Testimony: Reauthorization of the Export Administration Act

Testimony of Gary Milhollin

Professor Emeritus, University of Wisconsin Law School and
Director, Wisconsin Project on Nuclear Arms Control

Before the Senate Committee on Banking, Housing and Urban Affairs,
Subcommittee on International Trade and Finance

April 14, 1999

I am pleased to appear before this distinguished subcommittee to testify on the reauthorization of the Export Administration Act. I will begin by making some general remarks about the situation in which we now find ourselves with respect to export controls. Then, I will make specific recommendations on issues that now confront Congress. I would also like to submit two items for the record. The first is a study that my office has just completed on U.S. exports to China during the past ten years. The second is a study entitled “25 Myths about Export Control” that my office prepared a few years ago but which is still relevant to the issues we face today.

Export control is not a jobs issue

The most important single point for the Congress to appreciate is that export control is not a jobs issue. Export controls do not have a significant–or measurable–effect on employment. Of the total American economy, less than two tenths of one percent ($10.7 billion) even went through Commerce Department licensing in 1994, the last year for which I have been able to find licensing data. And more than 95% of licensing applications were approved. Only $141 million in applications were denied in 1994–which is less than one hundredth of one percent of the U.S. economy and roughly equal to six percent of the cost of one B-2 bomber. The figures today are roughly the same. Reducing export controls will not stimulate the U.S. economy; it will only stimulate the proliferation of weapons of mass destruction.

It is also important to realize that export controls are only a shadow of what they were during the cold war. Since 1989, applications to the Commerce Department have dropped by roughly 90%. Cases have fallen from nearly 100,000 in 1989 to 8,705 in 1996 and 11,472 in 1997. The reason is simple: fewer items are controlled, so fewer applications are required. Nor does export licensing take much time. The Commerce Department is meeting its licensing deadlines for 97% of its applications.

After cutting export controls to the bone to reflect the end of the cold war, we now need to strengthen controls to combat proliferation, the main threat of the post-cold war era.

Export control is a national security issue–now more than ever. The world for which Congress must legislate today bears little resemblance to the cold war world for which the existing law was written. The spread of weapons of mass destruction, rather than competition with the Soviet Union, is now the foremost strategic threat to the United States. Because mass destruction weapons are built mainly with dual-use equipment, the control of dual-use exports is of vital military and strategic importance. Rather than being viewed as commercial transactions with a military aspect, as they were during the cold war, dual-use exports must now be regarded as deeply affecting U.S. national security. It is illogical to say that 1.) the cold war is over and therefore proliferation is the main international threat, and 2.) that export controls, which are essential to contain that threat, should be reduced.

In light of the situation we face today, Congress should reevaluate the ability of the executive branch to respond to the proliferation threat. In particular, Congress should consider transferring dual-use licensing from the Commerce Department, which is concerned primarily with trade, to an agency, such as the State Department, that is primarily concerned with protecting U.S. national security. Or, if such a transfer is considered too big a step, Congress should strengthen the role of the national security agencies in the existing interagency process. In addition, Congress should find a way to increase public accountability for export licensing decisions, and a way to provide effective Congressional oversight of export licensing, which has been inadequate in the past.

Increasing the power of the national security agencies

The Defense, Energy and State Departments are the lead U.S. agencies on non-proliferation issues. These agencies house the experts who understand how dual-use equipment operates and what the risks are if such equipment is diverted for military purposes. They also know which countries and companies in the world are most likely to divert it. These experts are not at the Commerce Department. In order to bring the maximum amount of government expertise to bear upon export control decisions, the qualified personnel at the national security agencies must be able to decide what is controlled and who is allowed to gain access to it.

But that is not what is happening. The Commerce Department now has more influence than any other agency when it comes to determining what is controlled for export and who gets to buy it. Commerce now chairs the most important export control committees and can use its administrative preeminence to influence the outcome of licensing decisions.

I hope that this subcommittee will examine carefully the testimony given last June by Dr. Peter Leitner before the Senate Committee on Governmental Affairs. Dr. Leitner, who is a Senior Strategic Trade Advisor at the Department of Defense, explained how the influence of technical experts from the national security agencies has been diluted by making them subordinate to a committee of non-specialists chaired by the Department of Commerce.

I would also like to point to the testimony of Representative Christopher Cox, who testified in March before the Subcommittee on International Economic Policy and Trade of the House Committee on International Relations. Congressman Cox warned that mistakes are being made under the current process and that we can do a much better job of export control. He also testified that the national security agencies are not being given enough time to do a proper analysis of applications, and he warned that the national security agencies should not be outvoted on licensing cases. Because of the information Mr. Cox’s committee has recently gathered on export control, he is in a particularly good position to judge the results of our current process.

Congress should, in my opinion, insure that no license application is approved unless all the national security agencies concur. It makes no sense to allow cases to be escalated to the political level where the judgments of national security experts can be reversed by political considerations. If a national security agency takes a stand in opposition to an export application at the expert level, the case should end there.

And instead of being like poor relatives invited to dinner, the national security agencies should be put at the head of the table. Each interagency committee should be chaired by a national security agency. There is no reason to give this function to the Commerce Department, which has the least expertise in the subject matter. And the power to decide what to put on the control list should also be given to the national security agencies. Either the State or the Defense Department should be given the lead in formulating the export control list, with help from the Department of Energy for nuclear items. If export control is going to be a strategic question, instead of a trade question, then the strategic experts should be put in charge of it. This is the only division of labor that makes sense.

The Commerce Department is also burdened by a hopeless conflict of interests–it must promote exports as well as regulate them. The promotion function will always dominate, and will always cause the Commerce Department to champion the exporters’ point of view. As long as the Commerce Department is in charge of administering the export control laws, national security will take a back seat to trade interests.

Let me give an example to illustrate the problem. Last November, the State Department slapped trade sanctions on approximately 300 companies in India and Pakistan that are linked to nuclear, missile or military programs. The objective was to register U.S. disapproval of the nuclear weapon tests in May and to reduce the risk that American products would contribute to the nuclear and missile arms race in South Asia.

The Commerce Department, which opposed naming the companies in the interagency process, but which now administers the sanctions, has virtually interpreted them out of existence. Commerce has taken the absurd position that even though a company has been placed on the sanctions list, it is still okay to supply its subsidiaries or subdivisions. This is like saying that it is forbidden to sell to General Motors, but it is okay to sell to the Chevrolet division, the Buick division and the Cadillac division.

An example is Hindustan Aeronautics Limited (HAL). It makes the essential nose cones, guidance equipment and engines for India’s biggest rockets and missiles. HAL’s aerospace and engine divisions are specifically listed as “involved in nuclear or missile activities.” But the Commerce Department has taken the position that American exports are permitted to HAL’s aircraft division, its foundry division, and its design and development complex. How can anyone think that these entities are separate in any meaningful sense from their parent?

A more flagrant example is Bharat Electronics Limited (BEL) in Bangalore, India. It makes the electronic brains that guide India’s long-range nuclear missiles, the most powerful of which was tested just this week. Although BEL too is listed as an entity “involved in nuclear or missile activities,” the Commerce Department wrote a letter in March to a U.S. exporter declaring that sanctions did not apply to BEL’s “Components Division,” which is also in Bangalore, because that division was not specifically mentioned on the sanctions list.

According to an article in the Journal of Commerce, federal agents have received information that the Components Division is simply diverting imports to its parent. The Journal also reports that BEL has been faxing the Commerce Department’s letter to other U.S. exporters so that they too can supply BEL through its subsidiary. The result is that American products are continuing to fuel the missile and nuclear arms race in South Asia with the help of the Commerce Department. Commerce seems to care little about missile proliferation as long as the exports keep going out.

Congress should consider transferring jurisdiction over all dual-use licensing to the State Department, for essentially the same reasons that it just transferred jurisdiction over satellites. The Commerce Department is simply not a trustworthy guardian of U.S. national security.

The State Department already handles munitions licenses with the help of the Department of Defense. In the most recent fiscal year, State’s Office of Defense Trade Controls reviewed more than 44,000 licenses with a staff of about 55 persons. State could easily expand its efforts to cover the missile and nuclear items now controlled by the Commerce Department.

Congressional oversight

Another reason that export controls have not worked better is that Congress has not exercised sufficient oversight. The cost of this lack of oversight is shown by the example of Iraq. Congress essentially ignored export licensing to Iraq until the invasion of Kuwait. If Congress had used its oversight powers, it would have learned that the Commerce Department approved $1.5 billion worth of sensitive, dual-use American exports to Iraq from 1985 to 1990, and that many of these American products were sent directly to Iraqi mass destruction weapon sites. The Commerce Department approved the following:

  • special relays, capable of separating the stages of a ballistic missile, after the exporter told a Commerce representative that Iraq wanted the relays to be “tested for shock and vibration” and to operate 66 miles above the earth.
  • $57 million worth of navigation, guidance and other equipment for the Iraqi Air Force.
  • $557 million worth of computers and guidance equipment for the Iraqi Ministry of Defense.
  • $3 million worth of computers and diagnostic equipment for the Iraqi Atomic Energy Commission.
  • $2.7 million worth of computers, and testing and tracking equipment for Sa’ad 16, Iraq’s leading missile development site.

The United States should have learned an important lesson from the export debacle in Iraq. American pilots had to be sent to bomb what the American government had approved for export. And some of the equipment that the Commerce Department approved is probably part of what Saddam Hussein is still hiding from U.N. inspectors.

Unfortunately, the Commerce Department has a similar record on exports to China. My office has just completed a two-year study of what the Commerce Department approved for export from the United States to China from 1988 to 1998. The study, based on official Commerce Department records, found that during the past decade, Commerce approved more than $15 billion worth of strategically sensitive U.S. exports to China. The exports included equipment that can be used to design nuclear weapons, process nuclear material, machine nuclear weapon components, improve missile designs, build missile components and transmit data from missile tests.

The equipment, by definition, is of great strategic value. Only the highest performing machine tools, instruments, computers and other such items require a Commerce Department export license. This equipment has been placed on the U.S. export control list by U.S. experts who have judged that special care-and government review-is needed before releasing it to foreign countries.

Nevertheless, some of this “dual-use” equipment went directly to leading nuclear, missile and military sites-the main vertebrae of China’s strategic backbone. And several of these Chinese buyers later supplied nuclear, missile and military equipment to Iran and Pakistan.

The study shows that the military and strategic value of these legal imports exceeded by many times what China obtained by illegal means. What China got from the Commerce Department dwarfed what it got from spies. Even after purloining the design of a nuclear weapon, China still needed a large array of high-precision equipment to manufacture and test it. Commerce Department records show that it got that equipment from the United States.

More than half of the $15 billion was for computers. Until 1993, however, China was effectively denied access to high-performance computers. In that year President Clinton began to loosen export controls, and in early 1996 computer export controls were slashed dramatically. Under the relaxed rules, China has imported or is in the process of importing approximately 400 high-performance computers, the great majority of which have been or are being sold without an export license. Such machines can be used to encode and decode secret messages, to design and test nuclear warheads and to simulate the performance of a missile from launch to impact. China has refused to allow the United States to verify that these computers are being used for civilian purposes, so it must be assumed that China’s weapon scientists have access to them.

Some of the specific findings of the study are as follows:

  • The China National Nuclear Corporation was licensed to receive American computer and imaging equipment for uranium prospecting. This company then helped Iran prospect for uranium that U.S. intelligence believes will be used to make nuclear weapons.
  • The China Precision Machinery Import-Export Corporation was licensed to receive American equipment useful for building China’s new C-801 and C-802 anti-ship cruise missiles. This company then exported the missiles to Iran where, according to the U.S. naval commander in the Persian Gulf, they threaten U.S. ships and personnel.
  • The China National Electronics Import-Export Corporation was licensed to receive American equipment useful for developing radar. This company later sold Iran a powerful military radar that could someday threaten American pilots.
  • The Chinese Academy of Sciences was licensed to receive American computer equipment to help develop a nuclear fusion reactor. The Academy then exported the reactor to Iran, which U.S. intelligence believes is developing nuclear weapons.

This record of dangerous exports to both China and Iraq contain an important lesson. Congress must perform its oversight duty if American security is to be protected.

Congress should carry out this oversight in two ways. First, it should require each federal agency that participates in export licensing to file written, periodic reports with each Congressional committee and subcommittee that has jurisdiction over the agency’s performance. These reports should be filed automatically every six months. They should include a record of each licensing application on which the agency acted and should include the agency’s position at all levels of review. The record should also include the case number, the date received, the applicant, the consignee, the final date, the final action, the value, the license type, the end use, the relevant commodity information, and whether the application was approved, denied or returned without action. Receiving these reports automatically and in writing would greatly aid Congressional review of each agency’s role in the licensing process.

Second, Congress should create an independent Congressional office to review the government’s performance on export control. The office would have access to all licensing records, have subpoena power, and the power to conduct investigations. The office would function as an ombudsman. It would report to Congress and Congress would appoint the office’s director. The office could review the required periodic reports referred to above, and provide to Congress an analysis of the licensing performance of each relevant federal agency.


In addition to Congressional oversight, the licensing process needs transparency. Before the Gulf War, Iraq was able to buy sensitive equipment that the Commerce Department must have known was going to be diverted to weapons programs. The exports were approved primarily because the licensing process for dual-use equipment is secret. Neither Congress nor the public is permitted to examine Commerce Department licensing in the open. This means that only the exporters know what is being sold, and only the exporters’ voices are heard by the licensing officers when decisions are made. The effect is to freeze the public out of the process and to open the door to the worst forms of private lobbying. This is true despite the fact that dual-use licenses are supposed to be for civilian items restricted to peaceful use.

The experience of the Nuclear Regulatory Commission shows the benefit of public accountability. All of the Commission’s export licenses are granted on the public record and in the light of day. This is the main reason why there were no horror stories about U.S. nuclear exports to Iraq. Neither exporters nor regulators wanted to defend such transactions in public, so they did not happen.

To justify the present system, the Commerce Department argues that secrecy is necessary to protect proprietary interests. But the U.S. nuclear industry competes well on the international market, despite the openness of NRC licensing. That fact alone proves that secrecy is not necessary to be competitive. Indeed, there seems to be no evidence that any company would be disadvantaged if licensing data were made public. Each company would learn as much about its competitors as its competitors would learn about it.

It is also true that companies know their markets well. They know who is selling what to whom because their survival depends on it. Anything they might learn from licensing data would be only a small addition to what they already know through industrial intelligence gathering and marketing efforts. This is true of foreign companies as well as American ones. But even if there were a disadvantage to a company from having its past sales disclosed, this cost is outweighed by the strong national security interest in having an effective, publicly-accountable licensing process.

Congress should require the publication of all licensing decisions that are more than one year old. It is difficult to see what harm could result from all companies knowing what other companies had sold a year ago. Because exporters consider pricing information especially sensitive, Congress could decide not to release such information until it were two years old.

The licensing information would include the date of the application, the date and nature of the licensing decision, the applicant, the name and country of the ultimate end-user, a description of the item sold, its value, and a description of the end-use. This information already exists in a database. It could be printed by pushing a button.

The summary would include the name of the exporter. If a company is ashamed of having sold one of its products to a sensitive buyer, the company should not have made the sale in the first place. Reputable companies do not object to telling the truth about their business. If the sales are legitimate, and satisfy export laws, there is no reason to keep them hidden. The decision to license them is an official government act paid for with tax dollars. Pushing export licensing into the light of day would encourage exporters to be honest, encourage the government to be careful, and allow the public to find out whether American exports are undermining U.S. national security.

Multilateral vs. unilateral controls

One of the perennial issues in export licensing is the distinction between unilateral and multilateral controls. In my judgment, it does not make sense to pin national legislation on this distinction. A “multilateral” control cannot be defined without referring to the laws and practices of other nations. An attempt to do so leads one in a circle. To the laws of which foreign countries should one refer? How much compliance by them with their obligations is sufficient? Who decides whether the compliance really exists? If foreign countries change their laws, does Congress have to change U.S. law too? How does one avoid having the United States sink to the level of the lowest common denominator?

The use of this distinction is the main weakness of HR 361, which allows the United States to control only the things controlled by international regimes. All other controls, which are labeled “emergency” controls, lapse after twelve months. Under this approach, U.S. law would depend entirely on the laws of other nations. No self-respecting country should legislate in such a manner. It is the job of Congress to pass laws that advance U.S. interests and reflect American values, not the interests and values of other countries.

HR 361 would make it impossible for the United States to play its leadership role in export control, and would reverse a foreign policy stance the United States has maintained for over forty years. This would be an historic abandonment of America’s moral leadership. If the United States simply aped what other countries did, American companies would now be selling Iran sensitive machine tools because German firms are doing so. American companies did not sell poison gas plants to Libya and Iraq because Germany did, or sell large rockets to India because Russia did, or sell missiles to Pakistan because China did. There were no American logos on the Russian-supplied and German-enhanced Scud missiles that hit Tel Aviv during the Gulf War. U.S. exporters have “unilateral” controls to thank for that.

It is essential for the United States to be able to adopt strong controls first, and then persuade other countries to follow its example–the method by which every export control agreement since World War II has been created. U.S. diplomats are using this strategy today to help create export controls in the former East Bloc. Leadership is inherently unilateral. If the United States had waited for Europe, Japan and the Arab countries to agree on what to do when Iraq invaded Kuwait, Iraq might still occupy Kuwait today. Only by acting “unilaterally” was the United States able to forge a successful coalition.

Congress should give the President broad authority to control the export of any dual-use item that is judged relevant to the national security of the United States, the national security being taken to include combating the threat of proliferation of weapons of mass destruction and maintaining the military advantage that the United States now enjoys. The President should not be limited by statute to controlling what other countries control. If, because of the failure of other countries to control an item, the item becomes available from foreign sources, the question of maintaining U.S. controls should be dealt with through the foreign availability mechanism.