Mr. Chairman and members of the Subcommittee, I welcome this opportunity to discuss the export licensing for dual-use technology. The Subcommittee requested that I address the export control list process, as well as the export license application review process.
Dual-use technology consists of products and know how -- both tangible and intangible technology -- that have potential military use, but that are primarily commercial in design, and are in fact widely traded and used for non-military purposes. The United States and many of its allies have reviewed and controlled the export of such technology since the beginning of the Cold War. As I have testified previously before this Subcommittee, U.S. controls are coordinated with like-minded governments through various multilateral regimes.
Since 1993, in conjunction with the formation of the Wassenaar Arrangement, we have substantially streamlined both the U.S. and multilateral dual use export control lists, removing many items that were of marginal national security or foreign policy significance, particularly in the area of telecommunications and information processing. Having accomplished that major streamlining, we do not currently see major categories of items that require decontrol, except for lower level computers and microprocessors. We are always, however, considering a few individual items that are candidates for enhanced or reduced control. Since technology is changing and advancing, it is essential to continually update the control list just to keep pace with the technology.
We are constantly scrutinizing and evaluating the U.S. dual-use control list, known as the Commodity Control List (CCL). The Defense Department each year updates its Militarily Critical Technologies Lists. Each of the multilateral regimes periodically reviews its list, which may result in changes in the U.S. list. Companies, research centers, think tanks, and government agencies constantly bring items to our attention that they believe should -- or should not -- be controlled. For items that are not controlled either by the munitions list, maintained by the State Department to control munitions exports, or the dual-use list, the first issue is whether the item should be controlled as a munition. That decision is made by State and Defense. If they determine that an item does not meet the definition of a "munition" and is therefore not controlled by the Munitions List, it is then considered for possible dual-use controls. Adding an item to, or removing it from, the dual use control list is decided by consensus of the same agencies involved in the licensing review process. Regulations are issued to implement these revisions in the CCL.
One of the most difficult but important aspects of the export control process, Mr. Chairman, is determining when a product or technology which was originally used or intended for military purposes has become sufficiently commercialized to justify being treated as a dual-use item. This is particularly so now, when more and more military technologies are finding commercial applications and markets. Commercialization of military items began, I suppose, with the Jeep. Now it includes everything from night vision goggles, to traffic helicopter cameras, to satellites. All began as military products, but are now used commercially. This involves moving the item from the Munitions List to the Commodity Control List. Such transfers are determined by the State and Defense Departments. These "commodity jurisdiction" decisions can be escalated, and there is provision for appeals to the State Department, and the President, but that process has not been used often.
U.S. dual-use export controls, Mr. Chairman, consist of a combination of restrictions on products, countries, end-uses, and end-users. All U.S. products and technology are potentially subject to export controls. The regulations, however, provide detailed technical specifications of those products and technologies that actually require a license. Controls are imposed for general national security and foreign policy reasons, as well as specific non-proliferation reasons. The Commodity Control List consists of many categories of items that have been identified by the U.S. and the multilateral regimes as requiring control. It includes some items which the U.S. controls unilaterally. The CCL specifies the basis or bases for control on each item. In addition, the Export Administration Regulations specify the destinations and end-uses, and end users for which a license is required. Many items are restricted on the CCL for more than one reason. Certain machine tools, for example, are controlled for national security, missile technology, and nuclear reasons.
Evaluation of dual-use export licenses has always been a multi-agency process. The Commerce Department is both a decision-making participant, as well as the administrative custodian of the process. In its administrative role, Commerce maintains central records of license applications and decisions, prepares the relevant regulations and regulatory changes (with interagency concurrence), and issues actual licenses and license denials. In its decision-making role, the Commerce Department’s licensing officers examine and evaluate the commercial, national security, and foreign policy facts of each license application, along with counterparts in other agencies. More than half of Commerce's 50 license officers have advanced technical degrees, mostly in various engineering fields.
During the Cold War, the Department of Defense was the primary referral agency for national security cases to the Communist bloc, and was accorded the right to appeal any decision to the President under Section 10(g) of the Export Administration Act. The Departments of State and Energy reviewed foreign policy and nuclear cases, respectively.
With the end of the Cold War, and in recognition of the greater diversity and variety of threats, particularly from rogue nations, terrorists, and weapons of mass destruction, the Administration revised the interagency process. In December, 1995, President Clinton issued Executive Order 12981, which took effect in February, 1996. That Executive Order gave five Departments and agencies a direct and equal role in export licensing decision making -- Commerce, Defense, State, ACDA, and Energy. Under that Executive Order, each agency may review any and all license applications, and may appeal any decision with which it disagrees through a series of escalating interagency levels, ultimately to the President. In clearing that Order, all agencies agreed to respect the time limits and provide the statutory grounds for their positions. (As of April 1, ACDA was absorbed by the State Department and is no longer separately represented in export licensing.) The intelligence community receives applications for review at the same time they are distributed to other agencies, and participates actively as a non-voting adviser throughout the licensing process.
The three interagency dispute resolution bodies under Executive Order 12981 are the Operating Committee (OC) at the senior civil service level, the Advisory Committee on Export Policy (ACEP) at the assistant secretary level , and the Export Administration Review Board (EARB) at Cabinet level. Decisions are made at the OC by the Chair, a senior Commerce Department civil servant, usually on the basis of interagency consensus. Any dissenting agency at the OC may escalate a license decision to the ACEP. The ACEP is chaired by the Commerce Assistant Secretary for Export Administration, but decisions are made by majority vote. Cases can even be escalated beyond the EARB to the President. The EARB is chaired by the Secretary of Commerce, and decisions are likewise made by majority vote. To date, however, in the Clinton Administration, no case has been escalated beyond the ACEP.
Between 10- and 12-thousand dual-use license applications a year are being processed under these procedures. Of those, 85% are referred and decided interagency. The 15% that Commerce alone processes are mostly incomplete applications that are returned without action (RWA), some automatic denials, and a small number of routine cases which the other agencies have elected to delegate to Commerce. These Commerce actions are generally accomplished in the first 9-days of the license application process. With those exceptions, Mr. Chairman, no license is issued without the concurrence of at least two -- and usually four -- other agencies. All agencies receive referral cases simultaneously. Unless they choose otherwise, the participating agencies receive all cases. The Department of Energy has chosen to concentrate its efforts and expertise on nuclear cases, so it receives fewer referral cases than the other agencies.
Viewed as a whole, this interagency decision process accomplishes several important objectives:
First, it assures that a wide range of facts and opinions are brought to bear on each case.
Second, it encourages decision rather than indecision. With four (previously five) agencies participating, there is a strong possibility of indecision. But in dealing with commercial transactions, delay or indecision can easily constitute denial. To deal with that problem, Congress enacted specific but flexible time limits for dual use export control decisions. Those time limits are reflected in our regulations and procedures. But it takes more than time limits to achieve timely decisions. It takes effective dispute resolution -- or to put it positively, consensus building -- procedures. The current process provides those, as evidenced by the fact that 96% of all applications received are processed and resolved within the 90-day target called for in the regulations. On average, all agencies take less time than is allotted at each stage, with the Defense Department, for example, currently responding to cases in an average of 14 days rather than the allotted 30.
Some observers have suggested that the current time limits are inflexible, risking hasty decisions with incomplete information. That is simply not the case. If any additional information is necessary, any participating agency can "stop the regulatory clock" by simply posing relevant questions for the applicant or other sources. There is no limit on the number of questions that can be posed or the time available to answer them. This means of extending the time for analysis and decision is frequently used. Any agency’s request for a pre-license check, to conduct consultations with other governments, or consult with other governments are also grounds for "stopping the clock."
Third, the process allows cases that raise factual uncertainties, policy issues, and sometimes sharp differences of opinion to be escalated to the highest levels of government. Any agency involved in the process can appeal if they remain dissatisfied at any stage of the process. This provides a strong incentive for agencies to reach consensus by responding to each others concerns. That, of course, may be a consensus for approval or for denial.
About 93% percent of all cases are resolved by consensus among the licensing officers and their management without escalation. The first level of escalation is the Operating Committee. For cases in disagreement after 30-days or more of consideration, the OC is the "work horse" of this process. Of the 7% of all cases that go to the OC, all but 5% (less than 1 % of all cases) of those are resolved there at the OC through face-to-face interagency information exchange, debate, and consensus building. Nevertheless, some observers of export control policy have focussed on the OC, noting particularly the fact that the Commerce Department Chair decides cases at this level. Agency positions are noted and recorded in the OC records, but formal votes are generally not taken. On occasion, the OC Chair even rules against the Commerce Department position. More than half of the 777 cases escalated to the OC in FY 1998 were ultimately decided there by consensus. Only about 5% were decided by the Chair contrary to the majority, and virtually all of those were escalated to the ACEP for final decision by majority vote.
Of the 70 cases escalated to the ACEP in the past two years (Feb. 1997 to Feb. 1999), 5 were approved over the remaining dissent of one or more agencies, while 40 were approved unanimously. Eight were denied and 17 returned without action. Only 6 of these decisions overturned the decision of the OC Chair. The rarity of cases appealed to the ACEP and the high rate of approval of OC decisions are strong indications that this process works well to reach sound, timely decisions.
The technologies that have mostly frequently been the subject of OC and ACEP cases decided without complete consensus are:
ECCN Product Category Description
3B001 Semiconductor equipment
Various Nuclear items
Category 6 Sensors
6A003 Infrared cameras
2B001 Five-axis machine tools
There are various reasons for failure to reach consensus at the OC. In most cases, Mr. Chairman, the agencies themselves have concluded that a case should be decided at higher levels because it raises policy questions, and they hold to differing positions precisely in order to escalate the case. In such instances, it is misleading to suggest that the OC chair acted contrary to the wishes of the agencies by making a ruling that results in escalation, although that might appear to be the case.
If cases can be decided at the ACEP by majority vote of the agencies, why not use majority rule at the OC? Or even require consensus (unanimity)? A requirement of consensus at the OC would give every agency a veto over every case, removing the incentive to reach agreement and encouraging delay and indecision. Especially at lower decision levels, it is always easier to say "no", or take no position, than to defend a "yes." In the pre-1996 process, this led to too many cases falling into a "black hole" of indecision. We have replaced that with a "default to decision" through a combination of time limits, authority for the Chair to resolve cases at the OC level, and majority vote and the ACEP (and Export Administration Review Board) levels. Limited and preliminary as it is, the ability of the OC Chair to decide cases at the senior civil service level provides a needed incentive for agencies to reach consensus -- to make "default to decision" work.
Enabling the OC Chair to rule on cases, of course, means that some agencies must justify their positions or risk being overridden. But that does not mean that their views or concerns are ignored. Very often, the positions put forward initially by agencies are themselves preliminary. Often, the information and analysis assembled at the OC enables agencies willingly to change their positions on grounds that their concerns have been satisfied. In many cases, strict conditions are placed on licenses to meet particular agency concerns. In some cases, the proposed transaction is modified (with the concurrence of the applicant) in order to reach a decision.
Implicit -- and sometimes explicit -- in criticisms of this process is the contention that it is biased in favor one agency or another. Nothing could be further from the truth. That it is biased in favor of commercial interests is particularly unsupportable. To begin with, of the four (previously five) participating agencies, three (previously four) are charged primarily with furthering U.S. arms control and non-proliferation policies. If anything, the process is weighted in favor of national security and foreign policy concerns. Only if a majority of agencies can satisfy themselves on the basis of the facts of the case and relevant policy that national security and foreign policy concerns are met can any case be approved.
It is sometimes alleged that Commerce Department involvement constitutes a "dumbing down" of this process. Press reports often loosely refer to the Commerce Department's role in licensing as "lax", or worse. The fact is that most dual use applications reflect genuine commercial sales to non-military end users that pose little risk or threat to U.S. national security interests. But that is not assumed in the license review process. It must be clearly and persuasively established in each and every case. That is what the Commerce Department manages, and the other agencies participate in, so effectively. Simplistic studies of aggregate licenses issued to this or that country fail to take account of the facts of each case that are the basis for each individual decision. Anyone with experience in export licensing is aware that many cases are not what they appear to be on the surface. Conclusions drawn from studies that ignore the facts of each case are misleading at best.
No process is perfect, Mr. Chairman, especially when it comes to the difficult task of balancing economic, national security, and foreign policy interests in specific cases. This is a highly informed processed handled to a large extent by experts from the various agencies. But -- especially when the experts disagree -- it is also a democratic process, and like democracy itself, it can be said to be the worst method of making decisions except for all the alternatives. In such a process, there are always individuals who feel that their views and conclusions have not been accorded sufficient weight. Sometimes those individuals violate their professional responsibilities by leaking confidential information that supports their views knowing that the agencies cannot respond publicly without revealing additional -- sometimes very sensitive -- information. Neither the Congress nor any Administration should let such tactics rule the day.
Many of the criticisms of the dual use export control process seem, in fact, to reflect disagreements over policy. Such policy disagreements are legitimate. We should recognize them as such, discuss them, and work together to resolve them. We should not, however, try to resolve them by discrediting or placing undue blame upon a process that is working and reaching objective decisions through careful interagency consideration. The Administration looks forward to working with this and other Committees of the Congress toward that end.
In April of 2002 the Bureau of Export Administration (BXA) changed its name to the Bureau of Industry and Security(BIS). For historical purposes we have not changed the references to BXA in the legacy documents found in the Archived Press and Public Information.