Concern and frustration over the rapid spread of nuclear reactors, uranium enrichment facilities and reprocessing plants outside of the nuclear weapons club, to countries such as Brazil, South Korea, and the Union of South Africa, have recently led to suggestions that the United States place a ban on the export of conventional reactor technology, advanced reactor technology such as the breeder reactor, and fuel cycle technology until more acceptable safeguards institutions have been created. For example, in recent congressional testimony, David Lilienthal, the first chairman of the Atomic Energy Commission, suggested that such a unilateral U.S. strategy might be adopted.

There are several reasons why this approach to the proliferation problem is unlikely to be successful. First, the United States no longer possesses a monopoly over reactor technologies or fuel cycle processes. In fact, in the area of reprocessing and breeder reactor development the United States lags behind other industrialized countries. Countries such as France, Germany, Japan, Canada, and the United Kingdom have the capability to produce and export one or more key elements of the nuclear energy system and could quickly replace the United States as a supplier of nuclear facilities. Second, the nature of the nuclear industries in most of these countries makes it highly unlikely that they would go along with such an embargo; rather, they would be likely to exploit the opportunity for additional nuclear sales to the fullest. Third, such a policy would run the risk of wrecking the existing Nonproliferation Treaty (NPT) which guarantees countries nondiscriminatory access to nuclear technology for peaceful purposes. As a result, even if such a unilateral embargo policy could be effective in the short run, the long-term implications for international agreements in the safeguards area might be extremely undesirable.

These facts have led some, especially those in the domestic nuclear power establishment, to conclude that the current realities of the international nuclear energy industry not only rule out the possibility of an effective unilateral U.S. embargo policy but mandate the elimination of all trade restrictions on the domestic nuclear industry. This is not the conclusion that I come to in the discussion below. Rather, I seek to argue that recent developments in the international nuclear industry have considerably complicated the proliferation problem. Nonproliferation efforts must take account of these realities and recognize that trade in nuclear facilities can be effectively regulated only through cooperative agreements and international institutional arrangements, not by unilateral actions by the United States. The recent agreement to regulate such trade, signed by six key countries, is a step in this direction. But as I shall indicate below, this agreement still leaves much to be desired.


When a country considers whether or not to invest in nuclear steam supply systems for generating electricity, it must consider the economics of an entire nuclear energy system of which the nuclear reactor itself is only a part. For the light water reactors that the vast majority of countries are building, the generating facility itself is useless without a supply of uranium and uranium enrichment services which can turn the natural uranium into usable light water reactor fuel. The potential buyer must be assured that it can either buy enriched uranium fuel directly or contract separately for natural uranium, uranium conversion services, uranium enrichment services and fuel fabrication services-the "front end" of the nuclear fuel cycle. The availability of enrichment services or enriched fuel is the most important concern since it involves both advanced technology, which only a few countries now possess, and because the present cost of enrichment facilities of optimum economic scale is quite substantial.

The front end of the fuel cycle and the reactor itself are only part of the nuclear energy system. It is generally thought that the spent reactor fuel must be reprocessed for several reasons. First, the spent fuel contains uranium which can be salvaged, enriched once again and recycled as fuel. Second, in the process of nuclear fission, part of the heavy uranium isotopes have been transformed into plutonium, which has immediate value as a fuel mixed with enriched uranium in light water reactors, or can be stored for future use as a fuel in breeder reactors. Third, it is argued that the fuel should be reprocessed so that the radioactive wastes can be put into a convenient form for permanent storage.

At least the first two arguments depend on the costs of reprocessing spent nuclear fuel. There is growing evidence that such costs may be so high as to make reprocessing uneconomic. However, it is this "back end" of the fuel cycle that has caused the most safeguards concern since the plutonium produced in reprocessing facilities could be diverted and fabricated into nuclear weapons.

Therefore, the economic decision of a country that wants to use nuclear power strictly for peaceful purposes involves the consideration of a dynamic energy system of which the light water reactor itself is only a part. This economic decision ought to be made with consideration given to the costs and availability of front-end and back-end fuel cycle services to use along with conventional light water reactor technologies, and with an eye toward the future introduction of breeder reactors in which, if the costs of reprocessing and breeder construction are sufficiently low, the plutonium produced by conventional reactors can be effectively utilized.

To the extent that there has been a coherent U.S. nuclear export policy, it appears to be as follows: U.S. companies can export light water reactor systems, individual components, or technological know-how under International Atomic Energy Agency (IAEA) safeguards-but not enrichment or reprocessing facilities and technologies. (However, the United States is ready to sell enrichment and reprocessing services to foreign countries that wish to contract for them.) Since the Clinch River Demonstration Breeder Reactor is not likely to be operating before the early or mid-1980s, there is no policy as yet on future breeder reactor exports. The current export policy results in part from a desire to limit the spread of those fuel cycle facilities that could be used to produce either highly enriched uranium or plutonium which could then be fashioned into weapons.

For countries anticipating a major commitment to nuclear energy, the U.S. export policy is not particularly attractive, due to a number of important uncertainties. The availability and terms of U.S. enrichment capacity after the early 1980s remain unknown. The temporary moratorium on new enrichment contracts that was in effect in early 1975 created additional uncertainties about the dependability of the United States as a long-term supplier of enrichment services.

The situation in the back end of the fuel cycle is even more uncertain. At the present time there is no commercial reprocessing capacity in the United States, and there is not likely to be enough capacity even to serve U.S. needs until the early 1980s. More important are the general uncertainties of "environmental politics" surrounding the plutonium cycle in the United States. A decision to allow plutonium recycling has not yet been made, although one is due sometime next year. Even if the domestic plutonium industry passes through all of the environmental and safeguards hurdles, foreign governments remain concerned that the import and export of nuclear wastes to and from countries with reprocessing facilities will encounter serious internal political opposition. The Japanese have already begun to encounter such opposition in the United Kingdom, where they plan to send fuel to be reprocessed, though the United Kingdom has had by far the least opposition to nuclear energy of all the industrialized countries.

Any country which makes a large commitment to nuclear energy via a strategy of purchasing nuclear generating facilities or technology, but not fuel cycle facilities, finds itself in the position of being dependent on the internal political situation of the country or countries from which it contracts to purchase fuel cycle services. While political uncertainties in the United States are probably the greatest, they are substantial in several other countries that could supply such services. The nature of the fuel cycle industry is such that a country would find it extremely difficult to switch to other suppliers quickly if either enrichment or reprocessing services were suddenly curtailed or cut off. For example, our Energy Research and Development Administration (ERDA) and organizations in other countries currently selling enrichment services require long-term contracts to be signed well ahead of initial service. As a result, a country with substantial fuel cycle requirements would find it desirable either to contract for fuel cycle services with an international organization sheltered from the domestic political factors affecting export agreements, or to build the facilities themselves, both to serve its own reactors and to export such services to smaller countries wishing to diversify their sources of supply. By buying enrichment and reprocessing facilities or the technology with which to construct them, a country is of course also buying the capability to make nuclear weapons.


As long as the United States was the sole supplier of reactors, enrichment services and reprocessing, it could dictate the arrangements under which other countries, especially non-nuclear countries, could develop their nuclear energy systems. The purchase of a U.S. reactor carried along with it agreements as to safeguards procedures and the disposition of spent nuclear fuel. Normally a country would have to submit to inspections by the International Atomic Energy Agency after the NPT went into effect. Importing countries may not have liked the fact that they could not purchase key fuel cycle facilities or fuel cycle technology from the United States, but if they wanted nuclear reactors rather than alternative technologies, such as oil or coal, to generate electricity there was little that they could do but go along with the policy as long as alternative suppliers did not exist.

Today this monopoly position no longer exists. Several countries other than the United States are in a position to export enrichment technology, light water reactors indistinguishable from those supplied by U.S. manufacturers, and reprocessing technology and facilities. In most of these countries the economics of the domestic nuclear energy industries makes it almost imperative that they secure foreign sales in order to augment the domestic markets for nuclear power.

Let us consider the conventional light water reactor market first. U.S. exports of nuclear steam supply systems (NSSS), NSSS components and the associated technology have been accounted for primarily by Westinghouse and General Electric. During the 1960s and 1970s these companies sold reactor systems both directly and in cooperation with firms within particular countries. They concluded licensing and technical exchange agreements with companies in Belgium, France, Germany, Sweden, Switzerland, Spain, Italy and Japan. In addition, Westinghouse owns portions of companies which sell NSSSs in France, Belgium and Sweden. Between 1968 and 1971 the United States consistently provided over 90 percent of the world reactor export market. In most industrialized countries the idea was to use U.S. manufacturers as primary suppliers for initial reactors, but over the long term to transfer the technology and the production capability to domestic companies. Each of the industrialized countries felt that its natural interests dictated that it have its own independent industry for manufacturing the important components of the system and for putting the system in working order. In the case of Germany and Sweden, the major reactor vendors are completely independent of U.S. companies. In addition, Canada developed its own heavy water, natural-uranium-fueled reactor independent of U.S. firms. (The United Kingdom, while independent of U.S. technology, is no longer an important agent in the conventional reactor export market.)

By the end of 1975 there were four foreign firms active in the export market and in heavy competition with Westinghouse and General Electric. In France, Framatome produces pressurized water reactors (PWR) under license from Westinghouse. Framatome is a fully integrated company which can itself produce all major PWR components. It is partly owned by Westinghouse, but the French government has recently bought out most of Westinghouse's share in the company. While Framatome and its primary customer, Electricité de France, would like to continue the relationship with Westinghouse, there is no doubt that Framatome would go it alone if the U.S. government attempted to constrain Westinghouse's relations with it. In fact, there are groups within France, especially in the French Atomic Energy Commission, who would welcome an excuse to terminate the Westinghouse license agreement (which expires in 1982), both because the licensing arrangement with the U.S. firm restricts Framatome's freedom to export to certain countries without U.S. government approval, and to make the industry entirely French and more dependent on the existing French research and development establishment. Framatome currently has the capability to produce seven or eight NSSSs per year. The domestic market (of which Framatome has a monopoly) will probably absorb five per year, so that Framatome is trying to sell two or three NSSSs per year to foreign countries. Framatome has already negotiated sales of two reactors in Belgium and two in Iran and remains an active candidate for one or more reactors in South Africa. It has also been reported negotiating for sales in Egypt, Iraq, Kuwait and other countries.

In Germany, two major electrical equipment manufacturers joined together to form Kraftwerk Union (KWU), in 1967. KWU offers turnkey contracts to build a variety of nuclear steam supply systems, the primary one being a pressurized water reactor of German design independent of any U.S. company licenses. KWU, unlike Framatome, is primarily an engineering company and does not manufacture major reactor components itself. In making sales, its procedure is to sign a contract to deliver a complete nuclear generating station and then to subcontract to German and foreign firms for the individual components of the system. KWU is the major supplier to the German market and has already negotiated exports to Argentina, Brazil, the Netherlands, Spain and Iran, and is conducting negotiations for orders by Kuwait and Egypt. In 1975 KWU received eleven orders, more than the four U.S. reactor vendors received in total from both the domestic market and abroad. The company currently has the capacity to deliver at least four systems per year which can easily be expanded to six or seven per year with the domestic market absorbing perhaps five. This leaves a short-run export possibility of two reactor systems per year.

Framatome and KWU are the major competitors of the two major U.S. export companies in the light water reactor market at the present time. Together they could immediately provide exports of four or five reactors per year, a capacity which could easily be expanded over time if the demand warranted. Both companies can also provide a variety of fuel fabrication services, once presented with enriched uranium.

Several other existing or potential competitors are active in the international market. The Swedish company ASEA-Atom, which is similar in structure to KWU, sells boiling water reactors of Swedish design independently of U.S. firms. ASEA-Atom has secured only two foreign orders, both from Finland, but has been active in the export market. It was one of the bidders for the South African reactor orders. The company also provides a full set of fuel fabrication services. Due to reductions in domestic Swedish requirements, ASEA-Atom is likely to continue to seek foreign orders. Brown, Boveri and Company, a Swiss-German firm, has built or is building nuclear generating stations of both the pressurized and boiling water variety in Switzerland and Germany and was also a bidder on the South African orders. Brown, Boveri has generally operated in conjunction with a U.S. company and as a result should not as yet be considered an independent company without cooperation from General Electric or Westinghouse.

Three Japanese companies are potential entrants into the world market, but have not as yet consummated any major foreign reactor sales. Mitsubishi (Westinghouse license) sells pressurized water reactors in Japan and has been negotiating with the Soviet Union for the major components for ten PWRs. In addition, Hitachi (GE license) and Toshiba (GE license) are both selling boiling water reactors in Japan. The announcement of Mitsubishi's negotiations with the Soviet Union, combined with the slowdown in the Japanese domestic nuclear industry and the strong industrial base of these companies, indicates a distinct possibility that one or more Japanese companies will soon become an important international producer of NSSSs along with Framatome, KWU, General Electric and Westinghouse. As with Framatome, the U.S. licenses restrict somewhat the ability of these firms to export to communist countries without the approval of the U.S. government.

Finally, there are the Canadians, independently selling their heavy water reactor using natural uranium as a fuel. The CANDU reactor is desirable since it does not require enrichment services for the fuel, but it does require expensive heavy water as a coolant and moderator. The Canadians have been very active in the international reactor market, and their systems appear to be especially attractive to countries which seek independence from foreign enrichment technology and possess their own supplies of natural uranium. CANDU reactors have been sold in a number of countries including India, Argentina, Pakistan and Korea, with total sales of at least eleven overseas reactors. Canada has also signed a research and development agreement with Italy. The Canadians have been actively pursuing additional foreign sales, although for the near future Atomic Energy of Canada, Ltd. appears to have all of the orders that it can handle.

If Westinghouse and General Electric were forced to withdraw from the reactor market, Framatome in France, KWU in Germany and ASEA-Atom in Sweden could quickly replace them as alternative suppliers of light water reactors. In addition, the three reactor vendors in Japan must be considered as serious potential suppliers given their experience and production capacity. Finally, the Canadians, with some expansion in their production capacity, would be capable of supplying those countries willing to go with heavy water reactors rather than light water reactors. An attempt by the U.S. government to embargo the export of reactor components and technology would simply lead importing countries to turn to alternative suppliers who currently find themselves in an excess capacity situation. The United States might try to restrict exports of U.S. licensed components on national security grounds, but outside of exports to communist countries this does not appear to be a realistic possibility.


Let us turn now to enrichment. Currently the United States possesses virtually all of the non-communist world's commercial enrichment capacity in the form of three gaseous diffusion plants owned by ERDA and operated by U.S. corporations. ERDA has extensive contracts with foreign governments to provide enrichment services on a toll basis. Its capacity (to serve both domestic and foreign needs) is a little over 17 million separative work units (SWU) per year which is being increased to 27.7 million SWU per year by 1985. In addition, the Soviet Union has been selling enrichment services to non-communist countries for several years. Soviet deliveries are forecast at 3 million SWU per year by 1982. Nonetheless, the United States' near-monopoly position in the enrichment area is quickly disappearing.

At Tricastin in southern France, an international consortium composed of France, Belgium, Spain, Italy and Iran and called EURODIF, is building a large gaseous diffusion enrichment plant using French technology. The plant will eventually have 10.7 SWU of capacity, will begin operating in 1978 and reach full capacity in 1982. The same consortium is considering building another enrichment plant either in France or elsewhere, perhaps in Canada.

A second international consortium composed of Germany, the Netherlands and the United Kingdom, and known as URENCO, is building enrichment facilities based on a gas centrifuge technology. Pilot plants are currently operating in England and Holland and are being expanded. URENCO already has contracts for 2 million SWU per year in 1982 and is capable of expanding enrichment capacity to 10 million SWU per year if demand warrants it.

South Africa has built a pilot enrichment plant using a different process and is planning a commercial plant. An enrichment plant has been contracted for by Brazil to be built using German technology. Belgium is considering the construction of a large enrichment plant in Zaire using hydroelectric capabilities, and Japan and Australia have given consideration to enrichment plants.

In summary, the United States no longer has a monopoly in the enrichment technology area. Especially for reactors coming on line in the late 1980s, countries will have the opportunity to contract for enrichment services from a number of suppliers. They will also be able to construct their own plants if either their enrichment demand warrants it or if they have access to cheap hydroelectric power or abundant supplies of uranium which might make the domestic production and possible export enrichment services profitable. Such plants could theoretically be used to produce highly enriched uranium which could be fabricated into explosives.

The technology for reprocessing spent uranium fuel, the part of the fuel cycle that has attracted the most concern because it involves the separation of fissionable plutonium from spent reactor fuel, is also not a secret known only to the United States. The largest commercial plants in the world for reprocessing uranium metal fuel are at La Hague and Marcoule in France and at Windscale in the United Kingdom. A consortium of German companies is participating in the construction of uranium oxide reprocessing plants at La Hague and Windscale through the multinational reprocessing company, United Reprocessors, formed in 1971. In addition, an experimental reprocessing plant is operating in Karlsruhe, Germany, and a much larger reprocessing facility is planned by a German industrial-utility group in cooperation with United Reprocessors for the early 1980s. A small reprocessing plant is now operating in the Tokai area of Japan, and small reprocessing plants were also operated in the past in Belgium and Italy. While operating commercial reprocessing capacity in the world is well below current requirements, the technological capability to reprocess spent uranium fuel is known or being developed in at least ten countries which will be in a position either to build plants or to transfer technological capabilities to other countries in the 1980s. France recently sold a reprocessing plant to Pakistan, with IAEA safeguards approval already granted. A similar sale to South Korea was consummated but later cancelled by the Koreans under U.S. pressure.

It is perhaps in the area of fast breeder reactors that the notion that the United States can affect the international market through a unilateral embargo policy is the most questionable. Five other countries either have operating now, or will have operating around 1980, demonstration fast breeder reactors. An optimistic guess would be that the U.S. demonstration plant at Clinch River in Tennessee will not be operating until 1983 or 1984. The 250-megawatt-electric French Phenix fast breeder reactor has been operating with excellent results since 1973. A 250-MWe breeder built in the United Kingdom went critical in 1974, but as of December 1975 had not reached full power due to problems in the heat exchange system. Germany and Japan are each building 300-MWe demonstration plants which are scheduled to be operating around 1980. In addition, a 350-MWe breeder has been operating in the Soviet Union since 1972.

These countries are moving quickly to the construction of commercial-sized breeder reactors. Construction will begin late this year or early next year on a 1,300-MWe breeder known as the Super-Phenix in France as a joint venture between French, German and Italian utilities. The United Kingdom is moving forward with the construction of a 1,300-MWe commercial fast breeder reactor on its own. Finally, a 600-MWe fast breeder is nearing completion in the Soviet Union and is scheduled to go critical in 1977. France and the United Kingdom have both expressed interest in exporting breeder reactors. The French even have a 450-MWe export version of the Phenix which they have offered for sale. The British, whose early lead in conventional reactor technology has ended in dismal failure with virtually no export potential, are interested in reversing this trend through active sales of their breeder reactors.

When questioned about their response to a possible U.S. embargo on the export of breeder reactors and associated technology, the French and British participants in this area ask in turn: How can the United States embargo the export of something that it doesn't even have?

In sum, the international nuclear energy industry in 1976 is far different from the industry that existed even five years ago. In terms of operating reactors and enrichment facilities, the United States and U.S. corporations are in a commanding lead. But such figures reflect the past and not the present or future. A country wishing to purchase nuclear reactors can go to France, Germany or Sweden and obtain systems which are for all intents and purposes identical to those which could be obtained in the United States. One or more Japanese companies could and probably will enter this market in the near future. An importing country could also go to Canada and obtain a reactor based on a different technology. For enrichment technology or services, the country could go to France, Germany, the United Kingdom, South Africa and perhaps even Belgium or Japan. If a country wanted reprocessing capacity very badly, any of these countries and probably others could provide it. In the fast breeder reactor area, France, Germany, the United Kingdom, Japan and the Soviet Union will almost certainly be in a position to sell demonstrated technologies several years before the United States.


As a result of these developments in the international nuclear energy industry, the projected effectiveness of a U.S. embargo policy cannot be based on whether alternative suppliers could provide the requisite technologies, because they can, but rather on whether they would provide them in the face of a U.S. embargo policy aimed at forcing the international community to create better safeguards procedures and institutions. There is a very high probability that the suppliers in other industrialized countries would quickly and eagerly fill any void that an American embargo would create rather than go along with it themselves. There are many reasons for this conclusion:

First, each of the major industrialized countries has promoted the development of a relatively self-contained nuclear energy industry especially in the area of conventional reactors. By and large the domestic markets are not large enough to allow the reactor vendors to achieve scale economies in production. Most countries planned on having at least a small export market to allow them to operate their plants at minimum efficient scale and to help to spread some of the research and development costs. France's decision to abandon her own gas-graphite technology in favor of U.S. light water reactors was at least in part dictated by export market considerations. At the present time, due to slowdowns in domestic markets, there is substantial excess capacity in the world for producing conventional reactor systems. Competition for reactor sales is keen and growing. Seven different firms submitted bids to South Africa for light water reactors, and U.S. and European firms have recently found themselves competing for orders in Belgium, Spain, Brazil, Iran, Egypt and South Korea, in addition to South Africa.

Reactor exports not only help to maintain the viability of the domestic manufacturers, but can mean billions of dollars in foreign exchange, not only for the NSSSs themselves, but also for the turbines, generators and extensive technical services which are likely to go along with them. The general economic incentives to export these advanced technologies together with the current overcapacity in the market are not conducive to any kind of cartel agreement to restrict supplies of conventional reactors among the five or six major countries involved. For evidence of the desire of countries like France, Germany and Canada to gain export sales, one only has to look at the recent extensive efforts of high government officials to convince foreign countries to buy their own country's nuclear products. There are also strong feelings of national prestige associated with the export of such an advanced technology.

Second, the concerted efforts by the individual reactor vendors to export reactors, combined with cooperation and salesmanship by government officials and a very competitive marketplace, lead naturally to questions of sales of enrichment and reprocessing technology. A country like Brazil is in an excellent position to exploit the market to the fullest. Brazil is a country which might represent at least 20 reactor sales over the next 15 to 20 years. Three or four companies from different countries sought the business. The costs (to the extent they can be estimated) and the quality of the different bids were essentially the same. In addition, the Brazilians contended that the size of the Brazilian nuclear sector and the availability of extensive but remote hydroelectric sites made both enrichment and reprocessing desirable both in terms of straight economics and security of supply. Although such economic arguments are open to some debate, they are plausible and cannot be rejected out of hand. Given the intense competition, Brazil asked for a whole package including a complete fuel cycle, something U.S. firms could not offer. The German government, interested in obtaining the several billion dollars in exports that the order entailed as well as an opportunity to develop their own enrichment technology further, and faced with strong competition from France, was quite willing to agree to such a package over U.S. objections.

France's acceptance of Iran as a participant in EURODIF and agreements to share nuclear research and development, as well as the sale of a pilot reprocessing plant to South Korea, now canceled by the Koreans after strong U.S. pressure, are movements in the same direction. In addition, countries like the United Kingdom which have no significant role in the conventional reactor market are well advanced in reprocessing and enrichment technology, which at the present time represents their only chance to get a share of the nuclear export market. As pressure increases within the industrialized countries to curtail the import and re-export of the radioactive waste of other countries, the sale of technology and facilities will seem more and more attractive relative to the sale of services only.

Third, any nuclear embargo proposal set forth by the United States would be viewed with a jaundiced eye by most of the countries involved. The domestic market for nuclear power in the United States is just about as large as that in France, Germany, Japan, Spain and Italy combined. If U.S. companies cannot sell abroad for some period of time, they still have the large U.S. domestic market. They have already achieved substantial exports around the world, including exports to those countries which are now themselves in the export market. The French already view the controversy over the Concorde as an attempt to destroy a market that American industry cannot compete in, rather than as a legitimate environmental controversy. Will they look any differently at a nuclear export embargo? More importantly, would the United States be willing to make an embargo agreement more credible by encouraging the entry of French, German, Japanese and Canadian reactor vendors and fuel cycle companies into the U.S. market? I doubt it.

Given the large economic incentives of the countries involved to export nuclear facilities and technologies and the general perception that the burden of a multilateral embargo would fall unfairly on the industries outside of the United States, it is highly doubtful that an effective multilateral embargo can be achieved. On the contrary, Germany's sale of fuel cycle facilities to Brazil and France's sale of a pilot enrichment plant to South Korea, over the vigorous opposition of the U.S. government, is an example of how difficult it would be to get the other countries to go along with an embargo.

In opposing an export embargo, the countries need not appeal to the vulgarities of economics of scale, the profits of domestic manufacturers or even the venality of the United States. They will wrap themselves in the Nonproliferation Treaty which the United States has encouraged all countries to sign and which forms the only visible basis of U. S. nonproliferation policy. Article IV, paragraph I states: "Nothing in this treaty shall be interpreted as affecting the inalienable right of all the parties to the Treaty to develop research, production and use of nuclear energy for peaceful purposes without discrimination and in conformity with Articles I and II of this Treaty." Articles I and II deal with prohibitions against the transfer or construction of nuclear weapons. Since the sales of nuclear facilities are normally accompanied by either bilateral or multilateral safeguards agreements precluding (at least in theory) the production of nuclear weapons, other exporting countries will argue that an embargo would violate a key part of the Nonproliferation Treaty and as a result could cripple existing nonproliferation agreements.


The points elaborated so far should not be viewed as a brief for all-out free trade in nuclear facilities. Those expressing concern over the possibilities of additional nuclear proliferation arising from the current trade situation are quite correct. Continued free trade in nuclear facilities without adequate safeguards is frightening. The answer, however, does not lie in a unilateral abandonment of the international market by the United States, but in more vigorous efforts to create more effective international safeguards institutions. The current status of the international nuclear energy industry will, however, make such efforts far more difficult than in the days when the United States had a monopoly.

To say that the United States is no longer a monopolist in the nuclear energy area is not to say that it is powerless. It is and will remain a dominant force in the international nuclear industry as long as it continues to participate constructively in it. Effective policy measures to decrease the probability of the proliferation of nuclear weapons require that the United States recognize the realities of the international nuclear energy industry as it exists today and as it is evolving. The United States cannot attempt to make policy based on misperceptions of the interests, capabilities and motivations of countries like Germany, France, Japan, the Soviet Union and Canada as well as the legitimate interests of other countries in nuclear energy for peaceful purposes.

The common nuclear export policy announced last February between the United States and six other nuclear exporting countries is a step in the right direction. However, the current arrangement has some serious drawbacks and itself demonstrates how difficult nuclear export control has become. The seven countries agreed to sell nuclear energy facilities only under IAEA safeguards and after the importing countries agree not to use the facilities to produce nuclear explosives under any circumstances or to re-export reprocessing facilities or reprocessed plutonium. Other nuclear facilities could be re-exported by third countries if accompanied by the same international safeguards.

The United States was apparently unable to convince the other countries to agree to a complete embargo of enrichment and reprocessing facilities to third countries, the export of which had really been the main concern. Thus, the agreement does little more than extend the safeguards conditions of the NPT, with respect to only these seven major supplier countries-one of which, France, is not a party to the NPT (Japanese ratification was completed in late May, after the agreement)-and to those recipient countries which have not signed the NPT but wish to import nuclear facilities from one of the major suppliers. As a result, while the agreement demonstrates some international concern over the effects of accelerating nuclear trade, and the ability to come to some agreement, it does not really buy us very much.

The safeguards provisions continue to hinge on the detection capabilities of the IAEA. Even the current adequacy of these detection capabilities is far from universally accepted, and the technical problems of inspecting enrichment and reprocessing facilities are novel and formidable. In the face of these difficulties and of the "scaling up" problems the IAEA will face as it extends inspections to several hundred additional facilities over the next decade, its ability to fulfill the inspection tasks that have been given to it is highly doubtful. A detailed and realistic assessment of the actual rather than theoretical capabilities of the IAEA is certainly desperately needed at this point.

Even if the IAEA can do an adequate job of inspection, it has no practical enforcement or sanctions powers. Detection of a diversion might lead to public condemnation and a cutoff of nuclear supplies but little else. India produced an explosive device from material produced in a Canadian-built reactor which had bilateral safeguards assurances attached to it. The world just watched, voiced some strong objections, and added another member to the nuclear "explosives" club. Without some method of enforcement or some promise of serious sanctions, one's faith in the deterrent power of existing international safeguards institutions cannot be too strong.

The realities of the present structure of the international nuclear energy industry, the difficulties of inspection and the lack of sanctions do not lead me to an optimistic conclusion about the future of nonproliferation policy. The problem cannot be met through unilateral U.S. embargo activities. The United States can use its power and influence with particular countries and international organizations such as the International Monetary Fund and the World Bank to see that importing countries at least sign safeguards agreements and submit to inspections. Suggestions for the development of international fuel cycle facilities might be pursued with more vigor and perhaps even given substantial financial aid by the United States, which could also play a key role if more concrete enforcement or sanctions mechanisms are created. All this would add up to an important but limited U.S. role in developing more effective regulation of international nuclear trade by using the powers that it really has as a key actor in this arena, rather than seeking to invoke an imagined monopoly that is quickly disappearing.

But development of more effective regulation of international nuclear trade is not a "solution" to the proliferation problem. There is no easy solution at hand. As time marches on and the international nuclear industry develops further the problem will likely become even more difficult. In trying to find a true solution let us at least be aware of and take into account the realities of the international nuclear energy market from the viewpoint of both buyers and sellers, the problems that they entail, and the very difficult task at hand.

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
  • Paul L. Joskow is Associate Professor of Economics at the Massachusetts Institute of Technology. This article was based on research funded by the Ford Foundation.
  • More By Paul L. Joskow