The nuclear safeguard system is administered by the IAEA, established in 1957 with adoption of the Statute of IAEA. The IAEA grew out of the Atoms for Peace initiative declared by President Eisenhower in 1953, and embodied the dual objectives of enlarging the peaceful contribution of atomic energy and ensuring that the transfer of technology will not serve any military purpose. Ten years later, these two principles were included in the NPT, although the NPT took the concepts further by requiring that the conditions of peaceful use be applied not just to the specific material or equipment exchanged, but to all nuclear installations and material in the state.

In the tense political environment of the early Cold War, states were reluctant to entrust the Agency with tasks that would allow it to play a tangible political role, and at first it was unable to develop a meaningful safeguards document. With a temporary detente in 1964, it became possible to reach agreement on the text of INFCIRC/66, which still operates as the safeguards system for most non-NPT states.

The IAEA: Structure

Although the IAEA is autonomous, with its own statute, and is not a specialized agency of the UN, it is administratively a member of the UN, and reports on its activities once a year to the General Assembly. The two policy-making organs of the IAEA are the Board of Governors and the General Conference.

• The Conference, comprising representatives of all member states, convenes each year for general debate on the Agency's policy, budget and programme. Fifty nine member states participated in the first general conference; the current Agency membership is 124.
• The Board of Governors consists of representatives of 35 member states: 22 elected by the General Conference for two-year periods and 13 designated by the Board from among member states which are advanced in nuclear technology. Australia has held a seat on the Board since the IAEA began.

The Secretariat, headed by the Director General, is the operating organ of the IAEA. As can be seen from the organisational chart in Annex B, the Secretariat is divided into five departments, one of which is safeguards. The Secretariat has the sole responsibility for negotiating safeguards agreements, of which there are 206 in force, but the Director General must submit each agreement to the Board of Governors for its approval. The Secretariat comprises about 2200 professional and support staff. In the safeguards area, one significant group is SAGSI, the Standing Advisory Group on Safeguards Implementation, which is an advisory body to the Director General. SAGSI consists of experts from member states, not IAEA staff.

The IAEA: Functions

The Agency's functions can be divided into two main categories, safeguards and technical cooperation (assisting research on and practical application of atomic energy for peaceful uses). These are succinctly stated in Article II of the IAEA Statute:

The Agency shall seek to accelerate and enlarge the contribution of atomic energy to peace, health and prosperity throughout the world. It shall ensure, so far as it is able, that assistance provided by it or at its request or under its supervision or control is not used in such a way as to further any military purpose.

The legal obligation to invoke safeguards derives not so much from the statute as from other treaty instruments such as the NPT and the Treaty of Tlatelolco [8], and bilateral agreements between nuclear suppliers and their recipients. Once safeguards obligations have been incurred, however, the IAEA statute establishes a framework for implementation. Article XII of the statute sets out the rights and responsibilities of the Agency with respect to any situation where it is authorised to apply safeguards. The most significant of these, and certainly a right that was unique in 1970, is the IAEA's authority to send inspectors into safeguarded states, with "access at all times to all places and data" necessary to verify that there has been no diversion of material to military use.

The actual application of safeguards is conducted on the basis of safeguards agreements negotiated between the Agency and the safeguarded state, within the framework of the statute and Agency safeguards documents such as the 'model' agreement, INFCIRC/153.

In the area of technical cooperation, assistance is provided across the entire spectrum of applied and basic nuclear science. The area is not dominated, as may be thought, by the development of nuclear power stations, as the majority of IAEA members are developing countries, only a few of whom use nuclear power. Thus, in 1995, assistance in applying nuclear techniques to food and agriculture [9] accounted for almost one-quarter of the technical cooperation budget. This was followed, in order of support, by the physical and chemical sciences, health, industry and earth sciences (such as water resources development and environmental management) and radiation safety. Much of this assistance was provided in cooperation with other international organisations such as the UN Food and Agriculture Organisation and the World Health Organisation.

The IAEA: Budget

The regular budget of the IAEA, from which safeguards expenses are paid, has increased from US$5.8 million in 1960 to US$258.6 million in 1995. [10] However, since 1985 the Agency (in common with all UN agencies) has operated under the principle of zero real growth and has had to pay the increased costs of some programs, such as safeguards, by curtailing activities in other programs or by relying on extra-budgetry contributions. The regular budget is funded mainly by contributions made according to a formula of annual assessments on each member state. Under this formula, Australia's assessment amounted to A$4,708,321 in 1995.

The safeguards budget, the largest programme component of the regular budget, was US$88.6 million in 1995, plus a further US$14 million of extra-budgetry funds. The budget for technical cooperation from the regular budget was US$66.5 million. However, in addition to this figure there was US$61.5 million in voluntary contributions to the Agency's Technical Cooperation Fund, used to finance country-specific technology transfer projects.

The other major programme component of the regular budget is radiation safety, US$13.5 million in 1995. The remainder of the regular budget consists of management, administrative and general services, totalling US$90.3 million in 1995.

The IAEA: Is there a Conflict Between its Technical Assistance Functions and Safeguards Activities?

The progressively larger proportion of total Agency resources allocated to safeguards has led to criticism that the Agency's 'promotional' function, that is, its technical cooperation role, is incompatible with its regulatory function, that is, its safeguards activities. The following points can be made:

• The founders of the IAEA did not apparently see any conflict between the two functions, as is shown by the passage quoted earlier from Article II of the Agency's statute. It seems that the safeguarding of nuclear material was regarded as an essential control on 'enlarging the peaceful contribution of atomic energy'. In a sense safeguards are also 'promotional' in that they are pre-conditions for the further growth of nuclear power in the countries to which the IAEA gives assistance.
• David Fischer and Paul Szasz, in their book, Safeguarding the atom, point out that most of the agency's assistance has nothing to do with promoting or proliferating nuclear power:

Most of the IAEA's 'promotional' aid to the Third World relates to ingenious by-products of the main use of nuclear energy. These are isotope and radiation techniques applied in medical and agricultural research and development, medical diagnosis and treatment, plant breeding, food irradiation, controlling insect pests, and so on. [11]

Direct expenditure on fuel-cycle related programs - the areas usually considered by critics to be 'promotional' - is only 7.5% of the regular budget.

• Safeguards activities are undertaken by a separate Department of the Agency, entirely independent of the 'promotional' areas. To separate them further, for example, by forming another organisation, would result in unprofitable duplication of services, and could adversely affect the Agency's effectiveness.

Nevertheless the Agency's dual role has been a problem in that developing countries have insisted that any increase in safeguards funding be matched by an increase in the technical assistance budget.

How Effective is the IAEA?

To assess the effectiveness of the IAEA properly, we should look at both main functions, technical cooperation and the provision of safeguards. This paper is concerned only with the safeguards system, but the very fact that the developing nations have insisted that the technical cooperation budget remains broadly comparable with the safeguards budget suggests this technical support is valued by many countries. [12]

THE NUCLEAR SAFEGUARDS SYSTEM

How do IAEA Safeguards Operate?

Shortly after the NPT's entry into force in 1970, the IAEA Board of Governors established a safeguards committee to advise it on the contents of safeguards agreements to be concluded with non-nuclear-weapon states which had signed the Treaty. The existing safeguards document, INFCIRC/66, was unsuitable for the new requirements contained in the NPT. The document developed by the committee was published by the IAEA in 1971 as an information circular, INFCIRC/153, and this serves as a basis for the structure and content of other comprehensive safeguards agreements. Australia's safeguards agreement with the IAEA is based on INFCIRC/153. [13]

The verification system set out in INFCIRC/153 was developed in the late 1960s and was designed to fit the area where it was felt reassurance was most needed - the industrially advanced states such as West Germany which were or would be capable of making nuclear weapons. These states were keenly sensitive to any further discrimination beyond that already imposed by the NPT division of nuclear-weapon states and non-nuclear-weapon states. A major concern was to make the safeguards less 'burdensome' and 'intrusive', both to diminish the risk of raising operating costs of nuclear plants and to guard against industrial espionage by IAEA inspectors. Thus, where INFCIRC/66 agreements generally applied to nuclear plants, INFCIRC/153 focused on inventories and the flow of nuclear material. Where possible, 'instruments and other techniques' were to do the work of human inspectors whose routine visits would be confined to previously agreed points in the plant. A clear similarity between INFCIRC/66 and INFCIRC/153 was that detection of the diversion of declared and safeguarded nuclear material was the prime concern, and the detection of undeclared nuclear material was a secondary issue.

Even at this early stage can be seen the inherent tension in the system: on the one hand, the Secretariat's need to apply effective safeguards to the member states; on the other, the states with their sensitivity to encroachments on their sovereign rights and their control of the purse-strings of the system.

The first step in providing nuclear safeguards to a country is to establish an inventory listing all nuclear material present in the state. Verification by the IAEA of the information supplied by the state can now be carried out much more efficiently and comprehensively than in the past due to recent technology. The inventory having been established, the safeguards system then consists of three basic elements which together are designed to verify that no nuclear material is diverted for non-peaceful purposes: material accountancy; containment and surveillance; and on-site inspection.

• Safeguards based on these three basic elements is essentially an audit system, with material accountancy as the fundamental procedure establishing the quantities of nuclear material present within defined areas ('material balance areas') and the changes in these quantities that take place within defined periods of time. All information must be accurately collected by the operator and reported to the IAEA by way of the national authority concerned (in Australia, the Australian Safeguards Office). At the end of the specified period the operator takes stock - a 'physical inventory' - and records the amount of any material unaccounted for. [14]

Note that INFCIRC/153 begins at the stage of the fuel cycle where uranium metal or oxide

or uranium hexafluoride (UF6) are present. In Australia, this means IAEA inspection activity is confined to Lucas Heights. However, under measures introduced recently, the IAEA will have to be informed of production figures and of shipments of nuclear material that have not reached this stage of the fuel cycle. In Australia this means that the two uranium mines plus shipping companies must supply this information. In other words, there are two levels of verification: the first, involving IAEA inspectors, once nuclear material has reached a certain level of purity; the second, covering production, exports and imports of uranium ore, and relying largely on matching the figures provided when each shipment leaves Australia and arrives elsewhere. The second verification system is certainly a less rigorous system, but it can also be argued that the nuclear activities it covers are less immediately crucial from the point of view of diversion to weapons.

• Containment and surveillance are measures to complement material accountancy. Containment consists of making use of physical barriers in the plant, such as the pressure vessel of a reactor, so as physically to restrict access to the nuclear material inside and thus prevent clandestine movement of the material. It is usually necessary to use seals on the containers. Surveillance refers to both human observation and use of instruments such as automatic cameras or video recorders to detect any unreported movement of nuclear material or tampering with containment seals or other safeguards devices.
• On-site inspection is the crucial procedure to confirm the veracity of the data supplied, and thus achieve independent verification. For example, the inspector would be present at each physical inventory-taking, or very shortly afterwards, to make a thorough check. [15] The inspector will also look into any causes of 'material unaccounted for' and any differences between reports sent, for example, by shippers and receivers. During these inspections, the IAEA inspector has access only to the previously agreed strategic points of the plant. However, the agreement stipulates that there must be enough of these to provide "the information necessary and sufficient to implement safeguards" and verify information.

INFCIRC/153 lays down elaborate rules for calculating the maximum frequency of inspections required at various types and capacities of plant. Frequency is determined by the amount of nuclear material in a facility and the complexity of a country's fuel cycle, not by the amount of nuclear material in the state as a whole. Thus frequency can vary from one inspection a year for small reactors up to continuous inspections in the case of plants processing or handling large quantities of sensitive nuclear material (for example, separated plutonium or highly enriched uranium). In practice, actual routine inspection effort is kept to the minimum consistent with the Agency's responsibilities.

There is also provision in INFCIRC/153 for making special inspections entailing broader access rights if the information the Agency has received is inadequate for it to meet its verification obligations.

All this is aimed at "the timely detection of diversion of significant quantities of nuclear material". Thus the three grams of plutonium obtained by Iraq from secretly reprocessed spent nuclear fuel, and discovered in May 1991 in the first UNSCOM inspection mission, was well below the threshold for which the IAEA looks; a 'significant quantity' of plutonium is 8 kilograms, roughly denoting the amount needed to make a first generation nuclear weapon, or, more precisely, the amount for which the possibility cannot be excluded that a nuclear explosive could be made. 'Timely detection' relates to the time required to convert diverted material into the components of a nuclear explosive device.

How Effective are IAEA Safeguards?

The following discussion concentrates on the 'classical' safeguards system i.e. the system that operated prior to 1992, and which is still the basis of the system today. Since 1992, however, a number of additions to the 'classical' system have been, and are being, introduced, and these improvements will be discussed later.

Assessing safeguards effectiveness is not easy. One reason is that the purpose of the system has been stated only very generally. Then there is the question of whether effectiveness is to be judged in numerical terms or according to more vague concepts such as political confidence in the system. The approach in this paper is to discuss a number of aspects of the 'classical' system which have been criticised as weaknesses of the system, although most are better seen as limitations. As mentioned earlier, the expectation is widespread that the safeguards system should perform functions it was never intended to do. We would reemphasise the point made earlier that the system is only one mechanism in the non-proliferation regime. And it can be said at the outset that it has been a generally successful system, in its limited way. In speaking of the safeguards agreement, INFCIRC/153, Fischer and Szasz make the comment:

The model NPT safeguards agreement has stood the test of time and has provided the basis of a remarkably workable system of safeguards. [16]

IAEA safeguards broke new ground in showing that sovereign states are willing to accept international inspection. Now nearly 100 percent of the nuclear plants outside the five nuclear weapon states are under IAEA safeguards, with two regional treaties, Tlatelolco and Raratonga, as well as the NPT, requiring members to agree to full-scope safeguards. And in safeguarding these plants, the IAEA has, year after year, reached the conclusion that there had been no diversion of a significant amount of safeguarded material from peaceful uses. [17] Even the Iraq crisis did not spoil this record; indeed, Iraq's choice of a separate, secret facility rather than choosing the easier course of diverting from its standard plants, suggests it feared detection by the IAEA system, a view plainly shared by threshold states which remain outside the NPT.

At the end of 1996 there were 915 facilities under IAEA safeguards (or containing safeguarded material) in 69 states in which there are significant nuclear activities. During 1996, the Agency carried out 2475 verification inspections at facilities. [18]

Limitations and/or Weaknesses in IAEA Safeguards

• Fischer and Szasz argue that "the main inherent weakness of safeguards is the extent to which they depend upon stable international relations...and upon the willingness of governments to cooperate with the IAEA in helping it to apply safeguards to the national nuclear fuel cycle." [19] They point to major gaps existing in areas of tension, such as the Middle East and South Asia. However, in the 12 years since Fischer and Szasz wrote, a number of suspect countries, for example, Argentina, South Africa and Chile, have joined the NPT or, like Brazil, accepted full-scope safeguards, and the signing of the Comprehensive Test Ban Treaty has strengthened the international mood against nuclear weapons.
• The IAEA provides only an inspection mechanism for diversion of declared nuclear material. It cannot physically prevent diversion or, for that matter, the setting up of an undeclared nuclear program. Breaches must be reported to the Board of Governors, with the ultimate resource being the Security Council, to which the Agency has explicit legal access under Article XII.c of the IAEA Statute. It must then rely upon the the major powers to take action.
• Classical safeguards are limited to verifying compliance by a state with its legal obligations, and as such, they cannot read the future intentions of states, let alone control future conduct. They simply report on the existing situation and provide, hopefully, an early warning of any violation. Post-1991 improvements make some attempt to introduce an anticipatory element.
• One inherent limitation to material accountancy is that, as an auditing procedure, it can only disclose what has happened some time previously. And even when something is 'detected', what is being reported is an anomaly, such as a damaged seal or inconsistencies in records, not a definite diversion, let alone a diversion actually taking place. It is unfortunate that the precise numerical values assigned by SAGSI to the technical objective "timely detection of diversion" has been taken as a requirement to be met and as a criterion of effectiveness, rather than as a goal at which to aim. Certainly containment and surveillance measures can shorten detection times, but in the case of plutonium and highly enriched uranium, conversion to a nuclear device may sometimes require only a few days and the goal of 'timely detection' might be unattainable.
• The IAEA is popularly seen as the verifying tool of the NPT, but in fact there are obligations which states make under the NPT which the IAEA does not verify. For example, the IAEA does not verify the state's obligations under Article II of the NPT not to receive the transfer of nuclear weapons or control over them, nor to receive any assistance in the manufacture of nuclear weapons. Breaches here are to be detected by other elements in the non-proliferation regime, especially the export controls. One problem resulting from this system was that the IAEA had no right to obtain information about any new plant at a reasonably early date. In effect, "an entire reprocessing plant or enrichment plant could be exported to an NPT country (such as Iraq) without any notification to the IAEA." [20] As we shall see, this is one matter which post-1991 improvements attempt to take care of.
• Both the NPT (Art. III.1) and INFCIRC.153 (Art. 2) state that safeguards will be applied "on all source or special fissionable material in all peaceful nuclear activities within the territory of the State." The implication of the emphasis - all material in all activities - is that the IAEA has been responsible all along for applying a system that would safeguard undeclared as well as declared activity, although its focus has been only on the latter.
• Until the discoveries were made in Iraq, a certain complacency could be detected in IAEA statements that ensuring the nondiversion of declared nuclear materials was the be-all and end-all of an effective safeguards program. [21] As could be expected in an international organisation dependent on its members for resources and having negotiated with those members the right to intrude on their sovereignty in a unique way, the Agency is an inherently conservative and cautious body. Lawrence Scheinman comments:

Conservatism and self-restraint became internalized to the extent that the agency occasionally gave more ground in negotiating subsidiary arrangements that regulate the operational side of safeguards than perhaps was necessary. Thus, inspectors were not encouraged to raise questions about activities or structures outside defined strategic points when conducting routine inspections. Mind-set was based on verification of what was declared by states being inspected as literally specified by legal agreement. Asking too many questions was said to lead to difficulty with the state, and ultimately at headquarters. The defects in this approach were exemplified in the failure to question Iraqi activities in any of the 70-plus buildings at the Tuwaitha Nuclear Research Center except for their few declared areas. [22]

David Kay, formerly an inspection team leader for the IAEA in the early post-Gulf War nuclear weapons inspections in Iraq, says of the traditional IAEA safeguards culture:

For twenty-five years this culture has been dominated by a drive to achieve greater accuracy in accounting for declared nuclear material rather than by a desire to understand the totality of a nation's nuclear activities and to determine whether any of these are directed toward nonpeaceful ends. [23]

And in 1994, Harald Muller, of the Frankfurt Peace Research Institute, wrote that "the IAEA inspectorate had developed a bureaucratic 'work-to-rule' mentality, which is unhealthy for their watchdog task." [24]

When we look later at recent changes to the safeguards system, it is important to remember that improvements needed to be made not only to the mechanics of safeguards but also to the organisational culture in which these operated.

• David Kay also criticises the IAEA's pre-Gulf War reports on Iraq for seeming to imply that Agency confirmation that no safeguarded material had been diverted meant that the entire Iraqi program was peaceful:

The IAEA failed to communicate clearly the more limited assertion that the material it was charged to look at was in fact still there but that it could not speak authoritatively about anything else in the country. [25]

• Fischer and Szasz point out another consequence of the Agency's over-concern with states rights as "the somewhat excessive secrecy with which the IAEA conducts the safeguards operations." [26] Certainly a high degree of confidentiality is required in reporting on a country's nuclear matters, but Fischer and Szasz suggest that more transparency is possible and would be helpful in gaining the support of public bodies like the US Congress and of the media and public at large.
• The extent of the IAEA's safeguards activities in a particular country has been determined by the quantity of nuclear material in Material Balance Areas and the complexity of the nuclear fuel cycle in the country, on the grounds that the possibility of diversion can never be ruled out, and the Agency would never make subjective judgments as to where the danger of proliferation might be greatest. Non-discrimination between states is one of the underlying principles of the system. One consequence of this is that non-nuclear-weapon states with large nuclear programs account for a very large proportion of the Agency's inspection effort. The irony is that states that have recently become objects of proliferation concern generally have small programs. Thus around 50 percent of the inspection budget goes to just three countries: Germany, Japan and Canada. IAEA inspectors spend five weeks or more each year at each of Canada's nuclear power reactors, but inspect reactors in Iraq, Libya, Israel and North Korea only once or twice a year. [27] Another effect of this overconcentration of inspection effort on countries in which proliferation concerns are minimal is that it has further reinforced the dominant safeguards culture of lack of curiosity. As David Kay points out:

Why be curious and investigative about countries for which there is no plausible reason to believe that they have any intention of engaging in clandestine nuclear activities? [28]

The new criteria for the planning of inspection activities allow the IAEA to take other factors into consideration.

• Many commentators see budget limits as the biggest current restraint on the IAEA's effectiveness. As mentioned earlier, the safeguards budget has been frozen in real terms since 1983, while the amount of nuclear material under safeguards has been growing steadily over the last 15 years at about 12% per year. Although such growth does not demand a commensurate increase in inspection resources - many inventories require essentially the same inspection effort however much they grow - the figures do illustrate the relentless growth in the magnitude of the safeguards task [29]. As far back as 1990, Harald Muller, writing in the SIPRI Yearbook, warned that the zero budgetry growth policy was becoming "increasingly dangerous for the objectives of the NPT" [30], in view of the constant need to keep safeguards technology up to date in order to deal with new more complex facilities. Initially, savings could be made by improving efficiency and eliminating waste, but there are limits to such economies. Reiss and Litwak comment that "given the large returns that a minimum financial investment in the safeguards budget will bring, it is near scandalous that this state of affairs exists." [31]

Changes in the Safeguards System, 1970 to 1990

As peaceful nuclear activities have increased over the decades, there has been steady growth in the number of facilities being safeguarded. Whereas in 1970 only around 150 nuclear facilities were under IAEA safeguards, by 1984 a total of 875 installations of all kinds were safeguarded, and the figure at the end of December 1995 was 1112 installations. Making the application of safeguards even more demanding has been the increasingly complex facilities, requiring continual improvements in safeguards technology to maintain effective verification.

Thus, as Hans Blix, Director General of the IAEA, wrote in 1992, "safeguards are not a static concept...They are aimed at a moving target and always have to be in focus." [32]

One inevitable change has been the growth in the number of inspectors, from 48 in 1970 to around 200 in 1985, and this remains the present figure. Equally important, whereas in the 1970s many of the inspectors were on loan from states, in the 1980s a career inspectorate was established, together with a training section devoted exclusively to improving the training of inspectors.

Many changes have resulted from the five-yearly NPT Review Conferences, one of the most important examples of this being introduction of comprehensive full-scope safeguards as a condition of supply of nuclear materials. At the 1985 Conference, the Australian delegation proposed that the Conference should explicitly endorse full-scope safeguards, but West Europe, led by Germany, opposed the concept. However, the revelation of several illegal nuclear exports in the late 1980s led to full-scope safeguards as a requirement for future nuclear exports being recommended at the 1990 Conference (which unfortunately failed to agree to a final declaration), and eventual endorsement at the NPT Review and Extension Conference in 1995.

The main changes to the system have been technical improvements rather than conceptual changes. This is understandable in view of the conservative attitude of the IAEA as discussed earlier. Also most improvements in instruments can be introduced by agreement between a state and the Agency. In the 1980s, closed-circuit TVs began to replace twin cameras as the chief surveillance instrument. Fibre-optic seals began to replace cup and wire seals as the main containment device, and these in turn have been augmented by ultrasonic seals. CANDU reactors required new safeguards technology, with fuel bundle counters being introduced. Fischer and Szasz emphasise the importance of research help given by member states, and list ten main contributing countries, one of which is Australia.

In 1992 David Kay wrote of the need to change the "mechanically quantitative" approach of the IAEA. [33] Hopefully the impact of the revelation of clandestine nuclear activity in Iraq has been the catalyst to bring about significant changes in approach.

The Strengthening of Safeguards Since the Gulf War

Lawrence Scheinman points out that Iraqi activity should not be seen as a breakdown of the classical system of safeguards as designed, since this covered only declared nuclear material. It was more a circumventing of this system, exposing deficiencies rather than inefficiencies.

Scheinman also comments:

Safeguards have, by and large, done what was expected of them with regard to declared nuclear material. [34]

But the classical system was narrowly conceived, and certainly failed popular expectations, however misguided, that safeguards would uncover all illicit nuclear activities. The focus on declared, peaceful activities meant that the majority of inspectors had been where the largest civil nuclear programs in the non-nuclear-weapon states are located, in Canada, Europe and Japan. It also meant that virtually no attention was paid to the possibility of a nation developing a clandestine nuclear weapon program based on undeclared facilities. [35] Another limitation of conventional safeguards exposed by Iraq was the emphasis on nuclear material rather than both facilities and material. This emphasis on material allowed the construction of facilities without informing the IAEA.

The major challenge for the IAEA in the early 1990s was thus to develop alternative methods for the location of undeclared facilities. [36] Fortunately the first reforms could be taken reasonably quickly, requiring no new regulations but rather the full implementation of the Agency's existing authority base. Paragraphs 73 and 77 of INFCIRC/153 make provision for special inspections of undeclared facilities where the IAEA considers the information provided by the state is not adequate. Historically the right to conduct special inspections of undeclared facilities has not been utilised; the verification of declared nuclear materials has easily been met by less adversarial means when problems have arisen. [37] In February 1992, the Board of Governors reaffirmed the right to undertake special inspections. These are not likely to be frequent. Australia argued that they should be more common than they will, in fact, be, but at least the right to conduct special inspections is acknowledged.

A key question remained: how was the IAEA to obtain reliable information upon which to base a special inspection?

Access to information

• The first step in resolving the information problem was to increase provision of information by the state itself. INFCIRC/153 states that design information on new nuclear facilities should be provided "as early as possible before nuclear material is introduced into a new facility". Over the years, the IAEA had allowed the practice to develop of requiring this information 180 days before nuclear material was introduced. After Iraq, the decision was made that "as early as possible" meant at the time of the decision to construct or modify the facility.
• Another step to enhance nuclear transparency within the state was to expand the existing safeguards system of reporting the export and import of certain nuclear material to include all nuclear material and specified non-nuclear material and equipment. By mid-1996, more than 40 states (including nearly all the major nuclear exporting and importing states) had agreed to supply such information on a voluntary basis.
• A step has been taken to overcome the loophole which allowed Iraq to keep the amount of IAEA inspections far lower than required for good safeguards by distributing its nuclear material over a few small facilities. [38] Whereas previously the number of inspections was largely dependent on the amount of nuclear material present in the facility in question, revised criteria for the planning of inspection activity allow the IAEA to take into account the sum of all nuclear material present in a state.
• The most controversial information-gathering step was the invitation by the Director General for all states to assist with provision of relevant information on nuclear activities. Some member states were fearful that the United States would become the dominant source of information, turning the IAEA into a tool of Washington. In response, the Director General made it clear that information would be sought from a variety of sources and all information would be critically examined. There was also apprehension at the proposal that the IAEA should even develop its own intelligence resource. In fact, intelligence received from member states will be directly evaluated within the Director General's office, where it will be closely held unless circumstances demand that it be shared with other selected staff members. So the evaluation of intelligence information will be conducted separately from the assembly and evaluation of other information available to the Agency under the new safeguards flowing from 'Programme 93+2' (discussed later in the paper).

Harald Muller reports that there is still much concern about the scope of the special inspection authority and the proposed use of intelligence data. [39] David Kay, however, claims the intelligence initiative will be ineffective:

National intelligence communities will continue to be reluctant to provide, on a continuing basis, information to an international bureaucracy that does not even perform background checks on its own staff before or after hiring, has no real communications security, does not have document storage that measures up to national secure storage standards, and lacks any counterintelligence culture or capability. [40]

Kay considers a more effective way of coordinating national intelligence efforts would be to create a Joint Intelligence Staff under the Military Staff Committee, which was established under the UN Charter for the strategic direction of armed forces but rendered moribund by the Cold War. This Joint Intelligence Staff would have the role of analysing intelligence and reporting to appropriate UN bodies, such as the IAEA.

In view of the fact that national intelligence agencies failed to alert the international community to the comparatively large Iraqi nuclear program, there have been understandable doubts that they will detect smaller clandestine programs. However, since Iraq there has been a recognition on the part of both the intelligence and safeguards communities that their activities are highly complementary, a point that was illustrated to very good effect in North Korea. In general, however, there should be realistic expectations as to how much warning the IAEA can provide, even with intelligence assistance and expanded authority for special inspections.

Access to the Security Council

In view of the intrusive nature of some of the reforms and the Agency's own lack of enforcement power, it is important that access to the Security Council for backing and support be assured. The January 1992 statement by the President of the Security Council that the Council "will take appropriate measures in the case of any violation of safeguards notified to them by the IAEA", was therefore encouraging, and would be more so if it was turned into a binding Security Council resolution. Also encouraging was the situation with North Korea in 1993, with the safeguards reforms yielding quick results when put to the test. Good quality intelligence was received concerning clandestine nuclear activities at two sites, supporting the information received by the Secretariat from its own inspection activities. The Board of Governors promptly endorsed the call for a special inspection of the sites; and when this call was refused, the IAEA used its direct line to the Security Council to report the violation. However, it remains to be seen whether collective political will can be generated to carry through enforcement action when the political momentum created by the shock of Iraq has weakened.

Lawrence Scheinman summed up the impact of the Iraqi activities as follows:

Iraq was in a sense an opportune event, for it directed attention to the limitations and weaknesses of the regime and of safeguards at the very outset of the post-Cold War world, and provided the justification for evaluating what would be required to sustain confidence in the regime under new political circumstances. [41]

Post-Gulf War Strengthening of Other Non-proliferation Areas

Iraqi activity reflected a failure of the non-proliferation regime as a whole, with export controls failing to stem the flow of equipment relevant to nuclear weapon development, and national intelligence failing to identify clandestine activities. The weakness of existing export controls had been clearly revealed, with virtually all scientifically advanced countries contributing in some way to Iraq's efforts to develop weapons of mass destruction. In response, one positive result of the Gulf War was a more sustained and high-level attempt to improve the non-proliferation regime generally. Iraq's activities had shifted the focus from declared nuclear facilities to clandestine programs, and also to projects in countries that have not signed the NPT. The latter category is mainly outside the IAEA's reach, but is within the purview of supplier countries.

In 1991 the Nuclear Suppliers Group (NSG) ended more than a decade of inactivity by meeting in The Hague, resulting in a new set of London Guidelines for Nuclear Transfers being released in January 1993. The NSG has continued to meet annually and revise the Guidelines. Some major NSG initiatives are:

• A trigger list of 'dual-use' nuclear related technologies has been issued.
• The NSG members will not export nuclear and nuclear-related items to any country that has not signed an agreement with the IAEA for full-scope safeguards.

While the above may be effective non-proliferation measures, they can be irritating to many NPT members who could see them as denying their right to "the fullest possible exchange of equipment, materials and scientific and technological information" (NPT Art.IV.2).

The Potential of 'Programme 93+2'

The initial post-Iraq safeguards reforms, as described above, represented a move from the traditional non-intrusive verification methods. The second group of reforms, which are gradually being introduced, while still largely focused on the detection of undeclared facilities, aim at increasing transparency in nuclear activity and thus promoting confidence about compliance with the safeguards obligations. There is a wish to avoid reliance on adversarial special inspections, although the right to undertake these has been established.

Following recommendations by SAGSI to the Board of Governors, 'Programme 93+2' (so-named because it originated in 1993 with the intention of completing its proposals two years later) aimed at making safeguards more effective and more cost-efficient. In June 1995, the Board of Governors considered a paper which divided the new proposals into two groups:

Part 1, measures which could be implemented on the basis of the IAEA's existing legal authority; these are being introduced at present.

Part 2, measures which might require additional legal powers. In June 1996, an open-ended Committee of the IAEA Board of Governors (providing for participation of some 60 states) was established to draft a model Protocol giving the necessary legal authority to carry out Part 2 measures. From the outset members were supportive of the proposals. The negotiations have covered such matters as the degree of application to the nuclear-weapon states and the threshold states, together with numerous questions of detail. At the time of writing this paper, the Committee had reached agreement on the text, and the conclusion of the model Protocol is expected shortly.

While Part 1 by itself will not greatly increase the capability to detect undeclared nuclear activities, it does contain some important elements e.g.

• The taking of environmental samples at locations to which inspectors have access under existing safeguards agreements. Environmental sampling and analysis is designed to detect minute quantities of fissile material or certain fission products on vegetation, on the ground, on equipment, or on the inside surface of buildings. Any nuclear manufacturing process loses a fraction of the radioactive process materials to the surrounding environment. The technique has been used successfully in Iraq, South Africa and North Korea, and a number of countries, including Australia, have participated in tests. It will give the Agency some capability to detect certain illicit activities on declared nuclear sites.
• Systematic evaluation of all information available to the Agency concerning states' nuclear and nuclear-related activities. This will take on more importance when combined with Part 2 measures, especially the submission by states of 'expanded declarations', containing much greater detail than present declarations concerning their nuclear and nuclear-related activities and plans.
• The conduct of routine inspections with no advance notice. This will have enhanced value when combined with Part 2 measures providing extended access for inspectors to nuclear sites.

The new safeguards system will, like the classical system, be based on a declaration by the state, with the Agency carrying out inspections to verify elements of that declaration. There are two fundamental differences:

1. The expanded declaration will be much more comprehensive than information currently supplied by states, virtually giving a complete description of a state's nuclear fuel cycle.

Most important, a far wider range of information will be used in assessing the completeness and accuracy of the expanded declaration. Whereas present verification has been confined to routine inspection at declared sites, additional information will now be obtained from such sources as the following:

- environmental sampling and analysis

- extended access for inspectors, allowing access anywhere in and around a nuclear site (present access is limited to predetermined "strategic points"), to any location included on the expanded declaration, and to other locations for the purpose of taking environmental samples

- intelligence and open sources (possibly including commercial satellite imagery)

- IAEA databases

- reports on exports and imports under the voluntary reporting scheme.

The new system will place much greater emphasis on transparency, first from states to the IAEA, but also from the Agency to states, with more informative reports being presented to states on the range of safeguards activities carried out, especially those measures directed at undeclared facilities. [42] One reason why this openness is regarded as necessary is the greater flexibility of the system, with more emphasis on qualitative analysis, rather than total reliance on the precise quantification of the present system, and with inspectors' efforts directed towards more proliferation-sensitive areas. There is the possibility that incentives will be offered to states to participate by reducing the number of routine inspections if the state agrees to a high degree of openness in providing information. It is important here that all states are allowed equal opportunity.

Material accountancy will still be the cornerstone of the system; whatever innovations are made, the system must retain the confidence of members in the reliability of its quantification. A positive feature will be the element of unpredictability arising from such technology as environmental sampling and the possibility of special inspections, with a potential diverter finding it difficult to assess the risk of detection.

It is hoped that automation will progressively take over a number of the activities currently carried out by inspectors. This is not to suggest instruments can ever entirely replace human observation; rather, it will allow inspectors better to concentrate on relevant issues.

One aim of '93+2' is to improve the cost-effectiveness of safeguards, and a possible procedure here is to increase cooperation with national and regional safeguards systems. Already joint inspection arrangements with EURATOM have been introduced in order to lessen the load on IAEA inspectors.

Continuing Problems of the Safeguards System

The post-Gulf War reforms to the system have begun to move safeguards from its narrow material accounting base to the wider foundation of safeguards transparency, with a new focus on the undeclared as well as the declared activities, and greater attention to facilities in addition to nuclear material. However, problems remain in the system, and before moving to Australia's role in safeguards, an attempt will be made to summarise continuing areas of concern. Some of these have been outlined earlier in looking at limitations of the classical safeguards system.

• A major focus of the reform process has been the location of undeclared facilities. But to locate undeclared nuclear activities, an initial source of evidence will be required to enable the Agency to focus its detection efforts. However, there are as yet no high-confidence technologies to locate, especially from a long-range, clandestine enrichment activities. Possibly wide-area environmental sampling, as yet unproven, will eventually provide a solution to this problem, although some enrichment technologies will be more difficult to discover than others. For example, one system under development is isotope separation by laser techniques. This could have the advantages of taking only a small amount of space and energy and of being controlled to minimise detectable emissions. [43]
• The present lack of proven technology for long-range detection places added pressure on the new information sources, especially the intelligence provision, although there is still much uneasiness and scepticism about dependence on other countries' intelligence resources.
• The safeguards reforms emphasise transparency in nuclear activities, but there is traditionally a tension between transparency and confidentiality,which is the prerogative of sovereignty. The IAEA's concern with states rights, which some commentators have considered excessive, has been evident in restraints on inspections. The Director General, Hans Blix, argues that safeguards are not an infringement of a state's sovereignty but in fact help create the conditions necessary for peace, in which sovereignty can be enjoyed. [44] This may be objectively correct, but safeguards are a demand on a high technology industry, and many states are nervous about a possible intrusion of sovereignty. Article III.3 of the NPT is very much an injunction to the IAEA to be circumspect in the way it applies its safeguards. International organisations are not independent political entities, but rather creatures of their client sovereign states, and there is a need now, as Lawrence Scheinman points out, to find some balance between transparency, on the one hand, and legitimate commercial concerns on the other. [45] With political expectations on safeguards extended to include undeclared as well as declared material, the international community should be prepared to accept a broader concept of routine inspections. Nevertheless, continuing tension can be expected between transparency and sovereignty, and it emphasises the extreme importance of national cooperation to safeguards effectiveness.
• In one sense, the biggest threat to the safeguards system remains the countries outside the NPT, more specifically the three threshold states, and the danger that other states will choose to leave the NPT and join those outside. This seems unlikely at present, with a stream of new NPT members in the last five years, but changed world circumstances could conceivably see a country deciding that its security had deteriorated so much that it must have nuclear weapons. In addition, a significant number of mainly underdeveloped states, with little if any nuclear activity, believe that they would stand to lose very little by leaving the Treaty. It is thus extremely important that the IAEA (and of course the NPT) maintains the confidence and participation of its member states in its activities.
• Another continual problem for the IAEA is overexpectation of the role of safeguards. Safeguards can provide the international community with early warning of the risk of proliferation, but safeguards cannot prevent proliferation by a determined state. However, as was seen in the case of Iraq and clandestine nuclear activity, public expectation, making what some would say is a reasonable interpretation of the word 'safeguards', tends to demand a performance beyond the system's established role.
• The IAEA has spent a decade adjusting to frozen budgets, and funding difficulties can be expected to continue. While some of the new reform measures may offer savings in inspection effort, others may well cost more. For example, widespread application of environmental sampling could prove expensive. Resources must be found for the expanding number of facilities and quantities of material under safeguards, while on the horizon are new areas of possible IAEA involvement such as a cut-off convention and verification of material from weapon dismantlement. In 1992 the IAEA made some savings in its new "partnership approach" with EURATOM, which saw a rationalisation of the two organisations' joint inspection rules.
• A long-standing problem has been increasingly large stockpiles of plutonium. There are two separate issues:
• (i) The non-nuclear-weapon states signatories to the NPT can stockpile large quantities of plutonium as by-products of their civil production, providing these stockpiles are periodically inspected by the IAEA. However, these large stockpiles are, at present, only in nuclear weapon states. For example, Japan is one country which has accumulated sizeable stocks of plutonium, but these are held mainly in France and the UK. The comparatively small amount of plutonium in Japan serves as a buffer stock, for use in its civil reactors. The stockpiles are of reactor-grade plutonium which has been separated in the commercial process, and one somewhat confusing issue is the potential of reactor-grade plutonium for use in nuclear weapons. It is theoretically possible for reactor-grade plutonium to be used to make a nuclear weapon (the USA claims to have done this in 1962), but this is technically very difficult, with the result highly dependent on the skills of the designer. In addition, the yield of such a weapon would inevitably be significantly less than the yield of a weapon made with weapons-grade plutonium. The other option is for reactor-grade to be upgraded to weapons-quality plutonium through isotopic separation, although again this is by no means an easy procedure, with possibly only the USA being in command of the technology.
• (ii) The end of the Cold War has resulted in large numbers of nuclear weapons being retired and dismantled, and this has led to increases in stockpiles of HEU and plutonium not covered by safeguards. This accumulation of weapons-grade plutonium is a more dangerous situation than the safeguarded stocks of reactor-grade material discussed above, and there have been widespread calls for greater transparency of these stocks, perhaps with IAEA verification, to lessen the proliferation threat. Talks are taking place at present examining ways of dealing with the growing stockpiles in Russia and the USA. Relatively little progress has been made in the long-standing proposal for international plutonium storage, but in 1994 the USA agreed to put 7 tonnes of plutonium and 15 tonnes of HEU under IAEA inspection. Russia has now also agreed, at least in principle, to place fissile material excess to its military needs under IAEA safeguards.
• Australia has given consent to the reprocessing of uranium by several countries, including Japan, and the question has been asked: should we continue to give consent to reprocessing which may only contribute to growing plutonium stockpiles? It should be noted that Australia's consent is subject to review if circumstances change from the time it was given, and the growth of plutonium stockpiles in non-nuclear-weapon states would be regarded as a significant change.

AUSTRALIA AND NUCLEAR SAFEGUARDS

Background to Australia's Safeguards Role

When the Fraser Government agreed to uranium mining and exports in 1976, the commitment was that certain checks and balances would be put in place. This would give assurance to the Australian people that an undertaking which involved a potentially highly dangerous source of energy would be carried out in as safe a manner as possible. It would also be in line with the serious reservations which the Fox Report had placed on uranium development. The main international regulator was the IAEA, with the safeguards system as the mechanism to ensure the uranium produced was restricted to civil use only. The key internal organisations utilised to provide safety checks were the Office of the Supervising Scientist and the Australian Safeguards Office (ASO).

The Creation of the ASO

With its ratification of the NPT in January 1973, Australia undertook to negotiate an agreement with the IAEA for the introduction of safeguards. This agreement was concluded in 1974, and Article 7 required Australia to "establish and maintain a national system of accounting for and control of all nuclear material subject to safeguards". The agreement also required this State System of Accounting and Control to report its accounting activities to the IAEA and provided for its findings to be subject to independent verification by the IAEA. The Australian Safeguards Office was established within the then Australian Atomic Energy Commission (AAEC) as the Australian State System of Accounting and Control. Although functionally distinct, ASO's staff of two relied on the AAEC for facilities and administrative support and reported to the Minister for Minerals and Energy through the Chairman of the AAEC.

Uranium Mining Begins in Australia

The first Fox Report in October 1976 generally accepted the mining of uranium, providing it was properly regulated and controlled. Fox had been critical of existing safeguards as providing only "an illusion of protection", and in May 1977 Fraser announced a stringent set of safeguards, including the following requirements: [46]

• Australia would export only to countries which accepted the safeguards and inspections of the IAEA pursuant to the NPT (for non-nuclear-weapon states) or a sufficiently comprehensive voluntary offer agreement (for nuclear-weapon states).
• Those countries could not in turn re-export to third countries without Australia's consent at the time of sale.
• Similarly, they would have to agree not to enrich Australia's uranium beyond 20 per cent or reprocess Australia's spent fuel without Australia's prior consent.

The above restrictions remain today, and will be discussed later in the paper.

Development of the ASO

The ASO's workload increased substantially with the decision to export uranium. The first requirement was the negotiation of a number of bilateral agreements with prospective buyers. Currently Australia has 14 bilateral safeguards agreements covering 24 countries. There was also the responsibility to account for the Australian obligated nuclear material (AONM - nuclear material which is subject to obligations pursuant to one of Australia's bilateral safeguards agreements) as it moved through the nuclear fuel cycle under these agreements. At the end of 1995, the AONM abroad amounted to 55,974 tonnes.

Certainly the Chernobyl accident in April 1986 had a devastating effect on reactor sales and in turn on the uranium market, but there has been a steady increase in the export volume of uranium ore in the last four years, although the export unit value of uranium today is only half what it was in the early 1980s.

In 1987, specific safeguards legislation, the Nuclear Non-Proliferation (Safeguards) Act 1987, established ASO's independence outside the departmental structure through the creation of the statutory position of Director of Safeguards. It also provided the Director and the ASO with the powers necessary to carry out the functions of the office.

In February 1990 the ASO was moved to Canberra, a principal objective being the better integration of the Office into the policy-making and administration process. The ASO continued to be associated with the Department of National Development and its successors (Resources and Energy, then Primary Industries and Energy) until June 1994 when it was transferred from the Department of Primary Industries and Energy to the Department of Foreign Affairs and Trade, where it remains. This move partly reflected the increasing importance of international safeguards developments, and was also to take advantage of ASO's experience in international verification activities in establishing the proposed Chemical Weapons Convention Office (CWCO). The CWCO was established in February 1995 and co-located with the ASO.

The Role of the Department of Foreign Affairs and Trade

The move of the ASO to the Department of Foreign Affairs and Trade (DFAT) was logical, as DFAT has policy responsibility for Australia's network of bilateral safeguards agreements, and plays a key role in the formulation and implementation of uranium export policy. This derives from the portfolio's responsibilities for the conduct of Government policy in the field of international security on non-proliferation, arms control and disarmament, a major element of which is strong support for the international non-proliferation regime based on the NPT. The ASO is, of course, able to contribute significantly to relevant areas of Government policy.

DFAT's diplomatic efforts play an essential part in Australia's pursuit of its non-proliferation policies and objectives. Especially through its missions in Geneva, Vienna and New York, the Department seeks to influence the future disarmament agenda, engages in the negotiation of key agreements and contributes to the effective implementation and development of verification systems and export controls. [47] The Ambassador in Vienna, for example, is Australia's representative on the IAEA's Board of Governors.

DFAT points out that Australian uranium export policy supports the international non-proliferation regime in two main ways. [48] First, it gives expression to Australia's general obligation as a party to the NPT to engage in peaceful nuclear cooperation with other parties. Second, Australia makes a substantial contribution to international non-proliferation objectives by ensuring that a significant portion of the world's uranium trade is covered by Australia's stringent conditions.

The Functions of the ASO

The ASO's safeguards activity can be broken into three groups: the first is the safeguarding of nuclear material within Australia; the second is the tracking of Australia's nuclear exports to ensure AONM remains in peaceful use; and the third is evaluating and contributing to the effectiveness of the IAEA's safeguards, both the existing system and possible future measures.

The safeguarding of nuclear material within Australia

This involves nuclear materials accounting, and is the core of the ASO's functions. The main focus of safeguarded nuclear activities in Australia is the ANSTO site at Lucas Heights, and each month an ASO officer carries out an audit of the inventory of nuclear material at the site, and reports are then sent to the IAEA in Vienna. IAEA inspections of the site are also required, and during 1995-96 IAEA inspectors visited Australia on four occasions to carry out seven routine inspections [49], with an ASO officer in attendance to ensure the inspections proceed without problems and to resolve any inconsistencies that are found. One aim of an IAEA inspection is to uncover and evaluate 'material unaccounted for' (MUF), i.e. the difference between the records maintained by the operator and the physical inventory verified by the IAEA. Slight measurement differences are to be expected, but these have balanced themselves out over time. The IAEA has never found cause to comment adversely on Australia's accounting for and control of nuclear material.

Note that the IAEA is concerned to inspect only nuclear activity at Lucas Heights, the site of Australia's one functioning research reactor, i.e. it does not inspect Australia's two uranium mines. This is because safeguards start at the conversion plant stage, i.e. the point in the fuel cycle when uranium hexafluoride leaves a conversion plant. However, the ASO provides details of the amount of nuclear material of all types within Australia, based on information supplied from the mines and Lucas Heights. Indeed, the ASO operates a permit system which controls all nuclear items in Australia.

Ensuring peaceful use of Australia's nuclear exports

Australia exported 5286 tonnes of uranium ore in 1995-96. [50] The basic source data for the tracking of uranium exports is, first, the shipper's weight for each consignment and, secondly, the receiver's measurement confirming the first figure. The safeguards authorities in relevant countries are notified of each export, and these authorities in turn notify the ASO of receipt. The ASO also notifies the IAEA of each export. From time to time discussions are held with ASO's counterparts in other countries. As well as allowing valuable exchange of information, these meetings are used for reconciling all accounting details. Thus a tracking system operates world wide, with confirmation being provided with each move. The efficient operation of the system depends on rapid communication, and the ASO Annual Report records that reports from ASO counterparts "have generally been provided in a timely fashion". This implies that some safeguards authorities are not so diligent, at least on occasions.

Australia's policy requires that AONM be used exclusively for peaceful non-explosive purposes. In practice, this is achieved by using principles of equivalence and proportionality, which are employed by many other countries, including the USA and Canada. Confusion still exists about how the policy operates in the case of exports to Nuclear Weapons States, with the accusation being made that Australia's uranium is used on occasion in nuclear weapons.

The underlying assumption is that all uranium atoms are the same, regardless of source, and it is not possible to track the identity of identical atoms or quantities of nuclear material during processing, which involves the mixing of uranium from a number of sources. The equivalence principle provides that where AONM loses its separate identity because of processing, an equivalent quantity is designated AONM, based on the fact that atoms of the same substance are indistinguishable. In such circumstances, equivalent quantities of the products of such nuclear material may be derived by calculation or from operating plant parameters.

The proportionality principle provides that where AONM is mixed with other nuclear material, and is processed or irradiated, a proportion of the resulting material will be regarded as AONM corresponding to the same proportion as was AONM initially.

The application of the equivalence and proportionality principles provides Australia with the assurance that at all times a quantity of nuclear material precisely equivalent to the quantity exported is identified as being subject to Australian safeguards obligations and treated and accounted for as AONM. While it is possible that some atoms of Australian uranium, mixed with other uranium, may have ended in weapons, this possibility is severely limited by the fact that nuclear-weapon states which processed Australian uranium either had entirely separate facilities for military purposes or excluded civil material when military material was being processed . The official position on this is:

• There is no net benefit to any military activity from the Australian export.
• The argument is academic at present in that the USA, the UK, Russia and France have all ceased production of fissile material for nuclear weapons purposes.

It has been claimed that even if Australia's uranium is used as intended, only for peaceful purposes, its very presence frees other uranium sources for use in, for example, the French military nuclear cycle. The official response to this is that military nuclear requirements would receive priority at all times, so that Australia's exports would benefit only France's civil nuclear activity.

Concerning the so-called 'Nukem scandal' in 1988, the allegation has been made that, at least on this occasion, French authorities abused the equivalence principle by transferring Australia's civil-use requirement to a batch of uranium 'scraps' and enriching the Australian yellowcake without Australian consent. It is possible that some, at least, of the controversy is due to confusion over the word 'scrap', which in a technical sense does not signify waste material but material suitable for, and expected to be, recycled.

It does seem that Australia governments have been, on occasions, simplistic in major public statements on this matter of peaceful use, and have given unrealistic assurances. It is likely that most people interpret the 'peaceful purposes only' requirement in literal terms, and believe that uranium mined in Australia could never enter a weapon.

A possible diversion scenario

It is clear that the verification system covering Australia's uranium exports relies to a large extent on information supplied by the participants, and thus may be regarded by some as inherently vulnerable. The company attaches the seal to the container, the shipper weighs the container, and the importer at the destination provides confirmation of weight and of secure seals. All this information is conveyed to the ASO.

Because there is no independent check of procedures, a situation is theoretically possible where there is collusion between exporter, shipper and importer with the purpose of diverting a quantity of uranium ore. But there are several factors which make such a diversion most unlikely:

• Such a large-scale deception, expecially if repeated, has a high risk of being disclosed.
• While only a few countries have commercially viable sources of uranium, a majority of countries have access, for nuclear weapons purposes, to indigenous uranium ore. Minor quantities may be recovered as a by-product in most geologic terrains. For example, concentrations may be found in heavy mineral sands or mineral phosphates such as monazite. A would-be weapons producer would choose this indigenous source rather than a difficult international deception.

The fact is, yellowcake is considered to be of little immediate strategic significance, and its diversion is not a very useful tactic. SAGSI, as part of its quest to improve the system after the Iraqi disclosures, did consider the possibility of extending safeguards to mines. The idea was rejected for two reasons: the strategic value of the exercise would be small, and the Agency's already restricted resources are better concentrated on more sensitive stages of the fuel cycle. Any nuclear weapons-minded country could acquire fissile material either through uranium enrichment or by building a plutonium production reactor. Of course, a reprocessing plant and a fuel fabrication plant would also be needed, together with a variety of other facilities to develop a weapon design and assemble the device. This emphasises the importance of the post-Iraq safeguards improvements, with their focus on the location of clandestine activities. The development of undeclared facilities will remain expensive and is becoming more difficult.

Australia's contribution to the effectiveness of IAEA safeguards

While the above tasks, with their accurate accounting and reporting requirements, are essential, on the one hand, to the proper functioning of the IAEA, and on the other, to the effective operation of our bilateral arrangements, it is perhaps Australia's activity in the area of analysis and research which, despite the small resources involved, has given us a reputation as a valuable contributor to world safeguards. The ASO's recent efforts are especially appreciated in view of the urgent need to strengthen international safeguards. Australia is taking an active part in current safeguards developments through activities such as:

• Participation in SAGSI (the Standing Advisory Group on Safeguards Implementation). An Australian officer is one of the 18 members of SAGSI, which was the source of much of the IAEA's 'Programme 93+2';
• Australia is one of a small group of countries to offer itself for trials of the 'extended access' concept under 'Programme 93+2';
• Australia has also offered itself for trials of environmental sampling;
• Australia has made senior staff members available as part-time cost-free consultants to the IAEA, for work on 'Programme 93+2' matters.

In addition to making a major contribution to the development of Government policy and the provision of expert briefings and advice, it is important that the ASO maintains effective liason with the IAEA and counterpart safeguards authorities, for example by:

• Participation in IAEA working groups and conferences;
• Formal policy consultation, and discussions on various projects. Because of the expertise of its officers and its generally unbiased record, the ASO is invited to many projects.

Given, on the one hand, the continual need for research and development to generate new safeguards technology and procedures, and, on the other, the freeze on IAEA resources, it is essential that member states offer research support. Australia's support program is known as the Australian Safeguards Assistance Program (ASAP). ASAP is a modest program compared with those of some states, with direct expenditure around $200,000 per annum. There are 16 member state support programs of which the Australian program represents only some 1% of total expenditure involved. The Canadian program, for example, has a budget of $3 million. Because of the limited availability of funds, ASO strategy is to select projects which have a high priority for the IAEA, require only limited funding, and take advantage of expertise available in Australia. Recent areas of research include:

Digital video surveillance of nuclear facilities

Remote monitoring

Analysis of environmental samples

The ASO has successfully demonstrated an Australian-designed and manufactured remote surveillance system at the Australian Nuclear Science and Technology Organisation's research laboratories at Lucas Heights.

Australia has provided some assistance to the former Soviet republics, and a current project aims to help bring Uzbekistan's system for accountancy and control of nuclear materials up to international standards. The ASO is also exploring with Indonesia projects for future research collaboration.

The IAEA has expressed appreciation for "the valuable and vital contribution provided by the Australian Safeguards Assistance Program to the Agency's safeguards efforts." [51]

Since 1985, Australia and Japan have alternated in providing training courses on safeguards matters for personnel of countries in the region. The IAEA has written to the Director of Safeguards asking whether Australia would be willing to conduct another such course in 1998. As the Annual Report comments: "finding funds for a fourth Australian SSAC course in 1998 will not be easy in the current economic climate." [52]

Continual improvement of international safeguards is a key concern, and it is unfortunate Australia has not given the importance (with financial backing) to its contribution in this area as it has, for example, to the conditions it places on its exports of uranium.

Legislation and Agreements

The Nuclear Non-Proliferation (Safeguards) Act 1987, which came into force on 31 March 1987, gives domestic legislative effect to Australia's international obligations under: the NPT; Australia's safeguards agreement with the IAEA; Australia's bilateral agreements with various countries concerning transfer of nuclear material; and the Convention of the Physical Protection of Nuclear Material.

The Act also establishes the statutory office of Director of Safeguards and provides the legislative basis for ASO's activities.

Control over nuclear material and associated items in Australia is exercised under the Act by a system of permits for their possession and transport.

The Nuclear Safeguards (Producers of Uranium Ore Concentrates) Charge Act 1993 imposes an annual charge on yellowcake producers corresponding to a proportion of ASO's operating costs. A charge of $234,000 per producer was established in 1993-94. The legislation is scheduled for review during 1996-97. [53]

Agreements

The Agreement between Australia and the International Atomic Energy Agency for the Application of Safeguards in connection with the Treaty on the Non-Proliferation of Nuclear Weapons was negotiated as a consequence of Australia joining the NPT in 1974. This 25- page, 99-article Agreement provides for the operation of full-scope safeguards in Australia, and requires Australia to establish and maintain a national system of accounting for and control of nuclear material within its territory.

Australia's 14 bilateral safeguards agreements serve as a mechanism to apply conditions additional to IAEA safeguards e.g. with regard to retransfers, high enrichment and reprocessing. This can be shown by briefly outlining the contents of Australia's most recent bilateral agreement, the agreement with Mexico which came into effect in 1992. Looking at the major articles:

Article V provides an undertaking that AONM will be used only for peaceful purposes.

Article VI ensures coverage of all nuclear material by IAEA safeguards.

Article VII provides for contingency arrangements if IAEA safeguards cease to apply in either country.

Article VIII gives an assurance that internationally agreed standards of physical security will be applied to nuclear material in both countries.

Article IX provides that none of the following actions can take place without Australia's prior consent: Transfers to third parties

Enrichment to 20% or more of uranium-235

Reprocessing.

Article X provides for administrative arrangements to be established setting out the procedures to apply in accounting for AONM. It also ensures precautions are taken to preserve commercial confidentiality.

Article XI provides for regular consultations on the operation of the agreement.

Article XII provides for the removal of AONM in the event of a breach of the agreement.

Assessment of the ASO

Reviews of the ASO

Three reviews have been made of Australia's safeguards system or aspects of the system. The first inquiry was made by the Australian Science and Technology Council, which in its 1984 Report Australia's Role in the Nuclear Fuel Cycle concluded that:

Australian uranium and nuclear material derived from it are adequately accounted for and that Australia has the best possible guarantees that such material is being used solely within the civil nuclear programs of Australian customer countries. [54]

In 1988-89, a panel with John Nethercote as convenor reviewed ASO's operation of the bilateral system. A number of structural and other recommendations were made, but the panel's conclusion was also that the system was satisfactory:

On the basis of its enquiries the Panel has concluded that Australian safeguards administration is essentially sound. [55]

In 1991 ASO's procedures for accounting for AONM were subject to an efficiency audit by the Australian National Audit Office, which concluded that:

• Controls over the movements of yellowcake from Australia and the movements of AONM overseas are adequate
• Accounting records are accurately maintained and losses investigated
• Movements of AONM are restricted to those countries with which Australia has bilateral agreements

The Nethercote Review

The reviews have been generally reassuring, but it is worth examining some of the specific recommendations from the Nethercote Report to see to what extent criticisms have been met and recommendations implemented.

A number of administrative and structural improvements were proposed. For example, the recommendation for a more informative Annual Report has been very satisfactorily met. [56] In terms of size, the 1986-87 Annual Report was 35 pages whereas the 1995-96 Report is 116 pages. Much more information is provided on the functions carried out by the ASO in its various programs. And very informative background information is provided on a number of topics, such as the current strengthening of the IAEA safeguards system, the nuclear fuel cycle and Australia's bilateral safeguards agreements. The ASO has in fact set a precedent by publishing in the Annual Report the IAEA's report on its inspection activities. This initiative, which could well be followed by other countries, is aimed at promoting transparency, which is the broad goal of the stronger safeguards system.

More significant was the recommendation that, given the small number of staff and their travel responsibilities, it is important that all staff have a general proficiency in the full range of work duties to avoid gaps when officers were absent. [57] This has largely been accomplished.

The Nethercote Report criticised the ASO for some lack of imagination and initiative in R & D matters. [58] In 1989-90 the Australian Safeguards Assistance Program was established, and the description given earlier in this paper of the projects being currently tackled despite its sadly limited budget should be encouraging.

Problems of the ASO

Consideration needs to be given to the long-term future of the ASO. Although the office of Director is protected as a statutory office, threats to funding and staffing cannot be ruled out. The Nethercote Review in 1989 considered the staff of 14 (including the Director) at that time to be adequate, but forecast that "modest additions...will need to be made periodically in the early 1990s". [59] In fact what has happened since then has been continual attrition down to the current staffing level of eight. The cuts have exceeded the public service average and are contrary to the substantially increased workload. This staff level is not satisfactory in the long-term.

Even if funding is secured, there is still a staff recruitment problem. On page 24 of the Nethercote Report, the point is made that safeguards administration is a highly specialist and complex field and that recruitment in Australia is a problem because "it is difficult to find people with a background in nuclear matters." This recruiting problem has, if anything, deteriorated, as the field has become even more technologically complex, making it even harder for people without an appropriate background to adapt.

The ASO's situation typifies the plight of small, separate groups within large departments. In the case of the ASO, not only is the task important, but if uranium mining is to be expanded in Australia, the strengthening of ASO's capabilities would be important to give reassurance to the Australian people.

Some formula needs to be found to give security to ASO's funding and staffing. Two options are granting independence by establishing the Office as a statutory authority, or amalgamating with the Nuclear Safety Bureau. It is even conceivable that a single nuclear body could be set up to replace the present fragmented system whereby a number of separate organisations have some responsibility for monitoring nuclear science.

CONCLUSION

This decade promises to be the most significant period of development for nuclear safeguards since their inception. The Iraqi and North Korean violations have spurred the international community to accept a more intrusive system, and providing the cooperative momentum is maintained, IAEA safeguards will be far stronger in the year 2000 than they were in 1990. But they will not, and can never, be 100% effective. As David Fischer summed up recently:

We shall never have a completely effective non-proliferation regime nor safeguards that are completely foolproof. That is, of course, no reason for discarding what we have; it points rather to the need constantly to improve the regime and the efficacy of the IAEA's operation. [60]

Two components of the non-proliferation regime in which, hopefully, we will see progress are the nuclear disarmament process and the creation of additional nuclear-weapons-free zones. Probably the important attainments to be made are not in the areas of constraint, such as safeguards and export controls, but at the political level, removing major sources of insecurity and convincing countries that it is not in their interests to acquire nuclear weapons. [61] Australia has made valuable contributions in arms control. Hopefully it can broaden its thrust to help identify and address the underlying motivations of security and status that drive countries to seek nuclear weapons in the first place.

It is essential that the IAEA receives adequate resources both for the expanding number of facilities and quantities of material under safeguards and to introduce new safeguards technologies as needed. There is a tendency for authorities to cut resources to military and police forces when things seem to be going well and the forces not being fully utilised. Then the forces are roundly criticised for not performing effectively when the crisis arrives. It would be most unfortunate if the world community made that mistake with the nuclear safeguards system.

The ASO forms an essential part of the safeguards system, and it is important for governments to recognise the vital part this small office must continue to play, both in administration of safeguards in Australia and in contribution to the system internationally.

 

[Return to Table of Contents]

Footnotes

[1] David Anderson wrote this paper whilst on secondment to the Committee's secretariat.

[2] M. Reiss. Conclusion: Nuclear proliferation after the Cold War. In M. Reiss and R. S. Litwak (ed) Nuclear Proliferation after the Cold War, Woodrow Wilson Press, 1994:341.

[3] J. Simpson. Nuclear arms control and an extended non-proliferation regime. Arms Control and Disarmament (SIPRI Yearbook), 1993: 607.

[4] IAEA Annual Report, 1995: 63,77.

[5] D. Fischer and P. Szasz. Safeguarding the atom. SIPRI, 1985: 3.

[6] L. Scheinman. Safeguards: new threats and new expectations. Disarmament, No.2, 1992: 61.

[7] At the time of writing, only five states are non-NPT members; India, Israel, Pakistan, Brazil and Cuba.

[8] The Treaty of Tlatelolco, which entered into force in April 1968, established a nuclear-weapons-free zone for the Latin American region.

[9] The Agency's food and agriculture program assists member states in improving food security through sustainable and environmentally friendly agricultural development, and operates in such matters as soil fertility, crop production, plant breeding and genetics, animal health, insect and pest control, agrochemicals and food preservation.

[10] Details of the IAEA budget can be found in the IAEA Annual Report, 1995 and the Agency's Accounts, 1995.

[11] D. Fischer and P. Szasz. Op. Cit.: 110.

[12] David Simpson writes in the SIPRI Yearbook, 1990, 554: "The Agency enhanced its standing considerably by its professional handling of the situation after the 1986 Chernobyl nuclear reactor accident. It lent technical help to the Soviet Union, created quickly and effectively an enlarged programme for nuclear safety, and presented the framework for the negotiation and conclusion of two conventions...".

[13] See Annex C for the text of INFCIRC/153.

[14] 'Material unaccounted for' is defined as the difference between the physical inventory and the book inventory, the latter being a calculation of how much nuclear material should be present at this point.

[15] "The core of the standard light water reactor is normally opened at 12-18 month intervals, one-third of the spent fuel is removed and fresh fuel is inserted. This offers the opportunity for taking an inventory of all the nuclear material present in the reactor and its storage areas." D. Fischer and P. Szasz. Op. Cit.: 29.

[16] D. Fischer and P. Szasz. Op. Cit.: 84.

[17] There have been several temporary exceptions when it was unable fully to verify peaceful use at plants in two non-NPT countries, India and Pakistan.

[18] These figures are provided in the IAEA Annual Report. Each Annual Report also contains a list of nuclear facilities and their location in each state, together with the number of reactor units involved.

[19] D. Fischer and P. Szasz. Op. Cit.: 6

[20] Ibid: 80

[21] D. Kay. The IAEA: how can it be strengthed. In M. Reiss and R. S. Litwak (ed) Nuclear proliferation after the Cold War, Woodrow Wilson Press, 1994: 316.

[22] Quoted in Ibid: 316-7

[23] D. Kay. Op. Cit.: 319.

[24] H. Muller. The NPT and the post-1995 non-proliferation agenda. In M. van Leeuwen (ed) The Future of the International Nuclear Non-Proliferation Regime, The Netherlands, 1995: 284.

[25] D. Kay. Op. Cit.: 321.

[26] D. Fischer and P. Szasz. Op. Cit.: 83.

[27] D. Fischer. Towards 1995: the Prospects for ending the Proliferation of Nuclear Weapons, Dartmouth, 1993: 68. The proportion of the inspection budget spent on Germany, Japan and Canada was, during the 1980s, as high as 70%.

[28] D. Kay. Op. Cit.: 332.

[29] Australian Safeguards Office Annual Report, 1995-96: 29.

[30] H. Muller. Prospects for the fourth review of the Non-Proliferation Treaty. World Armaments and Disarmament (SIPRI Yearbook), 1990: 561.

[31] M. Reiss and R.S. Litwak. Op. Cit.: 345.

[32] H. Blix. IAEA Safeguards: new challenges. Disarmament, No,2, 1992: 44.

[33] D. Kay. Op. Cit.: 320.

[34] L. Scheinman. Safeguards: new threats and new expectations: 64.

[35] K. C. Bailey. Strengthening nuclear nonproliferation, Westview Press, 1993: 68.

[36] Another challenge to the IAEA at the time was to find resources to cope with the heavy new workload brought about by the break-up of the Soviet Union and the consequent creation of 14 new countries.

[37] K. C. Bailey. Op. Cit.:70.

[38] Although this loophole was not the cause of the Iraqi violation. Iraq's ploy was to bypass the safeguards system altogether.

[39] H. Muller. The NPT and the post-1995 non-proliferation agenda: 275.

[40] D. Kay. Op. Cit.: 326.

[41] L. Scheinman. Assuring the nuclear non-proliferation safeguards system: 41.

[42] Much of this section came from the following paper: J. Hill, J. Carlson and J. Bardsley. IAEA Safeguards: Perceptions of Effectiveness, INMM 36th Annual Meeting Proceedings, July 1995.

[43] K. C. Bailey. Op. Cit.: 76-7.

[44] H. Blix. Op. Cit.: 45.

[45] L. Scheinman. Safeguards: new threats and new expectations: 70.

[46] A. Cawte. Atomic Australia 1944-1990. NSW University Press, 1992: 154-5.

[47] Department of Foreign Affairs and Trade. Submission to the Senate Foreign Affairs, Defence and Trade References Committee. Inquiry into Nuclear Testing and Non-Proliferation. 1995: 2.

[48] Department of Foreign Affairs and Trade and the Australian Safeguards Office. Submission to the Senate Select Committee on Uranium Mining and Milling. July 1996: 1.

[49] This has been the normal number of inspections for at least the last decade.

[50] Australian Safeguards Office Annual Report, 1995-96: 27.

[51] Australian Safeguards Office Annual Report, 1995-96: 36.

[52] Ibid: 37.

[53] Ibid: 7.

[54] Australian Science and Technology Council. Australia's Role in the Nuclear Fuel Cycle. AGPS, 1984: 17

[55] Report of the Review Panel. Australian Safeguards Office, Canberra, 1989: 2.

[56] Ibid: 29.

[57] Ibid: 14.

[58] Ibid: 12, 16.

[59] Ibid: 19.

[60] D. Fischer. Op. Cit.: 79.

[61] G. Samore. Iraq. In M. Reiss and R. S. Litwak (ed) Nuclear Proliferation after the Cold War : 31.