“We continue to count on the valuable contributions of the Arms Control Association.”
How to Reach Consensus on an Arms Trade Treaty
Since 2006, government representatives to the United Nations have been engaged in progressively wider and deeper discussions toward the negotiation of a treaty to regulate the trade in conventional arms. In 2009 the UN General Assembly decided to convene a conference that established a timetable for crafting an arms trade treaty (ATT), which is to be “a legally binding instrument on the highest-possible common international standards for the transfer of conventional arms.”
This decision established a series of four preparatory committee meetings, of which three have been held since 2010. Under the agreed program, there will be a short preparatory committee meeting in February, primarily on procedural issues, and the negotiations will start in July.
This might appear to be a protracted process, but by the standards of multilateral negotiations, the pace has been remarkable. Nevertheless, with only 22 days of talks left, it is time to focus on issues that will increase the prospects for success and determine what a treaty might look like. A key issue in this regard is why the defense industry should be involved in the debate, a point that this article will address in some detail.
The aim of an ATT is to regulate global trade in conventional arms more effectively, not to reduce or to limit the scope for legal trade. The envisioned treaty would represent a new kind of instrument that would not fall neatly into the category of disarmament or arms control.
There is enormous support for an ATT among UN member states.[2] For example, 151 states voted in favor of the last UN resolution on an ATT in 2009. However, the five years of talks have shown that there are probably more than 100 different views of what the treaty should look like. As always in these negotiations, the devil is in the detail. Therefore, a key question for the negotiators is whether it is possible to avoid detail in the text of the treaty.
Prospects for Success
The chair of the process, Ambassador Roberto García Moritán of Argentina, has done a remarkable job in marshaling the discussions, gathering a collection of disparate views into a chairman’s draft that has been circulated to governments and observers in the preparatory committee meetings. He has said clearly and repeatedly that this is not a draft treaty text; indeed, it carries no formal status. Yet, that does not stop many people from thinking or claiming otherwise. That view needs to be dispelled. In the UN, such misconceptions often lead to considerable time being lost down rabbit holes.
Distilling from the draft text the elements that will constitute a strong, effective, and implementable treaty remains a significant challenge. Arguably, 70 to 80 percent of the issues are close to agreement; it is the remaining 20 to 30 percent that will lead to some late nights at the UN. Before addressing these issues, however, the delegates most likely will have to overcome some procedural obstacles.
During preparation of the 2009 resolution, a tremendous amount of heat, but little light, was generated in a debate over whether the negotiating conference should be undertaken on the basis of consensus. Following U.S. votes against ATT resolutions in 2006 and 2008, the need for consensus was a critical element in obtaining a new U.S. position of support.[3] Some long-standing advocates and observers felt this price was too high, arguing that it may give states the ability to stall negotiations and dilute the final treaty.[4] Nevertheless, having the United States, the largest exporter of conventional arms, positively engaged in the process was and remains vital to its success.
Clearly, the change of administration in the United States and the resulting desire to engage in a multilateral process alongside some of its closest allies and friends were major factors in the change of U.S. position. Yet, other factors also were important. First, a treaty in some form is likely to be concluded, if not in 2012, very soon after. Quite rightly, Washington wants to make sure the treaty does not undermine U.S. objectives: preventing arms from falling into the hands of those that would threaten its homeland, its military overseas, or its friends and allies. A weak treaty also could provide a fig leaf for governments less willing to address the need for responsible controls on arms exports, undermining years of efforts to deal with states that proliferate or divert arms.
Second, the United States needs to be at the table to protect its interests. Actively participating in the negotiation gives it the chance to promote the export control agenda that it has pursued since the early days of the Cold War, an agenda that focuses on promoting national responsibility and accountability for transfers. Moreover, once a treaty is completed, a state that is not a signatory will not have a place in meetings of states-parties that would define future implementation and adaptation of a treaty.
Other significant players such as China and Russia, which both had abstained on the 2009 resolution, now have taken a role alongside the United States, albeit from slightly different positions. Both had remained skeptical of the need for a treaty. One of the most significant breakthroughs came last July during the third preparatory committee meeting. The five permanent members of the UN Security Council—China, France, Russia, the United Kingdom, and the United States, known as the P5—made a joint statement[5] supporting the efforts aimed at establishing an “international instrument.” This choice of words was notable because an international instrument could be a document that is politically but not legally binding and therefore would be viewed by many as much weaker than a treaty. Perhaps this reflected remaining reservations on the part of China and Russia about the scope and purpose of any treaty as it did not entirely reflect the mainstream view in support of a comprehensive, legally binding treaty establishing the highest-possible common international standards for the transfer of conventional arms.
A consensus rule certainly helped to bring three of the largest arms traders—China, Russia, and the United States—to the table. Crucially, however, with Washington now at the table, actively pursuing a strong and robust treaty, it was perhaps unrealistic for Beijing and Moscow to continue to abstain.
It remains unclear whether the shift by China and Russia makes consensus more likely. Achieving consensus largely depends on a text the P5 can accept, as those five states represent the preponderance of the export and import trade in conventional arms. Nevertheless, they are not the only players in a process that must be open and transparent and take full account of the views of all member states. To have a significant treaty supported by a broad range of states will require compromise from all. There remains a critical need for a dialogue among proponents, the P5, and those less convinced of the need or feasibility of a treaty.
Industry Involvement Needed
As momentum builds in the diplomatic process toward the July conference, discussion now also must focus on practical aspects of the treaty’s scope and implementation. Creating idealistic or compromise language that fails to take account of the practical aspects of implementation will critically undermine the treaty’s longer-term success. Industry has an important role to play.
Governments and nongovernmental organizations (NGOs) sometimes do not fully understand the enormous scale and complexity of the global supply chain in defense goods and technology. From its experience in this area, the defense industry understands the dynamics and the practicalities, including the negative effects that could result from imprudent language that is impractical to implement. Along with governments, industry will be one of the major stakeholders in the treaty, the ones who will be most directly affected by its implementation.
The few industry representatives that have attended recent preparatory committee meetings have noted with some concern an absence of such expertise. The fact that the ATT process was established and subsequently developed within the UN General Assembly’s First Committee, the body that deals with disarmament and international security, has meant that an ATT is viewed through a disarmament lens and is being negotiated by disarmament diplomats. Although there was no other UN forum in which it could reasonably be discussed, the disarmament and arms control label is distinctly unhelpful.
It is both natural and essential that industry in all regions and of all views be involved. A number of the delegations and García Moritán are now calling for the close involvement of industry.
Perhaps because of a lack of clarity about the nature of an ATT, with the exception of some European companies, the larger aerospace and defense industry has been unaware of a process that could have a significant impact on its business. It also is possible that many merely thought the UN process would not progress as fast as it has. Some in the U.S. industry believed there was little chance in the early stages of the U.S. government signing a treaty, even if one emerged. British industry’s discussions with its overseas partners underscored this point, but many U.S. companies were simply unaware of an ATT as an issue.
This is not entirely industry’s own fault. Without government outreach to industry, the labyrinthine processes of the UN, involving resolutions, groups of governmental experts, open-ended working groups, and preparatory committees, is remarkably opaque to outsiders, and the potential implications often remain remote.
One exception to this lack of awareness was in the firearms industry. That sector has been sensitized to UN processes, often with a real fear that these would threaten or undermine its business. Such fears arose most notably in connection with the UN “Programme of Action to Prevent, Combat and Eradicate the Illicit Trade in Small Arms and Light Weapons in All Its Aspects” in 2001.[6] Alongside the firearms industry stood the sports shooting community, which also perceived real threats to its interests. At the time and on that specific initiative, some of those concerns probably were reasonable.
In the context of an ATT, these concerns swept in at a very early stage. Clearly, small arms and light weapons were on the agenda as they form part of a broad menu of conventional arms that would be discussed. Many states, particularly in Africa, attach the highest importance to issues of small arms and light weapons. This is entirely understandable as they are most affected by the proliferation of such weapons. They viewed an ATT as a means to address what they saw as failures by or omissions from the UN program of action process. Issues relating to small arms and light weapons became a dominant part of the early ATT discussions. As a result, the legacy concerns of the firearms industry and sports shooting community were resurrected. Regrettably, this has been erroneously skewed toward an unhelpful exchange between proponents of an ATT—governments and NGOs—on one side and those representing the sports shooting and civilian gun ownership community on the other. Calls from some states and a majority of NGOs to have the broadest possible definition of small arms and light weapons, including civilian firearms, have merely fueled that debate. This perpetuates the erroneous perception that an ATT is focused primarily on these weapons. More narrowly but no less importantly, the sports shooting community fears that issues of transfers within national borders and private ownership will be included in an ATT’s scope. Both of these fears should have been allayed. As the report of the 2008 UN group of governmental experts on an ATT concluded,
It was also mentioned that, were an arms trade treaty to be considered feasible, it would need to reflect respect for the sovereignty of every State, without interfering in the internal affairs of States or their constitutional provisions, and respect for their territorial integrity. Exclusively internal transfers or national ownership provisions, including national constitutional protections on private ownership within that State’s territory, should not fall under an arms trade treaty.[7]
That redline remains for many states, including many proponents such as the United Kingdom, and certainly for the United States. Because decisions must be made by consensus, any perceived encroachments on constitutional rights can easily be prevented.
Furthermore, it is important to note the global object and purpose of a future ATT. Issues of civilian ownership and sports shooting, important though they are in their context, simply have no place in an ATT. It would be a shame if this debate, over an issue explicitly ruled out at the earliest stage and reaffirmed in the 2009 resolution establishing the negotiating conference,[8] were able to continue sucking oxygen from the broader debate over the trade in conventional arms. One effect of the debate over small arms and light weapons is that it prevents the voice of the larger aerospace and defense industry community from being heard.
Covered Categories
The seven major categories of conventional arms likely to be included, taken broadly from the UN Register of Conventional Arms, are tanks, military vehicles, artillery systems, military aircraft, military helicopters, naval vessels, and missiles and missile systems. In addition to these, small arms and light weapons are being pursued as an additional category—an ordering that many in the earliest days called the package of “7+1.” During the First Committee debates in October 2009, one industry group noted the discussions on scope had been turned upside down and were more about small arms and light weapons than the seven major categories of conventional weapons and thus more like “1+7.”[9]
Discussion appears to be moving toward a broad interpretation of the seven categories plus small arms and light weapons. This broad interpretation is important, as is avoiding any attempt to define any of the categories. Such efforts have proven difficult in the context of the UN Register, and there is no reason to expect them to be different in an ATT. The direction now appears to be toward allowing states to define the coverage precisely in their own national legislation. This is the right way to go, in part because it would avoid just the kind of detail that would stall negotiations.
Including ammunition is a problem and will likely remain so for many states if there were to be any suggestion of reporting on ammunition transfers. Many states consider such information to be sensitive and, for large exporters, extremely burdensome to collect.
Another contentious issue is the inclusion of many thousands of dual-use items. A number of governments have recognized the practical problems with defining and including these items, and it seems doubtful that they will be part of the eventual scope. The inclusion of transfers of technology also raises practical problems. Many states define this term quite broadly;[10] the danger is that many millions of technology transfers across the defense industry, for example relating to minor design issues or product support, could be brought within the scope of a treaty. These would be irrelevant to its purposes and would prove bureaucratically difficult to license and manage.
One can envisage including technology in the scope of licensed production arrangements, where it could be part of the transaction, combining with hardware, machine tools, and other elements to form a defined package, rather than attempting to define it generally. In this way, there would be no ability to evade treaty controls by transferring complete production arrangements, including the associated technology, to a non-ATT territory. Industry favors closing such loopholes.
Why Industry Should Care
Since 2006, British and other European industry representatives have worked increasingly closely with NGOs and governments to drive the process toward an effective and implementable treaty.
The defense industry places the highest priority on compliance with export controls wherever it operates. The majority of the industry already is heavily regulated, but that regulation is by no means universal in scope or effect. This uneven landscape provides one of the key arguments deployed in support of an ATT: the need to plug gaps. For industry, the unevenness also presents enormous challenges in achieving concurrent compliance in all the jurisdictions in which it operates. Thus, an initiative that genuinely seeks to establish a global benchmark is to be welcomed and supported.
The moral argument for an ATT, which encompasses human security and social and economic development, is overwhelming. In no way is this incompatible with good business practice. Corporate social responsibility is a fundamental part of effective business strategy. In the majority of companies, there is a strong belief that reputation matters and that conducting business in an ethical and responsible way brings competitive advantage.
The current economic climate brings new drivers and challenges that are relevant to the ATT debate. At no time in recent memory has the concept of a “level playing field” been as important as it is today. Pressure on industry and governments has grown. Driven at least in part by smaller slices of the cake being available to high-value manufacturers, this pressure presents the increased risk of suppliers in some countries entering or revisiting markets that would have been considered off-limits under other circumstances. This outcome would be highly undesirable, certainly from a security and human rights perspective.
Many governments have expressed a desire to rebalance their economies, moving toward high-value manufacturing and new wealth-creating industries. They are looking at more-active approaches to strategic industrial policy, despite pressure on public resources. Some will be looking to capture higher shares of mobile investment. At the same time, ethical and reputational factors in public and private investment are playing an increasing part in where the smart money flows.
Much of the research, technology, and skill base in high-value industries is tied to the defense industry. That industry can be a catalyst for high-value manufacturing growth in developed and developing economies.
Openness in the world trade system is essential for successful growth. Such trade must be carried out responsibly and securely. A treaty that clearly establishes the highest common standards, coupled with transparency of implementation, can go a long way toward achieving this goal.
Industry would welcome the opportunity to conclude supply contracts with many countries that have relatively low costs and a skilled workforce. At the moment, industry’s ability to do that is limited by licensing restrictions, sometimes involving a number of governments. Industry would welcome an ATT that overcame this problem.
Possible Approaches
Although not new, two approaches have gained increasing currency. Both are welcome.
The first is an ATT “lite” that is relatively comprehensive and specific on the “what” but short on the “how.” This means that the object and purpose of the treaty must be clear for any state that signs and ratifies it but that the treaty must allow states to make national decisions on issues such as how to establish their export control systems and licensing policies. It also suggests that definitions of terms such as “brokering,” “transit,” “transshipment,” “retransfer,” “loans,” “gifts,” and “donations” need not be provided in the treaty text. Each state would be expected to implement the treaty in good faith through appropriate legislation.
Second, there are increasing calls for a treaty to be short, simple, and easy to implement, an approach reflected in the P5 statement but also increasingly echoed by others in the preparatory committee meetings. This suggests that many of the “wish list” items, particularly extensive documentation and record keeping at every stage of a transaction, will be ruled out, as they should be, as overly burdensome and difficult to implement.
Both approaches recognize some practical constraints: the limited time available to negotiate a text, the differing national mechanisms for implementing export controls, the differing security interests of states, and the differing capacities for implementation. No single issue will be the determining factor, but in combination, they represent powerful drivers toward a short but effective treaty text.
Five goals and objectives in the current draft text, listed below, appear to have broad support and should rightly provide the compass for negotiations and the creation of a core treaty text.
Promote the goals and objectives of the UN Charter. The 2008 group of governmental experts, which worked by consensus, said that the principles of the UN Charter should be “at the centre” of efforts to address the international arms trade.[11] Therefore, in practice, a treaty will have to respect the sovereignty of every member state. This points to national decision-making and precludes a supranational body that would be able to question the national decisions of states regarding their arms transfers.
Establish the highest-possible common international standards for the import, export, and transfer of conventional arms. The highest-possible common standards should be exactly that—not the lowest common denominator. Furthermore, they should be a floor, not a ceiling. It is fundamentally important that any treaty not be seen as weakening existing controls. It should raise the bar for some, and states should be free to go beyond if they deem necessary.
Prevent, combat, and eradicate the illicit transfer, production, and brokering of conventional arms and their diversion into the illicit market, including for use in transnational organized crime and terrorism. This issue has caused some polarization of views between states. Some states want to limit the scope of a treaty to illicit trade. This misses one essential point held as important by many other states: raising the bar on national export controls makes it more difficult for the illicit trade to operate, particularly if one of the criteria is risk of diversion. Part of the deal that must surely emerge is a recognition that strong and effective controls on all transfers are the only way to prevent, combat, and eradicate illicit trade.
Contribute to international and regional peace, security, and stability by preventing international transfers of conventional arms that contribute to or facilitate human suffering or violations of international law and obligations.
Member states have questioned whether the criteria for this goal[12] are objective. Where a UN Security Council resolution under Chapter VII of the UN Charter is relevant, by banning arms transfers, there clearly is no question regarding the legal obligation of states to comply. Such resolutions under Chapter VII refer to threats to international peace and security and are legally binding on all states. Any criteria referring to such a resolution would clearly be objective and should say that states “shall” not authorize a transfer of arms in situations that meet one or more of the criteria. Similarly compelling language should also apply to other international, regional, subregional, and national measures under which a state is bound or has chosen to be bound, where these relate to sanctions or embargoes on arms transfers. Whether the treaty will make any reference to regional embargoes is doubtful. Such language would certainly make holding together P5 support very difficult.
The objectiveness of other criteria will cause significant debate, especially where judgments over human rights law or international humanitarian law are required. Leaving aside the many debates over the applicability and interpretation of these important bodies of law, significant states will believe that these judgments will be inherently subjective and will therefore insist on the phrase “should not” or “take into account,” rather than the more restrictive “shall not” when referring to the decision to transfer arms.
Promote transparency and accountability in the import, export, and transfers of conventional arms. There is a laudable and principled call for maximum transparency to support a treaty, but this issue needs to be sliced a number of ways. Transparency of implementation has two parts. National reporting on the implementing arrangements will naturally require a comprehensive, one-time report to the UN on the legislative or administrative measures taken to ratify and implement the treaty in good faith. Subsequent reporting would merely provide updates materially affecting implementation. Through this kind of reporting, in combination with annual meetings of states-parties and regular review conferences, differences in national implementation can be discussed and improvements made. This will create a norm in the area of conventional arms transfers that has never existed, which is a prize worth pursuing in the longer term and not one that can be secured immediately.
Reporting on transfers will require more comprehensive annual reporting and would reflect the ultimate scope of the treaty. In order to avoid “reporting fatigue” and to recognize capacity problems in some states, the UN Register would be the logical vehicle because many states already report to it. If the scope of the treaty goes beyond the current seven categories, as it should, then treaty signatories would simply report to the register through an expanded reporting format. This is surprisingly simple to do in practice. The register could continue as a voluntary transparency and confidence-building instrument for any UN member states that are not parties to the treaty.
Cooperation and assistance will be an important element in the body of the treaty and will most likely generate a reporting requirement. States lacking a national export control system may need assistance in creating one to meet treaty requirements. States willing and able to provide assistance, for example on the establishment of export control licensing and enforcement arrangements, would be able to report this. Reporting would possibly identify both offers of assistance and those requiring it.
Perhaps in recognition of the drive toward a lite treaty, there is little appetite for any large secretariat to be established. A much smaller implementation support unit, like the one for the Biological Weapons Convention, acting as a repository and collating the information for the parties, would be sufficient to handle the reports. One possible additional role for the unit would be to act as a clearinghouse for cooperation and assistance, matching needs and resources.
Conclusion
An ATT would require states to consider their existing obligations under international humanitarian law and international human rights law, but many states appear to have differing views on how those obligations should be interpreted and brought to bear on arms export decisions. The drafting and implementation of such parameters must account for this. Simplicity is the best solution.
There is a balance to be struck between loading the treaty with complex text in an attempt to address every angle and keeping it simple for states to implement in good faith through their national implementing legislation. It should be possible to identify a simple menu of proven options for scope and implementation. Bringing these elements together into a coherent package, in a community of vastly different interests, will be the significant challenge over the remaining 22 days of discussion.
The challenges mentioned above certainly are not a reason to oppose an ATT. They acknowledge some practical and political problems that have defied attempts at consensus since the League of Nations. Now, in a world increasingly interdependent on economic and security issues, these global problems demand global solutions.
Andrew Wood is director of strategic export control for Rolls-Royce plc. He is responsible for the company’s global compliance program and is the British industry representative on the United Kingdom’s arms trade treaty negotiating team. Before joining Rolls-Royce in 2009, he served for 10 years in the Ministry of Defence, where he worked on arms control, nonproliferation, and export control policy. This article reflects a range of perspectives from British industry rather than a single corporate view.
ENDNOTES
1.UN General Assembly Resolution 64/48, A/RES/64/48, January 12, 2010.
2. It must be emphasized that UN member states, not the UN, have taken this work forward. Indeed, the UN as an institution could never make such a move.
3. See Hillary Rodham Clinton, “U.S. Support for the Arms Trade Treaty,” PRN 2009/1022, October 14, 2009 www.state.gov/secretary/rm/2009a/10/130573.htm (press statement).
4. Jeff Abramson, “U.S. Supports Arms Trade Treaty Process,” Arms Control Today, November 2009.
5. “P5 Statement at the 3rd Preparatory Committee on an Arms Trade Treaty,” www.un.org/disarmament/convarms/ATTPrepCom/Documents/Statements-MS/PrepCom3/2011-July-12/2011-July-12-Joint-P5-E.pdf.
6. United Nations, “Report of the United Nations Conference on the Illicit Trade in Small Arms and Light Weapons in All Its Aspects,” A/CONF.192/15, 2001.
7. UN General Assembly, “Towards an Arms Trade Treaty: Establishing Common International Standards for the Import, Export and Transfer of Conventional Arms,” A/63/334, August 26, 2008.
8. The June 2009 resolution “[a]cknowledg[es] also the right of States to regulate internal transfers of arms and national ownership, including through national constitutional protections on private ownership, exclusively within their territory.”
9. UN General Assembly, “Verbatim Records of the Proceedings of the Committee’s Meetings,” A/C.1/64/PV.9-18, November 13, 2009. See Elli Kytömäki, “Promoting Discussion on an Arms Trade Treaty,” 2010, p. 81, http://unidir.org/pdf/activities/pdf18-act431.pdf.
10. “Best Practices for Implementing Intangible Transfer of Technology Controls,” 2006, www.wassenaar.org/publicdocuments/2006/docs/ITT_Best_Practices_for_public_statement_2006.pdf. The Wassenaar Arrangement defines “Technology” as “[s]pecific information necessary for the ‘development,’ ‘production,’ or ‘use’ of a product. The information takes the form of technical data or technical assistance.”
11. UN General Assemby, “Towards an Arms Trade Treaty,” p. 16, para. 27.
12. The full text of this goal reads as follows:
Contribute to international and regional peace, security and stability by preventing international transfers of conventional arms that contribute to or facilitate: human suffering, serious violations of international human rights law and international humanitarian law, violations of United Nations Security Council sanctions and arms embargoes and other international obligations, armed conflict, the displacement of people, transnational organized crime, terrorist acts and thereby undermining peace, reconciliation, safety, security, stability and sustainable social and economic development.
For the full text of the paper, see “Chairman’s Draft Paper,” July 14, 2011, http://controlarms.org/wordpress/wp-content/uploads/2011/08/ChairPaper-14July2011.doc.