Wa, Wa, Wassenaar!
After the United States announced its new export control rules on semiconductors and semiconductor manufacturing equipment October 7 of last year, the first question asked was whether the controls were unilateral or multilateral. It has long been understood that in a globally integrated trading system where everything is made everywhere, unilateral controls are an exercise in futility. As my successor (several times removed) at the Commerce Department, Eric Hirschhorn, has said, unilateral controls are like damming half the river.
The answer to that question about the October 7 rules was initially, yes, they are unilateral, but over the succeeding several months, yes turned into no, as the administration obtained agreements from Japan, the Netherlands, and Taiwan to impose similar controls on chips and tools. Although details are still being worked out, this was a significant victory for the administration and a sign that producers of advanced technology products were prepared to stick together on a common approach to controlling them.
Subsequently, the public debate has moved on to focus on the questions many of us raised about the controls from the beginning—their effect on U.S. company revenues, how the Chinese will respond, and whether they will lead to another episode of third countries designing U.S. technology and components out of their products in order to avoid triggering U.S. controls. (The answers, for those of you following this issue closely, so far appear to be a negative impact on U.S. company revenue; an acceleration of China’s plans to develop its own technology and a search for retaliation that hurts us more than them, and the beginning of design out efforts that will take some time to mature.)
Lurking behind those questions, however, is the issue of how to create and maintain a lasting multilateral control architecture. In the short run, the administration has finessed that issue by cutting deals with the relevant foreign producers of the products we want to control. Fortunately for us, in the chip and tool sectors, that is a small number of other countries, and they are allies. If the United States decides to expand its controls in the semiconductor sector or branch out into other areas of critical technology, however, the list of other producers may not be so short, and the challenges of creating a multilateral united front will be much greater.
Multilateral approaches to export controls have a long history, beginning with the Coordinating Committee on Multilateral Export Controls (COCOM), during the Cold War, which forged an allied consensus on exports to the Soviet Union and its satellites. Subsequently, separate control regimes were developed for specialized purposes—the Nuclear Suppliers Group, the Missile Technology Control Regime, and the Australia Group (for chemical and biological weapons). After the collapse of the Soviet Union, former COCOM members developed what came to be known as the Wassenaar Arrangement (named after a town in the Netherlands) to control dual-use items and weapons not covered by the other regimes. Wassenaar differs from COCOM in its lack of a specific target and a process for vetoing members’ exports. It also differs from COCOM in that it includes Russia, Ukraine, South Africa, and India in its total of 42 members.
While Wassenaar has functioned successfully since 1996, recently its effectiveness has been questioned, largely for two reasons. First, Russia is a member, which has made reaching consensus on anything virtually impossible. Second, it moves too slowly for an era of rapid technological change. With only one decisionmaking plenary meeting per year and a long lead time in preparing for it, it has proved difficult for the organization to keep up with the times even when consensus might be reached. This has led to a debate over how to fix Wassenaar and over alternative architectures that might be better.
Proponents of the former note the advantages of an existing structure, secretariat, and control list over trying to build something new from scratch. The obvious, and probably only, “fix” would be “Wassenaar minus one,” meaning expelling Russia. The problems with that are that there is no expulsion mechanism in the Wassenaar Arrangement, and, in any event, other members would likely oppose such an action, so it would be Wassenaar minus three or four rather than minus one.
Proponents of a new architecture point out that today’s sophisticated technologies are not, in fact, made everywhere, and that it would make more sense to develop a series of “mini-regimes” each focusing on a specific technology sector and including only the small number of countries that actually produce the controlled items. This would make reaching consensus much easier while still including all the relevant producers. It is essentially what the United States has done with respect to semiconductor manufacturing tools.
Both approaches have their pros and cons. I favor the mini-regime structure because I think it would be easier to create and easier to operate successfully, but I can see the advantages of a reinvented Wassenaar. Two things, however, are clear: multilateralism is the only viable approach to high-tech export controls, and existing structures are not adequate to the task. The United States had great success in building a coalition that imposed sanctions and controls on Russia after its invasion of Ukraine. It should move more quickly to reconcile internal differences over how to build a multilateral structure that can best control high-tech exports to China so it can repeat its 2022 success.
William Reinsch holds the Scholl Chair in International Business at the Center for Strategic and International Studies in Washington, D.C.