May 2002, pp. 41-45
In March, President Bush marked the six-month anniversary of September 11 by warning us once again of the peril posed by weapons of mass destruction. From the south lawn of the White House, he cautioned that “terrorist groups are hungry for these weapons, and would use them without a hint of conscience.” The result, he said, would be “blackmail, genocide, and chaos.”
He was right. Yet in one crucial respect the actions of his administration have been completely at odds with these strong words of warning. For the White House is now pushing a bill in Congress that would make it easier for terrorists and the nations that support them to obtain precisely such weapons of mass destruction. The bill would weaken the limits on sales of “dual-use” technology-civilian items that can also help in developing nuclear, chemical, and biological weapons and the missiles to deliver them.
The term of art here is export controls-restrictions on the power of American companies to sell things abroad that could do us harm. As it happens, while Bush was still on the campaign trail in 2000, he pledged to reduce these controls if elected to office. In those bygone days before September 11, the world (it seemed) had become a peaceful place, and we could safely let American companies make a little more money. But now the question is different: how to ensure that our own technology does not help the people who want to kill us.
Export control is a hoary topic, dating from the years immediately following World War II. In our efforts to rebuild Europe and fashion NATO as a barrier to Soviet expansionism, it became obvious that the West would be inflicting serious harm on itself if American goods and equipment were being sold to the Soviets or the Chinese. There had to be a fence around the United States and its allies, one inside which trade could flourish but that would be strong enough to keep useful technology-especially of the military sort-away from rivals outside. The fence was COCOM, the Coordinating Committee for Multilateral Export Controls, born in 1949.
COCOM was a phenomenal success. It did not stop everything from getting out, any more than the best police force can stop all crime. But by systematically helping to deprive the East of the West’s technology, it left Russia and its allies in the dust. After the end of the cold war, COCOM officials toured the former Warsaw Pact to measure the historical impact of controls. What they learned from former East-bloc officials was that equipment had sometimes been smuggled in, but spare parts and service had been impossible to obtain. This made it risky to build manufacturing operations around such equipment. Even today, more than a decade after the end of the cold war, the dilapidated industrial infrastructure of Russia and Eastern Europe offers lingering testimony not only to the abysmal failures of socialist planning but to the power of export controls.
More recently, documents found by UN inspectors in Iraq after the Gulf war showed how export controls had seriously hampered that country’s nuclear-design teams, forcing them to spend time and money trying to engineer essential items they could not import. Controls also stopped Iraq’s drive to make a medium-range missile-one that would have been invulnerable to U.S. Patriot missile defenses. Without such restrictions on his purchases, Saddam Hussein might have come into possession of a nuclear weapon, a medium-range missile, or both, by the onset of the Gulf war.
Finally, export controls had beneficial effects on nonhostile regimes as well, contributing to a general atmosphere of restraint and caution. Thus, when both Argentina and Brazil agreed to give up their nuclear-weapons potential in the 1980’s, a principal reason was the cost imposed on them by American export controls, which had prevented them from buying vital technologies like supercomputers. Both countries preferred in the end to be seen as part of the solution to arms proliferation rather than as part of the problem. In effect, export controls had bought time, slowing these nations’ weapons programs until the political winds shifted, and had created an incentive for them to qualify as reliable recipients of advanced technology.
With the end of the cold war, export controls entered a period of decline. The Soviet Union and the Warsaw Pact were gone. The new threat facing us-a number of countries building weapons of mass destruction-was more fluid, and there was less agreement on how to deal with it. In 1993, R. James Woolsey observed at his confirmation hearing for the directorship of the Central Intelligence Agency that “We have slain a large dragon, but we live now in a jungle filled with a bewildering variety of poisonous snakes.” How to contend with those snakes was up to the Clinton administration, which took office that same year.
Things did not go well. At his own confirmation hearing for the position of Deputy Secretary of Defense, William Perry, who would later become Secretary of Defense, declared that controlling dual-use technology was a “hopeless task.” Control, he said, “only interferes with a company’s ability to succeed internationally.” This put Perry, a former electronics executive, in direct conflict with the UN inspectors then in Iraq who were reporting that without strict maintenance of export controls, Saddam Hussein would soon reconstitute his mass-destruction war machine.
Whether Perry realized it or not, virtually every part of a nuclear weapon is made with dual-use equipment. Iraq, in fact, had already imported dual-use “isostatic” presses to shape nuclear-bomb parts, dual-use mass spectrometers to sample bomb fuel, and dual-use electron-beam welders to increase the range of its Scud missiles. During the Gulf war, one of these increased-range Iraqi Scuds had killed a group of U.S. soldiers in their barracks in Saudi Arabia, prompting a Pentagon official to remark that “when you talk about export controls, you’re not talking about politics, you’re talking about body bags.”
There was (and there remains) no hope of thwarting the Iraqi-or the Iranian-bomb and missile program without controlling such dual-use exports. Back in the 80’s and early 90’s, under the Reagan and (first) Bush administrations, the Defense Department had successfully battled attempts by the industry-friendly Commerce Department to dilute controls. Under Clinton and Perry, this institutional counterweight disappeared. As one disillusioned Pentagon staffer complained, “We now have four layers of bosses who don’t believe in export controls.” The bosses included Ashton Carter, Perry’s assistant secretary, who actually proposed giving Pakistan the special electronic locks that make nuclear warheads safer, as well as supplying nuclear fuel to India, and Carter’s own deputy, Mitchell Wallerstein, who had directed studies decrying export controls for the National Academy of Sciences.
With the Pentagon thus neutered, the Commerce Department was able to give exporters what they wanted. The two greatest beneficiaries were the American computer industry and the People’s Republic of China (PRC). Before 1996, China had been denied virtually all access to high-performance computers. These computers can simulate the conditions inside an exploding nuclear warhead or the forces acting on a missile from launch to impact; they enable a country to build not only better missiles but smaller, more powerful nuclear weapons without explosive testing. Between 1996 and the end of 1998, China succeeded in acquiring over 600 such high-performance computers from American companies, all with the approval of the Commerce Department.
It is a standard condition of U.S. export licenses that a U.S. official must be allowed to verify that a dual-use item has not been diverted from its declared civilian use. In the case of these computers, China has refused to comply: by January 2002, it had rebuffed some 700 U.S. requests for verification. What the Chinese are hiding is indubitably the fact that the next generation of Chinese nuclear warheads is being developed with American equipment.
China needed help with its missiles, too. Before 1996, an embarrassing series of launching failures had destroyed valuable satellites and put in doubt not only the reliability of China’s space rockets, but its intercontinental ballistic missiles, which are pointed at American cities and bear thermonuclear warheads. Never fear: to the rescue came two of America’s leading satellite manufacturers, Hughes Electronics and Loral Space and Communications.
One or both of these companies helped China improve its guidance systems and nose cones, and taught the Chinese how to diagnose launch failures. In order to orbit their own satellites cheaply on Chinese rockets, the companies also gave China technology that could, in the words of a later congressional investigating committee, increase “the reliability of all PRC ballistic missiles.” In doing all this, the two companies, according to the committee, “deliberately acted without the legally required licenses and violated U.S. export-control laws.” Since getting American help, China’s rockets have enjoyed 21 successful launches in a row.
Both Hughes and Loral had special ties to President Bill Clinton. Hughes was led at the time by C. Michael Armstrong, a strong Clinton supporter whom the President chose to be chairman of an advisory group with a key role in shaping U.S. export-control policy. From this perch, Armstrong (who has since become the chief executive officer of AT&T) helped convince Clinton to dilute controls on satellites. The other company, Loral Space and Communications, was led by Bernard Schwartz, the largest single donor to the Democrats’ 1996 campaign fund. The illegal aid extended to China by the companies run by these men marked one of the lowest of the low points struck in the Clinton administration.
Would things change under Bush? In the first eight months of the new administration, little happened on the export-control front for good or ill. But in September, America was suddenly at war. Soldiers had to be sent to Afghanistan, and to make that possible the United States had to bolster its position in South Asia.
India was the first diplomatic port of call. U.S. officials were only too happy to entertain India’s warm public offers of help, and to accept Indian declarations of common purpose. The display was calculated to exert irresistible pressure on the real U.S. target, Pakistan. This beleaguered Islamic country, India’s neighbor and arch rival, was indispensable to U.S. operations in Afghanistan, but its president, Pervez Musharraf, had to be persuaded to sign on. Inviting in the West would anger Islamic extremists; not inviting in the West risked an American alliance with India that would leave Pakistan isolated and linked to terrorism. With America playing the India card, Musharraf had to do the right thing.
How would the United States reward India and Pakistan for their cooperation? There was money. There was the friendship and gratitude of the world’s only military and economic superpower. And there was the possibility of doing something about export controls.
In 1998, both India and Pakistan had set off nuclear test explosions, and the Clinton administration (in one of its periodic fits of clarity) had expressed American disapproval by announcing sanctions: henceforth, export licenses would be denied to Indian and Pakistani companies known to be participating in their countries’ nuclear and missile efforts. After September 11, however, the sanctions were quietly dropped. A long list of Indian companies, plus a few in Pakistan, had a green light to purchase dual-use equipment from the United States.
Among the Indian companies so benefited was Hindustan Aeronautics, a major producer of nose cones, engines, and fuel tanks for India’s largest rockets and missiles. Another was Godrej & Boyce, which also produces rocket components like engines, motor casings, and heat shields. A third was the National Aerospace Laboratory, which performs rocket and missile research, does wind-tunnel and ground-vibration testing, and analyzed the first flight test of India’s nuclear-capable Prithvi missile. And a fourth was Walchandnagar Industries, producer of coolant pumps, end shields, and reaction vessels for India’s nuclear reactors; many of these reactors operate outside the purview of international inspection, which means that the plutonium they make is available for use in atomic bombs.
All of these firms are unquestionably manufacturing weapons of mass destruction, and all of them have been cleared for sensitive American exports – a paradoxical consequence, to say the least, of our determination to respond forcefully and globally to a terrorist attack on American soil.
Which brings us back to the bill the administration is now supporting in Congress. The bill, unlike the present export law which it would replace, makes no attempt to strike a balance between national security and freedom of trade. Essentially a one-sided list of provisions advocated by commercial interests, it is at loggerheads with the administration’s own efforts to protect the American population after September 11.
One such effort involves the U.S. Customs Service. To help keep terrorists from getting weapons of mass destruction, Customs is sending out agents to warn American companies about selling approximately 100 items on a “most-dangerous-goods” list. The list was drawn up under the “Shield America” program, which urges companies to ascertain the true identity of the buyers of these dual-use goods.
Incredible as it may sound, the bill before Congress would decontrol many of the same items that the Customs Service is warning about. The fact that the bill enjoys the backing of the White House (as opposed to another version circulating in the House that adds provisions favoring national security) reveals what one Customs official has aptly termed “a massive disconnect between different agencies of the government.” It also reveals a fundamental incoherence in the administration’s own foreign policy.
Among the dual-use items now on the controlled list are high-precision electronic switches (known as “krytrons”), which release a powerful surge of electric current in a short, precisely controlled interval. These switches have a beneficent and almost miraculous civilian use: they trigger the machines that destroy kidney stones inside the body without surgery. They also have another use: setting off the chain reaction in thermonuclear weapons.
In 1998, Iraq (which is allowed to import medical equipment despite the UN embargo), tried to provide itself with a supply of these switches by purchasing a half-dozen of the machines that destroy kidney stones. The seller was Germany’s electronic giant, Siemens, which graciously accepted the claim that Saddam Hussein was concerned about kidney stones in his subject population. Each Siemens machine required one switch. When Iraq tried to buy 120 additional switches as “spare parts,” warning flags went up, but not quickly enough to prevent the shipment of at least some of them.
Under the bill, these switches would probably be decontrolled. The reason is that they fit the bill’s definition of a “mass market” item-a new concept, invented by industry lobbyists, according to which it is impracticable to prevent any dual-use item that is freely available inside our borders from making its way outside. But the concept is flawed. For decades, these switches have been on the export-control list of the Nuclear Suppliers Group, a consortium of industrialized countries that includes virtually every switch manufacturer. The fact that Saddam Hussein had to engage in subterfuge to import the switches from Siemens shows that he could not get them any other way.
In addition to the switches, a number of other dangerous technologies could be decontrolled. One is the high-strength, “maraging” steel used to make solid rocket-motor cases and propellant tanks for missiles. This steel can also be used to form the critical moving parts of gas centrifuges that bring uranium up to nuclear-weapons grade. Other candidates for decontrol are corrosion-resistant valves, essential for regulating the gases inside plants that produce weapons-grade uranium. Both Iraq and Iran hope to build such plants and will need these valves in great number. They have been controlled for export since 1981.
There is no telling how many other items might be released. The glass and carbon fibers that go into rocket nozzles, the gyroscopes that help guide missiles, the isostatic presses that fashion rocket heat shields and nuclear-weapons parts, the vacuum furnaces that melt and cast uranium and plutonium, and the fermenters that grow biological agents-all of these products are sold in sufficient quantity in the United States to meet the bill’s mass-market definition.
Under the bill, an item that meets the mass-market definition must be decontrolled by the Secretary of Commerce. He has no discretion in the matter. The only way an item can be kept on the controlled list is if the President reverses the Commerce Secretary’s decision within 30 days, a task the bill forbids the President to delegate to others. Somehow, one cannot imagine the President and his staff plowing, month after month, through data on electronic switches, maraging steel, and corrosion-resistant valves.
The logic behind the mass-market concept is flawed both practically and morally. Congress is being asked to believe, without proof, that things that can be bought in quantity here are bound to leak across the border, making controls ineffective. As I have already shown, we have abundant historical evidence to the contrary.
Industry spokesmen also contend that since other countries sell at least some of the same things we do, our exporting companies are made to suffer an unfair disadvantage. It is true that the United States did not sell poison-gas plants to Libya or Iraq just because Germany did, or nuclear reactors to Pakistan because China did, or missiles to Iran because Russia did. Is that a policy to be ashamed of, let alone to think of reversing? Today, North Korea is selling-to Iran, Syria, Pakistan, and Libya-rocket technology far inferior to that which U.S. firms could supply. Should we rush to compete, and help bring those weapon programs up to speed?
The truth is that our companies, far from being at a disadvantage, have never had it so good. Since the end of the cold war, and thanks to the Clinton years, fewer and fewer items have been under control, and fewer and fewer applications for export licenses have had to be filed with the Commerce Department. Since 1989, both the number of applications and the value of goods controlled have decreased by approximately 90 percent. The value of the applications actually denied each year is only a few hundred million dollars, the equivalent of less than one-hundredth of 1 percent of the U.S. economy. In sum, export controls are only a shadow of what they were under COCOM. We are now being asked to get rid of the shadow.
The spread of weapons of mass destruction, the President has reminded us, is the foremost strategic threat facing the United States. These weapons are built mainly with dual-use equipment. If, as the White House seems to be urging, we reduce controls on the export of that equipment, we will not contain the threat; to the contrary, we will actively abet its growth.
Gary Milhollin directs the Wisconsin Project on Nuclear Arms Control in Washington, D.C. His article “Can Terrorists Get the Bomb?” appeared in our February issue.