The Foreign Policy Essay: Preventing the Proliferation of Armed Drones
Editor’s Note: Drone warfare and its many implications is a favorite subject for Lawfare readers. Yet even as the United States develops policies for the use of drones on and off the battlefield, it must contend with their proliferation to other countries. Indeed, while many voices continue to call for limiting this new form of warfare, the market for drones, especially U.S. drones, is expanding. Sarah Kreps, a professor at Cornell and author of the forthcoming book Drone Warfare, presents us with several questions to consider as we ponder U.S. export policies on drones.
Published by The Lawfare Institute
in Cooperation With
Editor’s Note: Drone warfare and its many implications is a favorite subject for Lawfare readers. Yet even as the United States develops policies for the use of drones on and off the battlefield, it must contend with their proliferation to other countries. Indeed, while many voices continue to call for limiting this new form of warfare, the market for drones, especially U.S. drones, is expanding. Sarah Kreps, a professor at Cornell and author of the forthcoming book Drone Warfare, presents us with several questions to consider as we ponder U.S. export policies on drones. She argues that drones are a destabilizing technology and that the United States should foster nonproliferation norms and build institutions to counter their spread.
Second, is drone proliferation inevitable? If so, it makes little sense to worry about whether American export policy is liberal or stringent and, in fact, U.S. businesses may as well prosper. But there are good reasons to think that the technology will not otherwise seamlessly diffuse. In some countries, such as Germany, the domestic political environment is hostile towards acquiring armed drones and has put a pause on previous plans to acquire drones. A bigger reason is technological. While one can go onto Amazon.com and buy a rudimentary drone (basically just a remote-controlled airplane), constructing an advanced armed drone is no small feat. U.S. armed drones require sophisticated beyond-line-of-sight communications, access to satellite bandwidth, and systems engineering—from internal fire control to ground control stations—that are currently beyond the reach of most states.
Even countries that have relatively advanced aerospace programs—such as Russia, France, and Italy—have struggled to develop and deploy this systematic architecture of capabilities and processes. Russia, which has experienced a number of setbacks in its aerospace industry since the end of the Cold War, has been frustrated in its efforts to develop more advanced drones. In January 2010, an armed drone prototype of Russia’s Stork Unmanned Aerial Vehicle (UAV) crashed and burned as it attempted to take off, providing further evidence that Russia is decades behind the United States in UAV technology. France and Italy have not been able to produce the requisite technology indigenously and have therefore been limited to purchasing unarmed versions of the United States’ MQ-9 Reaper. Despite many countries expressing an interest in drones, only Israel, China, and perhaps Iran have indigenously produced advanced armed drones. This is not a story of rampant armed drone proliferation.
Third, what is the status of a nonproliferation norm when it comes to armed drones? Currently, the Missile Technology Control Regime (MTCR) regulates the transfer of unmanned aerial vehicles. But drones are a bolt-on to a regime that was intended to restrict the spread of nuclear weapons delivery systems. The regime was far from perfect in this original function since it was a non-legally binding agreement among 34 countries, most of whom are advanced industrialized countries. Excluded are countries such as China, India, Iran, and Israel (although Israel purports to adhere unilaterally).
The features of the regime that deal with drone exports are equally problematic. Because the MTCR was intended for nuclear delivery systems, the thresholds it establishes seem arbitrary when it comes to drones. It classifies technology into two categories: Category I systems, which exceed a range of 300 kilometers and a payload of 500 kilograms, and Category II systems, which are below that threshold. The guidelines state that exports of Category I systems are “subject to an unconditional strong presumption of denial”—that is, there is an assumption that countries will not export systems in this category. Category II systems are “subject to licensing requirements.” In general, Category I corresponds to the export of armed drones that are medium altitude long endurance (MALE), such as armed Reapers or Predators. But as technology improves, it becomes lighter, and armed drones that fall under that payload threshold will eventually be developed. For example, responding to requests from the United Arab Emirates (UAE), General Atomics has designed a Category II version of the Predator (XP) for export. While this is preferable to exercising the rare exception to the presumption of denial and exporting a Category I system, it points to the ways that industry can circumvent the arbitrary thresholds stipulated in the MTCR.
As this analysis suggests, armed drones are potentially destabilizing and their proliferation is not inevitable, but the nonproliferation regime intended to regulate the transfer of the technology is far from airtight. The United States could make the argument that a number of other potential drone exporters such as China and Israel are not bound by the MTCR, so why should it play with one hand tied behind its back? This scenario of other countries capturing market share—as we saw happen in the satellite industry, where tighter U.S. export regulations in the 1990s enabled foreign companies to gain market share at the expense of U.S. companies—is frequently put forward as an illustrative tale by those who advocate greater liberalization of U.S. drone export policies.
But the answer is not for the United States to step away from its MTCR commitments. Rather, the answer is to try to bring other potential drone exporters into the nonproliferation fold. One way to do this is to create a UAV-specific nonproliferation regime that would focus on the regulation of drones based on mission type rather than payload, focusing on armed drones to avoid the arbitrary thresholds embedded in the MTCR. The regime would include not just the 34 advanced industrialized countries that are part of the MTCR, but also the countries that actually have or could soon acquire armed drones.
To be sure, such a multilateral arrangement raises thorny issues. Any modifications to the MTCR, which would essentially be needed if drones were handled by a separate regime, require a consensus vote of MTCR members, and members such as Russia might not share the same nonproliferation incentives as the United States. More nettlesome is the challenge of co-opting non-MTCR members—especially China, Israel, and Iran—since their exports might not be confined to their own regions, and their interests are not as global as those of the United States (both of which minimize any destabilizing consequences). In both contexts—modifying existing institutions and eliciting a broader membership for a new institution—U.S. confidence-building concessions in terms of its own use—for example, increasing transparency on how it applies international humanitarian law and self-defense law to its policy of armed drone strikes (and whether it applies international human rights law)—would be a likely prerequisite. These types of concessions have been typical in the context of nuclear non-proliferation—for example, in 1995 when the nuclear weapons states agreed to make more aggressive progress on their own disarmament in exchange for non-nuclear states agreeing to an indefinite extension of the Treaty on the Non-proliferation of Nuclear Weapons (NPT). Concessions in the context of increased transparency on its drone program would be a small price for the United States to pay for tying the hands of would-be armed drone proliferators.
In short, the answer to the export dilemma is not a race to the bottom, a free-for-all where states vie for market share, since even some armed drones in the wrong hands would have destabilizing consequences for regional and international security. The answer is to develop appropriate nonproliferation institutions that will help regulate the transfer of this technology. The United States, working with other countries, should take steps in that direction.
***
Despite lingering questions about whether armed drone strikes are legal or ethical, a number of countries have indicated that they want what the United States has and are trying to import American technology. After all, what’s not to like about the capacity to conduct counterterrorism missions without incurring meaningful risk? To date, the United States has only exported armed drones to the United Kingdom, but a question under consideration by an interagency review process set up by the Obama administration is whether the United States should liberalize its exports. Especially in an era of declining defense budgets at home, the prospect of selling more armed drones abroad looks attractive as a way to prop up the U.S. industrial base. Whether it should or not depends on the answers to three questions. First, are drones qualitatively different from other weapons platforms and is this technology destabilizing? Yes, drones should be treated as a distinct class of weapons and yes, their attributes can cause them to be used in ways that are potentially destabilizing. The main difference between drones and other platforms is that they are unmanned and therefore pose no risk to those operating them. For the United States, this means that drones have expanded the military’s range of operations to include many that would have been too risky to attempt with other platforms. Of the estimated 465 non-battlefield targeted killings undertaken by the United States since November 2002, approximately 98 percent were carried out by drones. If the U.S. experience is any guide, states equipped with armed drones will be more willing to use force in ways and in areas they might not otherwise have. Armed drone proliferation in regions that are already crisis prone such as the Middle East, the Caucasus, or East Asia could potentially lower the threshold for using force, making these combustible regions even more volatile.
Second, is drone proliferation inevitable? If so, it makes little sense to worry about whether American export policy is liberal or stringent and, in fact, U.S. businesses may as well prosper. But there are good reasons to think that the technology will not otherwise seamlessly diffuse. In some countries, such as Germany, the domestic political environment is hostile towards acquiring armed drones and has put a pause on previous plans to acquire drones. A bigger reason is technological. While one can go onto Amazon.com and buy a rudimentary drone (basically just a remote-controlled airplane), constructing an advanced armed drone is no small feat. U.S. armed drones require sophisticated beyond-line-of-sight communications, access to satellite bandwidth, and systems engineering—from internal fire control to ground control stations—that are currently beyond the reach of most states.
Even countries that have relatively advanced aerospace programs—such as Russia, France, and Italy—have struggled to develop and deploy this systematic architecture of capabilities and processes. Russia, which has experienced a number of setbacks in its aerospace industry since the end of the Cold War, has been frustrated in its efforts to develop more advanced drones. In January 2010, an armed drone prototype of Russia’s Stork Unmanned Aerial Vehicle (UAV) crashed and burned as it attempted to take off, providing further evidence that Russia is decades behind the United States in UAV technology. France and Italy have not been able to produce the requisite technology indigenously and have therefore been limited to purchasing unarmed versions of the United States’ MQ-9 Reaper. Despite many countries expressing an interest in drones, only Israel, China, and perhaps Iran have indigenously produced advanced armed drones. This is not a story of rampant armed drone proliferation.
Third, what is the status of a nonproliferation norm when it comes to armed drones? Currently, the Missile Technology Control Regime (MTCR) regulates the transfer of unmanned aerial vehicles. But drones are a bolt-on to a regime that was intended to restrict the spread of nuclear weapons delivery systems. The regime was far from perfect in this original function since it was a non-legally binding agreement among 34 countries, most of whom are advanced industrialized countries. Excluded are countries such as China, India, Iran, and Israel (although Israel purports to adhere unilaterally).
The features of the regime that deal with drone exports are equally problematic. Because the MTCR was intended for nuclear delivery systems, the thresholds it establishes seem arbitrary when it comes to drones. It classifies technology into two categories: Category I systems, which exceed a range of 300 kilometers and a payload of 500 kilograms, and Category II systems, which are below that threshold. The guidelines state that exports of Category I systems are “subject to an unconditional strong presumption of denial”—that is, there is an assumption that countries will not export systems in this category. Category II systems are “subject to licensing requirements.” In general, Category I corresponds to the export of armed drones that are medium altitude long endurance (MALE), such as armed Reapers or Predators. But as technology improves, it becomes lighter, and armed drones that fall under that payload threshold will eventually be developed. For example, responding to requests from the United Arab Emirates (UAE), General Atomics has designed a Category II version of the Predator (XP) for export. While this is preferable to exercising the rare exception to the presumption of denial and exporting a Category I system, it points to the ways that industry can circumvent the arbitrary thresholds stipulated in the MTCR.
As this analysis suggests, armed drones are potentially destabilizing and their proliferation is not inevitable, but the nonproliferation regime intended to regulate the transfer of the technology is far from airtight. The United States could make the argument that a number of other potential drone exporters such as China and Israel are not bound by the MTCR, so why should it play with one hand tied behind its back? This scenario of other countries capturing market share—as we saw happen in the satellite industry, where tighter U.S. export regulations in the 1990s enabled foreign companies to gain market share at the expense of U.S. companies—is frequently put forward as an illustrative tale by those who advocate greater liberalization of U.S. drone export policies.
But the answer is not for the United States to step away from its MTCR commitments. Rather, the answer is to try to bring other potential drone exporters into the nonproliferation fold. One way to do this is to create a UAV-specific nonproliferation regime that would focus on the regulation of drones based on mission type rather than payload, focusing on armed drones to avoid the arbitrary thresholds embedded in the MTCR. The regime would include not just the 34 advanced industrialized countries that are part of the MTCR, but also the countries that actually have or could soon acquire armed drones.
To be sure, such a multilateral arrangement raises thorny issues. Any modifications to the MTCR, which would essentially be needed if drones were handled by a separate regime, require a consensus vote of MTCR members, and members such as Russia might not share the same nonproliferation incentives as the United States. More nettlesome is the challenge of co-opting non-MTCR members—especially China, Israel, and Iran—since their exports might not be confined to their own regions, and their interests are not as global as those of the United States (both of which minimize any destabilizing consequences). In both contexts—modifying existing institutions and eliciting a broader membership for a new institution—U.S. confidence-building concessions in terms of its own use—for example, increasing transparency on how it applies international humanitarian law and self-defense law to its policy of armed drone strikes (and whether it applies international human rights law)—would be a likely prerequisite. These types of concessions have been typical in the context of nuclear non-proliferation—for example, in 1995 when the nuclear weapons states agreed to make more aggressive progress on their own disarmament in exchange for non-nuclear states agreeing to an indefinite extension of the Treaty on the Non-proliferation of Nuclear Weapons (NPT). Concessions in the context of increased transparency on its drone program would be a small price for the United States to pay for tying the hands of would-be armed drone proliferators.
In short, the answer to the export dilemma is not a race to the bottom, a free-for-all where states vie for market share, since even some armed drones in the wrong hands would have destabilizing consequences for regional and international security. The answer is to develop appropriate nonproliferation institutions that will help regulate the transfer of this technology. The United States, working with other countries, should take steps in that direction.
***
Sarah Kreps is an assistant professor of government and adjunct professor of law at Cornell University and a Stanton Nuclear Security Fellow at the Council on Foreign Relations. She is the author of Coalitions of Convenience: Military Interventions after the Cold War and, most recently, the forthcoming book Drone Warfare. Previously, she served as an acquisitions and foreign area officer in the United States Air Force.
Sarah Kreps is the John L. Wetherill Professor of Government, adjunct professor of law, and director of the Tech Policy Institute at Cornell University. She is also a nonresident senior fellow at the Brookings Institution and a member of the Council on Foreign Relations. She is currently finishing a book on the policy responses to emerging technologies from nuclear weapons to artificial intelligence.
