The past two years have brought radical changes in the nonproliferation policies of the Unites States and a massive international study of proliferation issues. Impending commitments to new nuclear power technologies threaten to give many nations quick access to weapons-usable material, and this prospect has thrust the issue of power-cycle technology and materials, and their control, to the center of proliferation concerns. There is a general desire to achieve a proper balance between the possible benefits of new nuclear systems and the attendant proliferation risks. But there is disagreement, here and abroad, about how to strike this balance, and what policies are appropriate for achieving it.

The current focus of discussion of these issues is the International Nuclear Fuel Cycle Evaluation (INFCE), called for by the United States in 1977. More than 50 countries are participating in this effort, which is due to end in early 1980. In the interim, the U.S. Congress has acted on its own, setting new terms for U.S. participation in nuclear trade. Under the 1978 Nuclear Nonproliferation Act, new and strict conditions are placed on the technology activities of any nation that wants to buy nuclear goods and services from the United States. In effect, the Act is a unilateral step toward revision of the worldwide nonproliferation regime, with the intended source of leverage being the dominant role of the United States in the supply of uranium enrichment services, a vital step in the production of fuel for most reactors.

Serious questions are now emerging, however, as to whether this legislative formulation can fit within a viable and internationally acceptable nonproliferation regime. First, it appears increasingly unlikely that the INFCE deliberations will yield general agreement in support of the Act's tight restrictions on technology. Second, recent changes in world nuclear fuel markets have reduced drastically the control over nuclear fuel supply posited as the basis of U.S. leverage. Third, the Act-especially as seen abroad-does not provide the long-term flexibility and potential for accommodation that will be necessary to reconcile U.S. policy with the widely differing circumstances of other nations.

There is also a more general objection to present policy. To the extent that the United States uses fuel supply unilaterally to coerce policy abroad, it lessens other nations' assurance that they will have secure access to low-enriched uranium fuel. But unless uranium fuel supply is perceived as reliable, these nations will feel a strong incentive to acquire indigenous fuel-cycle facilities or proliferation-sensitive plutonium fuels. Such an inconsistency might be tenable in practice, but only if U.S. control were unchallenged: in the case of the fuel cycle, this will soon not be the case.

Thus there are important deficiencies in current U.S. policy, and the changes now evident in nuclear fuel markets will magnify the problem. To keep pace with events, U.S. policy will have to change, and this will probably call for further amendment of the Nonproliferation Act.


Hindsight often gives great clarity, and looking back we can see that there have always been, in theory, three paths that could be followed in dealing with the weapons proliferation threats posed by sensitive nuclear technologies and materials. The first would be to construct technical and institutional safeguards adequate to allow plutonium fuels and uranium enrichment technologies to be used around the world without presenting unacceptable dangers of diversion to weapons use. A second approach would involve a uniform world regime with severely restricted use of plutonium and enrichment. Finally, a third way would be for nations to recognize a wide variance in reasonable needs for plutonium and enrichment facilities, leading to a situation in which these are regarded as appropriate for use in some countries, but to be avoided in others.

Very early the United States moved down the first path. In the 1950s and 1960s, when the United States held a near-monopoly in nuclear fuel supply and nuclear technology, it led in establishing bilateral controls and the multilateral safeguards of the International Atomic Energy Agency (IAEA). The basic bargain of the 1970 Nonproliferation Treaty (NPT) was that non-nuclear-weapons states were assured access to nuclear technology and fuels in exchange for a no-weapons pledge and safeguards on all nuclear activities. For nations not signing the NPT, access was more limited: they could receive supplies only if they accepted safeguards on each transfer of technology or fuel.

The current debate centers on the reprocessing of spent fuel from present "conventional" reactors, and the ensuing use of plutonium. Though high-enriched uranium is as much a weapons material as plutonium, at present the spread of commercial uranium enrichment is less of a threat due to its technical complexity and high cost. With new technologies under development, and growing nuclear programs, this aspect of the connection between nuclear power and nuclear weapons will eventually surface as a major new challenge to the nonproliferation regime. And there is today one case, in Pakistan, where the fear is that clandestinely acquired enrichment technology may be employed to produce weapons material. But for the most part the question of enrichment facilities is less pressing than that of plutonium-producing technology and facilities.

In the early years of the industry, and indeed up to about 1976, it was the view of the U.S. Atomic Energy Commission and the nuclear industry that the separation of plutonium from spent fuel and its use as a fuel in conventional reactors (called recycling) was a natural component of an efficient light-water reactor cycle, as well as a necessary step in effective nuclear waste disposal and a logical move toward the expected development of breeder reactors fueled by plutonium. Although the United States refrained from exporting reprocessing equipment, the assumption was that plutonium fuels would become a normal part of the international fuel cycle. There was some awareness of its special risks: early agreements provided for U.S. buyback of excess plutonium, over and above fuel cycle needs. But beyond these very loose controls there seems to have been little concern that the proliferation risks of a plutonium fuel cycle were significantly greater than for the uranium-based cycles. Indeed, it was on this basis that the United States declassified the technology for reprocessing of spent fuel to recover plutonium. It was assumed that safeguards would be equally effective for all cycles-in effect decoupling technical decisions in system design from problems of nonproliferation.

Confidence in this approach faltered in the mid-1970s. First, there was the Indian explosion in May 1974. Soon thereafter, new nuclear exporters became active. France and Germany considered or negotiated sales including pilot reprocessing or enrichment facilities to Brazil, Iran, South Africa, Pakistan, and South Korea, marking what a colleague labeled in mid-1976 "the end of the American monopoly."1 These events also stimulated a reexamination of the assumption that safeguards would be adequate to deal with plutonium, and with the spread of enrichment facilities. The issue of technological choice as a fundamental factor in proliferation risk was brought suddenly to the fore.

The issue surfaced in the Congress in early 1976 and found wider expression in the presidential campaign, when both candidates called for a pause in plans to recover plutonium from U.S. spent fuel. The Carter Administration carried this move further, with an indefinite domestic deferral of reprocessing and the commercial development of breeder reactors. These changes had international implications as well. In effect, the United States began trying to reverse the direction it had set for the world in its two decades of technological leadership and to move-at least for a time-toward the second path above, with plutonium to be withheld uniformly from fuel cycle use.

The new caution was based on a conviction that existing safeguards could not provide a probability of detection great enough to deter decisions to acquire nuclear weapons. Or, even if detection were possible, the warning would not come in time for international response prior to the realization of a weapons capability. The issue is not only that sensitive commercial nuclear technologies and materials might be sought out of a desire for nuclear explosives-there are simpler and more expeditious routes to weapons-but that the more immediate availability of a weapons option from commercial programs would alter the domestic political processes affecting weapons acquisition decisions. By acquiring these facilities or materials, nations would move closer to weapons without having to make, or acknowledge, explicit decisions to do so. Moreover, countries would have to assign a higher probability to the possession of weapons by potential adversaries, thereby destabilizing international relations.

Unfortunately, this recognition came just as the technological leverage of the United States was slipping away, to be replaced by vigorous competition with France and West Germany for reactor export sales. At first, there were informal proposals to seek accommodation with these new forces, notably the market-sharing suggestion of Senator Abraham Ribicoff,2 but there was little interest abroad in such arrangements. There also were efforts to use non-nuclear sources of American influence: for example, the United States persuaded South Korea to reverse the purchase of a reprocessing plant from France, and the 94th Congress passed the Symington Amendment to the foreign aid bill, which cuts off U.S. aid to countries importing or exporting enrichment or reprocessing equipment without guarantees of safeguards on all nuclear power activities. But the ultimate focus of legislative action in 1977-78 was toward a new basis for U.S. nonproliferation policy, largely through conditions on the export of fuel enriched in the United States. Enrichment services were seen to be the remaining source of direct U.S. nuclear leverage.

A year and a half before the passage of the Nuclear Nonproliferation Act of March 1978, Senator Ribicoff had proposed threatening "a cutoff of enriched uranium fuel to supplier nations that refuse to join in meeting basic nonproliferation objectives" if there was insufficient interest in the market-sharing approach.3 This possibility was echoed a year later in President Carter's response to a question following his April 7, 1977 announcement of major changes in U.S. nuclear policy: "If we felt that the provision of atomic fuel was being delivered to a nation that did not share with us our commitment to nonproliferation, we would not supply that fuel."4 This condition was incorporated into the U.S. nonproliferation legislation of 1978.

However, if alternative supply sources exist, countries will accept U.S. conditions only if the benefits of doing so outweigh the costs. Therefore let us examine, first, the provisions and effects of the Nonproliferation Act, and then the comparative costs to other nations of compliance and noncompliance with its terms.


Title I of the Nuclear Nonproliferation Act is concerned with the assurance of adequate supplies of low-enriched uranium fuel. It proposes unilateral and multilateral actions intended to "provide a reliable supply of nuclear fuel to those nations and groups of nations which adhere to policies designed to prevent proliferation." This appears to conform with the fuel-assurance obligations of the United States and other nuclear weapons countries under Article IV of the 1970 Nonproliferation Treaty.

However, the Act's fuel-assurance provisions are then linked directly to Title III of the Act, which imposes new and more restrictive nonproliferation conditions on specific export licenses. In addition, any U.S. participation in international undertakings requires explicit congressional approval.

From the foreign perspective, the Act thus appears to subordinate low-enriched uranium fuel assurance to satisfaction of new U.S. nonproliferation requirements. Indeed, the inflexibilities of the Act's requirements, and the complex procedures proposed for implementing them, clearly compromise fuel supply security even for those nations that share the U.S. commitment to nonproliferation.

The evident intent of Title III of the Act is to restrict the spread of reprocessing and enrichment facilities and to discourage commitment to plutonium fuels for recycle or breeders. In regard to export policy for nuclear fuel, the Act specifies new criteria and procedures both for bilateral government-to-government Agreements for Cooperation and for individual licenses of specific exports under those Agreements.

Some of the criteria-application of safeguards to particular exports, prohibition of use for explosives, physical security requirements, and application of safeguards to subsequent generations of material or equipment derived from exports-are comparable to those in prior Agreements and to those imposed by other suppliers. But the Act goes considerably beyond these common conditions and beyond past practices, imposing new conditions on most of its nuclear trading partners and, in many cases, doing so retroactively, even if the exports were covered by existing agreements and contracts.

Specifically, the Act requires prior approval for reprocessing of spent fuel of U.S. origin or fuel used in reactors provided by the United States; similar conditions apply to the enrichment of uranium produced or processed by the United States and to subsequent retransfers of nuclear materials or technology. Severe restrictions are placed on the granting of such approvals. The Act further requires that any recipient of fuel, if it is a non-nuclear-weapons state, must accept safeguards on all nuclear activities (full-scope safeguards).5 It also demands renegotiation of existing Agreements for Cooperation, though the Agreements no longer play such an important role. Instead, the primary locus of U.S. control is shifted to individual export licenses and subsequent approvals.

The implementation of the new conditions in export processes and subsequent arrangements is extremely complex. For example, the issuance of an export license for fuel requires that the Secretary of State, on behalf of the executive branch, certify to the Nuclear Regulatory Commission (NRC) that the export "will not be inimical to the common defense and security or that any export in the category to which the proposed export belongs would not be inimical to the common defense and security because it lacks significance for nuclear explosives purposes." This broad judgment by State will itself require coordination among the Departments of Energy, Defense and Commerce and the Arms Control and Disarmament Agency (ACDA), and notification must also be given to appropriate committees of the House and Senate. The NRC must then make an independent "reasonable judgment" (possibly with public participation) that the criteria above are being met. If the NRC issues the license, it is still subject to judicial review under administrative law.

The key difficulty for some countries is the need for prior approval by the United States for each specific retransfer or for reprocessing of spent fuel which is of U.S. origin or which has been used in a facility supplied by the United States. Such "subsequent arrangements" must be approved on a case-by-case basis by the Secretary of Energy in consultation with the Director of ACDA, the NRC and the Secretary of Defense, with the concurrence of the Secretary of State, and only after justification to appropriate congressional committees. The Act also interposes barriers to the further construction of reprocessing plants: approval is to be given for reprocessing only in facilities which had been commercially in operation (or had received U.S. approval) prior to passage of the Act, unless it can be demonstrated that the reprocessing "will not result in a significant increase of the risk of proliferation." In judging the latter, the likelihood of timely warning to the United States of diversion, well before construction of nuclear weapons, will be of primary importance.6

Without the guidance of experience, or the existence of precise definitions and procedural details, these license and approval processes lack predictability. Changes in the personnel or policies of any of the numerous governmental entities statutorily involved may create new uncertainties. These problems might exist even for countries totally conforming to U.S. nonproliferation conditions. For the great majority who do not yet conform, the prospects for predictable long-term nuclear trade with the United States are severely compromised.

In recognition of the stringency of its conditions and the extent of its departure from past practice, the Act does contain several potentially mitigating provisions. The President may exempt an Agreement for Cooperation from any criterion if he determines that its inclusion "would be seriously prejudicial to the achievement of United States non-proliferation objectives or otherwise jeopardize the common defense and security." Moreover, in decisions on licenses of specific exports or approvals of subsequent arrangements, he can override the NRC by Executive Order in specific cases, subject to 60-day review by Congress, which may deny the export by concurrent resolution. And the executive branch may defer the application of the new conditions in certain defined cases.7 These temporary exemptions and interim flexibilities have been used by the executive to advance near-term nonproliferation objectives.

But what is most notable about these allowances for exemption and deferral is not that they exist but that they are in every case provisional. The Act gives countries no basis for confidence that such special rights will be extended to them consistently and reliably in the long term.


From the beginning, long-range nuclear power planning had been based on the free use of plutonium fuels (under IAEA safeguards). It is thus not surprising that the shift in approach contained in the Nuclear Nonproliferation Act should present difficulties for countries with major nuclear programs. By imposing criteria under which any further commitments to reprocessing or breeders are to be regarded as deviations from U.S. policy norms, the Act suddenly called into question the basic assumptions on which these foreign programs are based and created a threat to their future viability.

Nowhere is this conflict more critical than in the area of nuclear waste management. Operating under technical assumptions which are in retrospect questionable, nuclear leaders abroad (as in the United States) fostered the public impression that reprocessing of spent fuel was essential to responsible management of nuclear waste.8 As a result, the nuclear programs of Japan, West Germany, Sweden and other countries depend, legislatively and from a public opinion perspective, on waste management arrangements that involve reprocessing. By requiring prior U.S. approval for transfer and reprocessing of each batch of fuel of U.S. origin, the Act creates uncertainty not only about the economic and technological success of these foreign programs but also about the near-term political acceptability of the nuclear option itself.

In the longer term, nuclear development in Japan and Western Europe has been predicated on recovery and use of plutonium, and on technological efforts to achieve greater self-sufficiency through the development of plutonium breeder reactors. By controlling spent fuel reprocessing, the United States controls access to plutonium by these nations. Thus, from their perspective, compliance with the new Act threatens their long-term domestic nuclear policies and energy security.

For those countries that are nuclear suppliers, there are additional problems resulting from U.S. policies. The export of nuclear technology and reprocessing services can have an important effect on trade balances; nuclear exports can also be used to bolster international relationships which yield access to sources of raw materials and to future markets for other products. Perhaps most important, exports provide a way to finance and sustain nuclear industries which might otherwise not be cost-effective for the new suppliers, based on domestic markets alone.

Thus, Japan and Western Europe in particular are strongly constrained in their ability to respond to U.S. initiatives by interdependencies among foreign and domestic nuclear policies, commercial and trade relationships, and energy security concerns; all of these factors contribute to the momentum toward the use of plutonium fuel cycles in these areas. On any realistic political analysis, it seems clear that the more serious proliferation risk is centered not in Japan or Western Europe, but rather in the developing countries that might be swept into a plutonium-oriented system of technologies and fuels. Nevertheless, the new U.S. policy, if strictly interpreted, unilaterally imposes uniform and stringent conditions on all countries. It is hardly surprising that the initial response by the industrialized countries whose programs would be most severely affected was one of categorical rejection.

On technical and economic grounds, the developing countries might seem to pose fewer problems for U.S. policy. Here the proposed U.S. bargain-ample supplies of low-enriched uranium fuel in exchange for limitations on sensitive technologies and more comprehensive safeguards-should be a better deal. For some decades yet, these countries will not be able, technologically or economically, to make extensive use of plutonium or indigenous reprocessing or enrichment facilities, but will continue to depend on industrialized supplier states now placing relatively stringent conditions on access to such sensitive technologies and materials. Moreover, although the technology path followed by industrialized countries affects desires and technological opportunities in developing countries, the developing countries do not yet have large commitments to particular technologies; nor do they have public opinion barriers to altering their nuclear energy strategies.

Thus, the costs of their compliance with U.S. conditions are relatively low. But though the costs of compliance are low, so also are the costs of avoiding dependence on the United States. By late this year, or early next, it may be possible for most of these countries to arrange fuel supply without recourse to the United States.


As recently as 1977, the United States provided over 70 percent of the enrichment services for foreign reactors (outside the "centrally planned economies" of the Soviet Union, the COMECON countries, and China). Though there were some voices to the contrary, many of those concerned with the formulation of U.S. policy in 1976-78 appear to have assumed that this dominant position would persist for some years yet, giving time to exercise U.S. leverage. True, new European enrichment suppliers, URENCO and EURODIF, were coming into operation.9 But they were not expected to result in major changes: high demand levels and tight market conditions would limit the ability of nations to alter their enrichment contracting patterns. Those concerned about the effects of new supply possibilities on the U.S. position in the market generally focused on the more distant future, on the possible entry in the late 1980s or the 1990s of South Africa, Brazil, or Japan.

Since then, however, there has been an unexpectedly high rate of slippage in the nuclear growth plans on which enrichment contracting and enrichment facility construction were based; there has also been a substantial increase in enrichment contracting between West European nations and the Soviet Union. The current result is a considerable oversupply of enrichment and the creation of numerous opportunities for consumer nations to alter enrichment supply patterns to fit their political or other interests, and, in some cases, to escape further dependence on the United States.10

Taking the Soviet Union first, its contracts with Western Europe have increased to the point where they now provide, through 1990, for quantities of enriched fuel comparable to those contracted for from the United States in this period. The largest quantities go to the countries with the largest nuclear programs: France, Spain, West Germany and the United Kingdom. The Federal Republic could cover nearly all of its needs this year with scheduled Soviet imports (due to over-contracting), but the fraction of needs provided by the U.S.S.R. drops rapidly to about 40 percent by the end of 1982. France has less dependence on the U.S.S.R.-and on the United States for that matter-and indeed becomes essentially self-sufficient with the startup of the EURODIF facility. Overall, the Soviets will provide 50-80 percent of Western Europe's enriched fuel needs for each of the next four years. The U.S.S.R. does not yet appear to have contracted to supply enrichment services outside Europe, though there have been recurrent reports of negotiations with Japan.

The second major source of new enrichment supply will be the EURODIF facility in France. When this facility was planned, it was expected that the equity partners (France, Italy, Spain, Belgium and Iran) would utilize virtually all of its output. However, actual reactor deployments have not supported these demand expectations, and EURODIF will now have capacity in excess of the equity partners' needs even after contracts with Japan, Switzerland, and the Federal Republic are met. If the equity partners maintain their present contracts to purchase enriched uranium from the United States and the U.S.S.R., the EURODIF excess capacity in the 1980-84 period could range from nearly five to as much as seven million "separative work units" (SWUs) annually, enough to provide annual fuel reloads for 45 to 60 gigawatts/electric (GWe) of reactor capacity (comparable to total West European capacity over this period).11 This excess is projected to decline gradually over the remainder of the 1980s, but even under optimistic nuclear growth assumptions the excess does not vanish until the end of the decade. If the equity partners were to drop their contracts with the United States, substituting EURODIF capacity, the surplus available for others would be reduced to about five million SWU in 1981 and about one million SWU in 1987.

In this situation, it will be technically possible for a number of countries to find new supply arrangements that avoid, or significantly reduce, dependence on the United States-at least for a time. Indeed, as of 1980, the combination of excess capacity and Soviet exports is such that it would be physically possible for all non-U.S. demand to be satisfied by non-U.S. sources. This complete independence could be sustained only until about 1984 (perhaps 1987 if all excess enrichment capacity were used to produce enriched fuel which was then stockpiled) when projected reactor growth would begin to strain planned non-U.S. capacity.12 But if the nuclear growth rate declines further, or if new enrichment capacity is constructed, independence from U.S. supplies could be extended indefinitely.

Which countries might find it advantageous and possible to secure alternatives to dependence on the United States? The list clearly includes some of the major nuclear consumer and supplier states and some of the developing countries. Countries like West Germany and Japan may wish to free a fraction of their fuel from U.S. conditions (and from uncertainties about future conditions) in order to make decisions on breeders or other technological steps independently of the United States. Some of these countries also have international trade interests that are threatened by their dependence, or that of their customers, on the United States. For example, sales of reactors and reprocessing or other services are often tied to fuel supply arrangements, and the possibility of intervention by a third party may affect the competitive advantage of different suppliers.

Among the major nuclear fuel consumers, France is already nearly independent of the United States and is generally oversupplied. West Germany is in a more difficult situation since current contracts are comparable to requirements from the early 1980s on, and nearly half of the contracted supply is from the United States. Alternative sources of supply include the U.S.S.R., expanded URENCO capacity and purchases from EURODIF. With a domestic centrifuge industry in need of orders, URENCO expansion appears most likely; indeed, there are industry reports that Germany will soon cancel eight enrichment contracts with the United States (55 percent of its total contracted U.S. supply) and expand URENCO accordingly.13

The third major consumer of interest is Japan, all of whose enrichment currently comes from the United States. Under present contracts, Japan appears to be unavoidably dependent on the United States until the late 1980s. Clearly, however, it would be in Japan's interest to diversify, obtaining some enrichment services from EURODIF or the U.S.S.R. until a domestic centrifuge capacity can be built up in the late 1980s. Other industrialized countries, such as Spain or Sweden, would be in a good position to end enrichment dependence on the United States, though the incentives do not appear to be as great as for the three major consumers already discussed.

The disincentives to such a shift away from U.S. supply should not be considered prohibitive. While the United States would find the loss of influence and income unattractive, it could hardly argue that changes in suppliers of low-enriched uranium fuel constituted an immediate proliferation danger and thus justified strong political opposition. Moreover, the economic penalties are small: slightly higher non-U.S. prices, and even cancellation penalties, would add at most a few percent to the cost of nuclear electricity. On balance, the trend toward substitution of non-U.S. supplies of enriched uranium for those from the United States will continue, unless there is some adjustment in U.S. policy (and probably some shift will occur even if there is).

In short, the position of the United States in the supply of uranium enrichment services is simply not powerful enough to affect strongly the decisions of the major industrialized nations on future nuclear technology. Indeed, these nations are in fact much more dependent on a handful of countries for their supplies of raw uranium than they are for enrichment services. To the extent that there now exists any substantial supplier leverage over major nuclear fuel consumers, it is gradually passing from the United States to the currently largest actual and potential sources of raw uranium, notably Canada and Australia.14 And even these two countries may have only transient influence, given the rapid expansion of output in Africa and intensive exploration elsewhere.

In the case of the developing and middle-income nations, neither the position of the United States in enrichment capacity nor that of Canada and Australia in raw uranium seems to offer significant direct leverage on technology choices. These countries have relatively small fuel requirements over the next decade, and have yet to make commitments for the major portion of projected capacity. Therefore they can pick and choose among available suppliers. For enrichment, EURODIF is the most likely new source of supply; for uranium, African suppliers with expanding output (South Africa, Namibia, Niger and Gabon) will appear attractive to countries wishing to avoid conditions likely to be imposed by the United States, Canada and Australia. Detailed calculations for a representative set of developing or middle-income countries, with a total nuclear capacity projected to grow to about 30 GWe in 1990, show that the requirements of these countries would fit easily within the EURODIF excess capacity window until about 1990 (until later if there is further slippage in European demands on EURODIF capacity).15 By 1990, additional enrichment capacity could be built. Raw uranium requirements are also easily met by African suppliers and can even be met until the mid-1980s by domestic and foreign uranium production controlled by France and in excess of that country's domestic needs. A considerable quantity of uranium will also be freed by Iran's abrogation of uranium supply contracts with South Africa.

If there is no convergence in the policies of the United States and other suppliers, the developing countries are likely to exploit remaining differences, perhaps to the detriment of nonproliferation goals. France, as the dominant partner in EURODIF and with its favorable access to uranium through its special relationships with several African suppliers, is the key supplier country involved. If there are major differences between French and American policies on nuclear fuel supply, it is not difficult to imagine a gradual drift toward the French sphere of nuclear influence.

These market changes are of profound importance to U.S. nonproliferation policy, because countries perceiving an advantage in independence of the United States will soon be able to achieve it. This perception undoubtedly will be affected by the way in which the United States uses fuel to achieve broader constraints on nuclear development abroad or on export policies of other suppliers. If the stakes are raised too high, or if large uncertainties about the security of supply from the United States continue, the balance of some nations' interests may be shifted away from the United States, to the detriment of both its nonproliferation influence and its commercial interests.


The preceding analysis reveals several deficiencies in the current U.S. nonproliferation legislation. As noted at the outset, the assumption that fuel-cycle leverage can be used to discourage commitments to plutonium or indigenous fuel-cycle facilities is not self-consistent. This problem would be less serious if the United States were to retain total fuel-cycle control, so that it could raise the costs of noncompliance with U.S. policy above the costs of complying. But U.S. market leverage is declining rapidly and so too is any influence based on this leverage. The continued U.S. use of fuel leverage only tightens near-term supply conditions and increases uncertainties, adding to the pressure to decide in favor of the very activities we are trying to restrain. In the larger perspective, the Act's new conditions on access to U.S. supply, and its apparent willingness to impose high costs on recipients-in terms of technology planning, energy security or sovereignty-have not only created new fuel-assurance fears, but raised questions about balance and stability in U.S. policy.

How might the United States respond to this situation? There are two potential approaches: maintain the strict policy of the Nonproliferation Act by increasing effective fuel-cycle leverage, or adapt U.S. policy to a changing world. We believe the first choice to be a dangerous one. The United States could try to make more vigorous use of its own remaining fuel-cycle leverage, for example by threatening total cutoff if there is inadequate response to U.S. proposals. But this would likely make things worse, threatening a number of important relationships and strategic interests, and increasing the stakes in a situation in which the United States ultimately cannot win.

The United States could also try to increase fuel cycle leverage by working in tandem with Canada and Australia. Although, as we have just noted, the uranium export policies of these countries will in fact have more enduring relevance to other nations' decisions, their use of fuel leverage involves the same questions of self-consistency discussed above. Canada has retreated from its unilaterally imposed conditions on exports to Europe for the duration of INFCE but will subsequently have to reach an accommodation which protects Canadian commercial interests as well as international security. Canadian export conditions thus do not represent a very promising channel for U.S. foreign policy. Australia poses an even more difficult problem. Present Australian policy on uranium development and exports appears to be premised on a satisfactory international nonproliferation agreement following INFCE. If such an agreement cannot be achieved, it is possible that Australia will not be able to sustain the fragile domestic political consensus which allows large-scale exports of uranium. The actual or potential loss of this supply would then accelerate commitments to plutonium. In short, uranium supply from Canada and Australia is susceptible to the same concerns we have raised about the efficacy of a U.S. nonproliferation policy based on fuel leverage, with additional complications.

Thus, to attempt to use even strengthened fuel leverage, alone or with others, is simply not a workable policy. But if the alternative is a policy of adaptation, what are its essential features?

They are, we believe, threefold. First, the conditions for U.S. supplies of nuclear technology and materials to foreign consumers must be made more predictable. Second and directly related, the United States must seek more effective, and less self-contradictory, sources of influence than that provided by fuel-cycle leverage. But these two essentials cannot be met without a third-U.S. policy must be responsive to the widely varying situations of different countries, and founded on realistic assessments of where the balance lies as between reasonable nuclear power needs and weapons proliferation dangers in each country or group of countries. This third condition-in essence a more selective policy on the part of the United States and, one hopes, others of like mind-raises basic questions about the distinctions now enshrined both in the Nonproliferation Treaty and in the 1978 U.S. Nonproliferation Act.

Under the NPT, the only distinction between nations is on the basis of possession of nuclear weapons. Non-nuclear-weapons states, without distinction, are assured access to nuclear power technologies and materials (subject to a no-weapons pledge and the acceptance of safeguards). The U.S. Act, in effect, preserves the weapons versus non-weapons distinction by a system of conditions applicable equally to all non-weapons states. And, it can be argued, such a system is consistent with the Treaty requirement of not discriminating among non-weapons states wishing to pursue nuclear power programs.

However, others argue that the Treaty need not be construed to mean that all non-weapons nations must in practice receive the same kinds of technology and materials, especially if the obtaining of these would create clearly different risks of weapons proliferation. Under this interpretation, all nations should be able to obtain what they reasonably need for nuclear power purposes, but particular kinds of supply could take into account both need and proliferation risk. This view might also seem both reasonable and in accord with the basic objectives of the Treaty. The resolution of these conflicting interpretations awaits full international consideration, especially in the upcoming NPT Review Conference.

In practice-and in large part because of technological developments and perceptions that have arisen since the Treaty was adopted in 1970-it is now clear that all non-weapons states do not have similar needs, though there is a sensitive question of assessing proliferation risks and who should be entitled to make such assessments. For a few non-weapons countries such as Japan, West Germany, and perhaps others, the pursuit of plutonium-related technologies is both a matter of national policy and a response to energy needs and assessments which it would be difficult to refute conclusively. For others, however, it is possible to argue persuasively that these sensitive technologies offer little or no nuclear power benefits.

Indeed, just such a selective distinction has been pursued in several contexts by the Ford and Carter Administrations since 1975. In bilateral discussions, the Administration has indicated willingness to consider use of plutonium in major industrialized non-weapons countries like West Germany or Japan. In addition, American initiative led to the formation and operation of the London suppliers' group, the initially secret group of representatives of nuclear exporting countries that began meeting in London in 1975. Now in the open and institutionalized, the London group has been effective in restraining the spread of sensitive technologies and materials beyond the few countries in which their use for nuclear power is at least plausible.

But the efforts of the Administration must be seen-and are seen abroad-against the background of the 1978 Act, which legislatively entrenches U.S. policy in a uniform set of restrictions. What negotiating flexibility the United States has is currently limited to the exemption procedures of the Act. Foreign nations cannot yet have a clear view of the extent to which U.S. policy can ultimately adapt to their needs; under worst-case planning, they will assume that the inflexibilities of the Act will prevail. This prospect tends to unify opposition to U.S. efforts, with Japan and West Germany becoming allied with each other, with other states in the European Community, and, potentially, with some of the developing countries.

At present, many of these issues are held in unstable equilibrium by the INFCE process. American officials have hoped that INFCE will result in an international consensus on economic and technical factors that would automatically lead to distinctions between nations on these grounds, and thus avoid the need to make distinctions on the basis of nonproliferation risks. For example, a collective judgment that recycle of plutonium in light-water reactors (thermal recycle) was without significant merit could lead to the restriction of plutonium use to the few advanced countries experimentally pursuing breeder reactors. While this approach is attractive, its success is problematic. And even if the executive succeeds in achieving such a general agreement, it cannot fit comfortably within the confines of the U.S. Act.

There are, of course, several other directions in which the world could move when INFCE ends. Unless the United States takes action soon, there is a significant chance that the outcome will be unsatisfactory to the United States and even threatening to the existing nonproliferation regime. While further unilateral actions cannot guarantee a better outcome, there are steps that could have a constructive influence on the way the world moves at the end of INFCE.

The critical needs are for more flexibility in dealing with different nations, and greater predictability once the terms of bilateral relations are agreed to. As discussed above, the present Act has difficulty in meeting these needs. Without changes in the Act, only a few options are plausible: the executive may commit itself to intervene through exemption procedures, or the NRC and the Secretary of Energy might commit themselves to blanket licenses or prior approvals, perhaps on a reactor- or country-specific basis. However, these actions are not explicitly provided for in the Act, and uncertainties about congressional approval, and the longevity of commitments by agents with limited tenure, raise questions about their effectiveness in any but short-term situations.

The attractive alternative to these limited executive measures is congressional action. We believe that the Act should be revised, and that particular attention should be paid to the role played by Agreements for Cooperation, as opposed to export license and subsequent approval procedures. As the Act is presently constructed, Agreements play a subordinate role: licenses and approvals generally require satisfaction of Title III's technology-oriented nonproliferation criteria even if some of these criteria are explicitly waived in the negotiated Agreement. It is this rigid policy focus upon individual licenses and approvals that creates most of the uncertainty and inflexibility.

The most obvious solution would be to restore the Agreement for Cooperation to its original role as the basic instrument of long-term bilateral nuclear relations. Title III of the Act could be revised so that a license or approval would be granted as long as there was no evidence that the prospective purchaser was in violation of its Agreement with the United States. This reversal of the burden of proof for export licenses would reduce uncertainties and renew incentives for nations to seek accommodation with the United States on nonproliferation objectives.

This revision would also provide a firmer and more realistic framework for responding to the diverse needs and proliferation risks in different countries. The basic criteria for Agreements laid down in the 1978 Act could remain, protecting essential nonproliferation objectives and avoiding (at least in part) the appearance of discrimination. However, the ability of the Administration to negotiate, and propose to Congress, exemptions from particular Agreement criteria for particular nations would remain and would have much greater value in negotiations because such exemptions would apply to all licenses unless a nation violated the terms of its Agreement. Thus the President could propose that sensitive facilities or materials could be used in particular non-nuclear-weapons states, under specified conditions. In addition, it would be possible to require more evidence of compliance for some states than for others.16

In sum, we believe that amendment of the 1978 Act is a prerequisite to a shift to a more realistic and effective U.S. policy. Only such amendment can give the U.S. negotiators the necessary flexibility for pursuing with other supplier and consumer nations the kind of multilateral agreements and arrangements that would truly establish a workable balance between nuclear power needs and the risks of weapons proliferation.


A more successful U.S. policy requires not only adaptive changes in approach but also a clearer recognition of the relationships between the ends of U.S. policy and the means at our disposal. The evident goal is to retard proliferation. In the near term this involves slowing commitments to sensitive technologies and materials. In the long term-if enough time can be gained-the problem is one of altering the structure of incentives and disincentives to weapons acquisition and constructing political, technological and institutional mechanisms which reduce the remaining proliferation risks. Full discussion of this longer term task is beyond our scope. However, our analysis does indicate that success will require a sensitivity to the growing importance of other nations in nuclear affairs.

The approach chosen in the 1978 Act is unilateral, as may have been appropriate to the initiation of a period of interactive policy shifts here and abroad. But the limits of unilateralism are now being strained. Dealing with proliferation is increasingly an international task requiring the cooperation of a number of key nations. U.S. policy must have the flexibility to accommodate the vital interests of these nations or it will fail. The measures we have suggested will not guarantee success, but only remove our own internal impediments to success, putting the United States in a more flexible-and believable-position to work out solutions bilaterally and multilaterally.

If the United States does not find ways to adapt, lowering the uncertainty of fuel supply, the evolution of nuclear fuel markets will soon provide opportunities for other countries to avoid involvement with us. In this event, the United States could only take comfort in having clean hands with respect to the further spread of enrichment facilities or plutonium, but it could not take credit for having led the world to a safer nonproliferation regime. Indeed, the uncertainties it has created about fuel supply could, eventually, make things worse rather than better. This is not to argue that the United States has thus far made things worse; past changes in American policy have called attention to important problems and set the stage for significant improvements in the nonproliferation regime. But without further changes in policy, the United States will not have the freedom to act on this stage.


3 Ibid., p. 785.

4 White House Briefing Room Transcript of April 7, 1977. Reaction to this statement was undoubtedly intensified by earlier Canadian actions in embargoing uranium contracted for by Japan and Western Europe.

5 In practice, the full-scope safeguards condition in the Act applies to non-signers of the Nonproliferation Treaty (notably Argentina, Brazil, India, Israel, Pakistan, South Africa, and Spain), since non-nuclear-weapon signatories to the NPT are already obligated to accept full-scope safeguards.

6 Note that this may involve a judgment as to the efficacy of the IAEA safeguards system in providing such warning to the United States of possible diversion. It is not obvious that the system provides such warning at present.

7 The imposition of the "subsequent arrangement" conditions can be delayed by action of the Secretary of State until March 1980 for particular nations or groups of nations agreeing to renegotiate Agreements as stipulated in the Act. This exclusion, which may be extended for additional 12-month periods by Executive Order, is a critical negotiating issue with EURATOM, which has its own safeguards system and prohibits barriers to free movement of nuclear materials within the Community. Additionally, the "full-scope" condition is imposed only for export license applications occurring after September 10, 1979 (or actual export after December 10, 1979), and the President may defer this requirement annually thereafter, under special conditions and subject to congressional review.

8 This assumption was based on a plausible technical logic, widely prevalent prior to 1977, that radioactive materials needed to be reduced in volume by reprocessing and encapsulated in glass prior to disposal. Subsequent studies suggest that waste management does not require, or necessarily benefit from, reprocessing of spent fuel. See, for example, the Report to the American Physical Society by the Study Group on Nuclear Fuel Cycles and Waste Management, July 1977, in Reviews of Modern Physics, Vol. 50, No. 1, Part II, January 1978.

9 URENCO is a consortium of the United Kingdom, the Netherlands and West Germany; it began furnishing enrichment services in 1978 from a centrifuge enrichment facility but is currently committed to a capacity only about one-tenth that of the United States. EURODIF'S gaseous diffusion plant in France is financed by France, Italy, Spain, Belgium and Iran, each of which is entitled to a share of the annual output (nearly half as large as current U.S. capacity). EURODIF has just begun operation and will reach full output in 1982.

10 This section is based on research sponsored by the Office of International Programs, U.S. Department of Energy (which has no responsibility for opinions expressed) and by the MIT Center for Energy Policy Research. See T. L. Neff and H. D. Jacoby, "Nuclear Fuel Assurance: Origins, Trends and Policy Issues," MIT Energy Laboratory Report No. MIT-EL 79-003, February 1979, and "Supply Assurance in the Nuclear Fuel Cycle," Annual Reviews of Energy, Vol. 4, 1979 (forthcoming). Enrichment requirements are based on the OECD/NEA-IAEA nuclear growth estimates of December 1977, Uranium Resources, Production and Demand, OECD, France. These growth estimates are now regarded as high, and adjustment to more recent unpublished estimates would only strengthen the conclusions discussed in the text.

11 Enrichment services are measured in separative work units; about 240,000 SWU are needed to provide the initial fuel for a 1,000 MWe light-water reactor and an annual reload of fuel requires about 110,000 SWU.

12 These calculations assume Soviet exports at currently contracted levels (a peak of 4.8 million SWU annually in 1982), URENCO growing to only 2 million SWU, and EURODIF coming on line on its present schedule, with a peak capacity of 10.8 million SWU by 1982. U.S. capacity-not fully utilized at present-is assumed to be increased through the currently planned improvement and upgrading of U.S. diffusion plants and the development of a U.S. centrifuge capacity of 2.2 million SWU by 1988, bringing total U.S. capacity to 29.5 million SWU by the late 1980s.

13 Nuclear Fuel, February 19, 1979, p. 9.

14 Although the United States still has substantial reserves of raw uranium, it does not play a major role in the world market. Foreign consumers turn to U.S. sources for spot needs, in preference to long-term contracts. This is perhaps due to uncertainty about the stability and acceptability of U.S. export policies.

15 The countries examined were Argentina, Brazil, Egypt, India, Iran, South Africa, South Korea, and Taiwan. Of the 30 GWe in 1990, about six GWe would be heavy-water reactors not requiring enrichment. These projections are ambitious; lower growth would provide even better opportunities to achieve independence of the United States.

16 The balance between Congress and the Administration under an amended Act would require careful consideration. The present Act specifies considerable review powers for Congress in regard to executive branch actions. While these safeguards may well be superfluous in the case of nonproliferation policy, where there is general bipartisan agreement, they do not represent a serious impediment to executive branch initiatives and could be preserved if desired. The more important problem is that discussed in the text: the severe prior limitations on the diplomatic opportunities and flexibility of U.S. negotiators imposed in the Act's present formulation.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • Thomas L. Neff was the senior staff member for the Ford-MITRE Nuclear Energy Policy Study published in 1976, and has served as a State Department adviser. He is now Manager of the International Energy Studies Program of the MIT Energy Laboratory. Henry D. Jacoby is Professor of Management at MIT, and Director of the Energy Laboratory's Center for Energy Policy Research.
  • More By Thomas L. Neff
  • More By Henry D. Jacoby