Delegation success and policy failure: collective delegation and the search for Iraqi weapons of mass destruction.
In late 2002 and early 2003, the U.S. government was preparing for war and working desperately to convince the other members of the United Nations (UN) Security Council to authorize the use of force against Iraq. According to the U.S. government, Iraq had repeatedly failed to comply with multiple UN Security Council resolutions, especially its commitments to eliminate a Weapons of Mass Destruction (WMD) program that included biological, chemical, and nuclear weapons. In September 2002, President George W. Bush addressed the UN General Assembly and argued that U.S. patience was wearing thin and that the stakes for the institution were high. "The purposes of the United States should not be doubted. The Security Council resolutions will be enforced. The just demands of peace and security will be met--or action will be unavoidable, and a regime that has lost its legitimacy will also lose its power." Bush continued by asking, "Will the United Nations serve the purpose of its founding, or will it be irrelevant?" (1) Although Bush was clearly signaling his willingness to employ unilateral military force, he had both international and domestic political reasons to seek a UN mandate. Such a mandate would catalyze military and financial support from other governments for a U.S.-led military operation, and it would strengthen Bush's domestic support among U.S. voters and members of Congress.
Practically speaking, this meant that Bush needed a new UN Security Council resolution, explicitly authorizing the use of force against Iraq. To realize this goal, the United States would have to persuade the other four permanent members of the Security Council plus at least four of the ten rotating members to vote for a resolution authorizing the use of force. (2) In pursuit of this goal, the U.S. government tried repeatedly to shape the behavior and the language of the United Nations Monitoring, Verification, and Inspection Commission (UNMOVIC) and the International Atomic Energy Agency (IAEA). The UN Security Council had previously granted these two international bodies authority to monitor Iraqi compliance with UN resolutions, (3) to conduct inspections within Iraq, to report their findings back to the UN Security Council, and to provide opinions on whether or not Iraq continued to possess prohibited weapons. (4) The U.S. government knew that in order to get reluctant members to vote in favor of the use of force, it would need UNMOVIC or the IAEA to report that Iraq was in material breach of previous resolutions and that Iraq's WMD program posed a direct threat to international peace and security. Despite its great power (and its great efforts), the U.S. government could do very little to shape the inspectors' behavior or the official reports of these international bodies.
These facts led some analysts to conclude that UNMOVIC and the IAEA were "bad agents," either because they were incompetent, biased against the United States, or corrupt. (5) Consequently, the U.S. government paid significant "sovereignty costs" by delegating authority to agents that could not be controlled or directed ex post. Such conclusions stem from a misspecification of agency models or a misunderstanding of the source of the authority that was delegated to weapons inspectors. A clear elaboration of common agency problems, and the principal-agent (P-A) models that are typically used to analyze such delegations of authority, help us to understand this case. Moreover, the framework developed below should help analysts avoid modeling errors in analogous cases in which groups of states pursue their interests through an international delegation of authority.
This case suggests several lessons for scholars interested in analyzing the legal and political issues raised by international delegation. First, the legal mandates and the formal decision rules of international institutions are often efficacious, as they were in this case, and they have substantively important and predictable consequences for international outcomes. Although a single case study cannot prove that formal rules within all international organizations (IOs) are always efficacious, if they shape outcomes involving the great powers contemplating war, then it should not come as a surprise when they influence behavior and outcomes in trade policy, foreign aid, human rights, or other issues areas that are more highly institutionalized. Scholars should invest time studying IO decision rules and authoritative mandates issued by groups of member states to international bodies. Second, scholars should ensure that their theoretical models fit the empirical questions and political domains they seek to explain. Conceiving of international delegation as a grant of authority by a single state to some international body is often misleading, since most international delegation originates from group decisions within existing institutions. (6) Third, as the Bradley-Kelley typology suggests, different states can pay very different sovereignty costs as a result of delegation to the same international body. Further, previous delegations can structure the political costs and benefits of subsequent decisions. Fourth, when new typologies and analytic categories meet actual cases, scholars can make initial judgments about the utility of the typology for organizing empirical analysis. In the case of UN weapons inspectors in Iraq, the Bradley-Kelley typology illuminates a variety of different outcomes, and raises additional questions that could be explored in this or in other cases.
Perhaps more important than these four lessons for scholars, the case of UN weapons inspectors in Iraq suggests that delegation to international bodies can shape the political costs of foreign-policy choices by powerful states. If generalizable, this finding provides lessons for policymakers and citizens as well. Once authority has been delegated, even if the authority is limited to monitoring compliance and providing information, this prior decision can have profound effects on outcomes in international politics--such as the probability and the costs of war. Finally, both for domestic and international political reasons, there are good reasons to think that democratic states (even powerful ones) will be increasingly reluctant to use military force without the imprimatur of the UN Security Council.
STRATEGIES FOR RESEARCH: PRIVILEGING FORMAL RULES IN IOs
If we want to explain (or judge) the choices of states, the behavior of IOs, or the outcomes within and around international institutions, then a focus on the formal decision rules within IOs and official mandates issued by member states is an underappreciated research strategy. Such a strategy offers several advantages to scholars interested in questions of international delegation. First, it will help us to develop deductively sound and falsifiable hypotheses on the causes and consequences of international delegation. Since informal rules and norms are more difficult to identify ex ante, analysts will disagree on the substance of informal rules purportedly guiding behavior in any given case. Under these conditions, "testing" hypotheses can degenerate into spin control, cherry-picked cases, and literary criticism. When rules are clearly specified ex ante (in official documents, articles of agreements, and constitutions), analysts are constrained to focusing on their effects rather then debating, interpreting, or constructing their content to fit a particular legal theory or normative commitment.
Second, a focus on formal rules enables legal scholars to make judgments-both about the efficacy of international legal commitments and about the legality of specific acts undertaken by governments, IOs, and other international actors. A focus on formal rules does not imply that official policy directives, procedures, and formal measures of voting power within IOs can account for all the empirical patterns of interest. However, there are clear benefits to studying international delegation and formal institutions within a principal-agent framework, and there are some underappreciated costs when attempting to incorporate soft law, informal rules, and norms into social science explanations or into legal interpretations. Fortuitously, the number of formal international institutions is increasing, and the specificity and precision of regulative rules and policy mandates are also increasing. (7) Of course, whether these formal rules are efficacious or epiphenomenal is an empirical question.
Expanding beyond legal and formal rules, to incorporate social norms or nonauthoritative power relationships is incompatible with the meaning of delegation and, thus, the conceptual limits of P-A theory. International organizations (or other agents of national governments) may indeed behave in certain ways because they are responsive to global norms and they may also be responsive to third parties rather than (or in addition to) their principals, who have delegated authority to them. Hence, these other factors may be important in explaining outcomes in international relations; but incorporating them within P-A models leads to flawed theory and concept-stretching. Worse, these amendments to P-A models will lead analysts to claim that their "P-A model" explains more than it really does. In order to specify the utility and limits of P-A models for understanding IO behavior and change, one must deduce specific, observable implications from P-A models and then test them empirically. A focus on legal and formal rules at each stage of any delegation chain will enable such efforts and will provide analysts with a baseline against which to assess various claims about the accountability of agents and the legality of particular actions.
Of course, these claims raise the question: How should scholars study the influence of formal rules and decisionmaking procedures within IOs? One approach assumes that once states have agreed to pooh their sovereignty within some international body, (8) the formal rules governing collective decisions are actually efficacious. (9) This idea is identical to the assumption that many scholars of American and comparative politics make when they study the behavior of legislators within committees or voters at the ballot box. The predictive accuracy of these institutional models will be shaped both by the validity of the models themselves (are they internally logical?) and by the efficacy of the formal decision rules within the institutions under study. The first issue is addressed briefly in this article and at greater length elsewhere. (10)
Of course, the most airtight theoretical deductions will be useless in explaining IO behavior, measuring sovereignty costs, or making legal judgments relevant to the real world if our assumptions about the efficacy of formal rules are overly optimistic. This approach presumes the efficacy of formal institutions and is thus identical to the problem faced by new-institutionalist scholars who study domestic politics. If stuffing the ballot box is a common occurrence, then the predictions of comparativists (11) will suffer. If politicians listen to campaign donors or to narrow special interests and disregard the preferences of voters in their districts, then the predictions of Americanists (12) will be inaccurate. Similarly, if the management and staff of an IO (or any other international agent) are primarily responsive to bribes, narrowly focused nongovernmental organizations (NGOs), or large commercial banks, then P-A models that assume formal rules of representation are effective at holding IOs accountable to legally authorized coalitions of member states will provide inaccurate predictions about IO behavior. (13)
The institutionalist literature from the domestic realm has grown rapidly and has shaped scholars' thinking about politics precisely because these P-A models have been able to account for a wide range of outcomes and behavior. As positive empirical findings pile up, scholars' confidence in the utility of these models has grown. There has been no parallel effort involving large numbers of scholars over an extended period that applies P-A models to the study of international politics. Therefore, one strategy is to take formal rules more seriously than researchers have in the past, deduce observable implications from P-A models, and then test these implications empirically. (14)
Analyzing the case of the Iraqi WMD inspection regime, this article focuses on the formal mandates found in UN Security Council resolutions that create new international bodies (United Nations Special Commission (UNSCOM) and UNMOVIC) (15) or identify existing international bodies (IAEA) as the agents of the UN Security Council. These agents were delegated authority to monitor implementation of various UN resolutions and agreements made between the UN Security Council and the Iraqi government. The original design of these delegation contracts and the selection of particular agents can be explained by the decision rules within the Security Council, the distribution of preferences, and the voting power of member states within the Security Council at time T1 (during deliberations over whether to delegate). Efforts to direct the agent at time T2 (after authority has been delegated) can also be explained by the decision rules, the preferences of member governments, and the voting power of the various members that constitute this collective principal (UN Security Council) to which the agent (UNSCOM and UNMOVIC) is formally accountable.
INTERNATIONAL DELEGATION AND PRINCIPAL AGENT THEORY
The key definitions offered in the introductory article by Bradley and Kelley are narrowly drawn and largely consistent with definitions used by a growing number of political scientists studying international delegation. (16) Bradley and Kelley's definition highlights formal delegations of authority from states to international bodies, from states to other states, from states to private actors, and from international bodies to either states, private actors, or other international bodies. For these reasons, the Bradley-Kelley definition will enable research findings to accumulate and should also enhance communication between political scientists and legal scholars working on these issues. The Bradley and Kelley approach to international delegation differs conceptually from those adopted by the growing P-A literature in political science in only a few ways. Most of these are mainly rhetorical or reveal themselves only during model specification or empirical application.
A. A Plea for Counter-Proliferation ... of Conceptual Terms
New ideas often require new language to describe them and to distinguish them from existing ones. However, renaming existing concepts that can accommodate new ideas actually inhibits the accumulation of knowledge and interdisciplinary research. Scholars of international delegation in economics, political science, and law are now becoming familiar with the language and concepts of P-A theory, so jettisoning the language without challenging the basic paradigm offers little payoff. Bradley and Kelley offer a very clear, and possibly the most comprehensive, discussion of the concept of international delegation currently in print. No other book or article provides as general a discussion of the numerous links in any real or hypothetical chain of delegation. With minor amendments, all of their discussion can be reframed in the language of P-A theory without losing any analytic rigor or reducing the number of research questions that emerge from their discussion. If this claim is true, then one should question the wisdom of inventing new words for old ideas.
In addition to easing communication between different scholarly communities, clarifying concepts surrounding international delegation allows scholars to more accurately identify the empirical referents of "principal" and "agent" in the real world that researchers are studying. (17) This point is illustrated by offering a general definition of delegation, discussing different types of delegation relationships, and distinguishing between single, collective, and multiple principals. These distinctions bear directly upon the subsequent analysis of UNSCOM and UNMOVIC behavior from 1991 to 2003 presented in Part IV of this paper.
B. Conceptual Distinctions and Behavioral Expectations
Delegation occurs when an actor X (or actors XYZ) who is authorized to make a decision or take some action conditionally designates some other actor (or actors) to make that decision or take that action. Notice that this definition implies some preexisting set of laws or rules that establish property rights. Succinctly put, authority implies hierarchy. (18) Typically, in the economics context these are rights of ownership. (19) In the context of domestic politics, a constitutional rule or a statute authorizing specific actors to make particular types of decisions is typically the source of authority. (20) For example, legislators are granted the authority to make laws, whereas courts are granted the authority to adjudicate disputes. In international politics and international law, the institution of sovereignty implies that states are the ultimate locus of decisionmaking authority, and hence the actors that can choose to delegate various types of authority internationally. (21)
However, although such grants can vary in terms of the tasks a designated agent is authorized to perform and even how easy such delegations are to revoke, they must be revocable in principle. Grants of authority must be conditional, or else they are not a delegation of authority, but an abdication of authority. (22) This is the central distinction that Coase and Williamson make when distinguishing between markets and firms in their seminal work on delegation. (23) A market exchange takes place between two vertically equivalent actors. Integration of production within firms to overcome market failure necessarily implies a hierarchy--and thus a delegation of authority from a principal (owner of property right) to an agent (employee). As soon as the owner forfeits all her property rights to an employee, there is no hierarchy, and one is no longer in Williamson's world of delegation under hierarchy, but in Adam Smith's world of free exchange in a market. (24) As legal scholar Dan Sarooshi explains about the first link in any chain of international delegation, "The first definitional element of a delegation is that the State has the competence under the instrument of conferral to revoke the conferral of powers at its own discretion." (25)
When one actor delegates authority to another actor, the former is acting as a principal and the latter becomes her agent. More generally, principals are the actors within a hierarchical relationship in whom authority ultimately rests. (26) Agents are the actors who are hired (and potentially fired) by principals. Agents are conditionally designated to perform tasks in the principal's name and have the requisite authority to do so. By definition then, principals and agents exist in a hierarchical relationship that is defined by a delegation contract. As Moe explains, "The logic of the principal-agent model, therefore, immediately leads us to the theoretical issues at the heart of the contractual paradigm: issues of hierarchical control in the context of information asymmetry and conflict of interest." (27)
Scholars do not gain conceptual clarity or empirical leverage by calling agents "entities," "bodies," "trustees," or any other broader or narrower term. If some other actor has conditionally granted these actors any authority, then they are agents by definition. If they do exactly what the principal wants them to do with their delegated authority, they are still agents. If they do nothing that the principal wants them to do with their delegated authority, they are still agents. Such variation in agent behavior should be explained rather than used as an opportunity to proliferate new names for existing concepts.
Equally important, an agent can be an IO, an NGO, another state, a group of states, or any of the other things that the Bradley and Kelley typology suggests. (28) That previous scholars have used P-A models to describe delegations of authority to IO secretariats, (29) NGOs, (30) international courts, (31) subgroups of states, (32) an individual state, (33) emanations of existing IOs, (34) and even individual bureaucrats, (35) demonstrates that the P-A approach is not analytically limited to the lower, left-hand cell of Bradley and Kelley's Table 1, (36) but instead is flexible enough to handle all the types of international delegation described therein. Although individual research projects using P-A theory may focus on just one of these entities, there is no conceptual reason to associate P-A theory with the study of "third party" agents. After all, the seminal work on "internal delegation" is located squarely within the P-A tradition, (37) and novel applications of internal delegation within the international relations literature also employ P-A logic and language. (38)
Perhaps the perceived need to change the names of agents to something else stems from the fact that some applications of P-A theory are closely associated with empirical claims by these P-A proponents that principals tend to get what they want and agents either respond to principal instructions, or else they are fired. Hence, "agent" has become associated with "good agent that does bidding of principal," even though nothing in the conceptual apparatus of P-A theory leads one to this conclusion. In fact, P-A theory offers plenty of reasons why analysts ought to observe significant variation in agent behavior resulting in varying levels of "sovereignty costs," which is one dependent variable of interest to the contributors to this issue. P-A theory is an ideal framework to explain such variation.
Bradley and Kelley are right to emphasize that there is no delegation if there is no "entity" to which authority has been granted. Bradley and Kelley also emphasize the under-studied legal and political dynamics of cases they call "collective redelegation"--when groups of states pool their sovereignty and subsequently empower their representatives (or a subgroup of state representatives) to delegate authority to some entity. (39) This point may be especially important, since such delegations of authority may be the most common in international relations, (40) and since they are almost certainly the most consequential. In fact, no individual state is typically a principal in its own right with authority to direct, hire, or fire any IO agent. (41) Instead, states are typically members of a collective principal.
C. Common Agency: Collective Principal or Multiple Principals
A delegation relationship can have one or more principals, and a principal can be either an individual or a corporate entity containing more than one individual. Following Kiewiet and McCubbins, a single agent with more than one contract with organizationally distinct principals (42) has a delegation relationship with multiple principals. The U.S. Congress and President are both the principals of any given agent within the U.S. bureaucracy. Yet, neither the U.S. Congress nor the President requires the consent of the other branch to monitor, reward, or sanction that agent. (43) The European Commission is responsible to both the Council of Ministers and the European Parliament. (44) Hence, either of these principals can re-contract with the agent independent of the actions of the other principal. Whereas the Council certainly possesses tools of control that the Parliament does not, the recent use of parliamentary authority to censure and force the resignation of the commissioners en masse demonstrates that this formal authority has political effect within the context of the European Union.
The agency literature has largely overlooked another type of complex principal in which an agent has a single contract with a principal, but the principal happens to be composed of more than one actor. (45) Such actors can be designated a collective principal. (46) The most familiar delegation relationships in politics and government involve a collective principal. Groups of voters delegate to politicians, legislators delegate to party leaders, and nation-states delegate to international organizations. In all these situations, a group of actors comes to a decision among themselves and then the group negotiates a contract with an agent. None of these situations permits any individual member of the collective principal to re-contract with the agent directly. (47) If the group cannot come to a decision a priori, then they cannot change the status quo. This conclusion holds for initial hiring decisions, for proposals to renegotiate the agent's employment contract, or for giving the agent novel authoritative instructions. In all these scenarios, there is a single contract between the agent and his collective principal. The single principal and both types of complex principals are depicted in Figure 1. (48)
[FIGURE 1 OMITTED]
Ironically, although collective principals are much more common in politics, political scientists have focused greater attention on the question of multiple principals. In the field of American politics, there is a vigorous debate about the independent influence of Congress on bureaucratic behavior. This debate has spawned considerable general theoretical work on the problems of agent control that are faced by multiple principals. (49) Despite this growing knowledge of the multiple-principal case, political scientists have often incorrectly characterized individual members of a collective principal as multiple principals in their own right, (50) and, as suggested below, this modeling choice can lead to a variety of errors.
Many scholars implicitly assume either a single or multiple principal (such that each state can unilaterally re-contract) in situations that require a collective decision. This inappropriate modeling choice leads to a confusing set of questions that are frequently posed by critics of P-A models. As Chong and Weller ask, "Is this IO, as an agent, accountable to one, two, or the majority of states or to those with the most political or economic power? Or is it supposed to be accountable to those most affected by its actions?" (51) One clear answer to these questions is to identify principals and agents in terms of the actual legal authority that has been delegated. If scholars allow the formal rules that are written into delegation contracts to guide their operationalization of P-A models, then clear answers to these questions about accountability and the identity of the principal will be apparent. Once these conceptual questions have been clearly answered, then empirical questions can be addressed in a systematic fashion. Without such agreement on these basic theoretical and conceptual issues, accumulation of positive or negative findings will be impossible.
For example, if the North Atlantic Treaty Organization's executive body, the North Atlantic Council (NAC), has delegated some authority to a military commander in Kosovo, then it will require unanimity among the members of the NAC to re-contract; hence, that agent is formally accountable to the NAC as a whole and no individual state can alter the delegation contract unilaterally. (52) If the World Bank Executive Board is doing the delegating, then it will require a majority of shares, which could mean as few as eight of the 181 states represented on the Board. If one were analyzing the Legislative Assembly of the World Health Organization, one would properly identify any simple majority of member states based on the one-state, one-vote rule. The possibilities are numerous and varied; one should not despair, however, since the principle by which one answers the questions posed by Chong and Weller is crystal clear--it depends on the formal lines of authority within the delegation contract. Whether one can explain anything using this approach is an empirical question, but it will certainly allow scholars to deduce falsifiable hypotheses and to answer positive and normative questions about accountability at the international level.
D. Differential Delegation
Not all members of an international organization have delegated the same amount of authority to that organization. Bradley and Kelley offer the interesting example of the International Court of Justice (ICJ), but such differential delegation is also present in IOs (unlike the ICJ) whose management and staff have real authority to spend the resources of member states. For example, changes to the articles of agreement or changes in membership of the institution require a supermajority of eighty-five percent of the voting shares on the boards of the IMF and World Bank executive boards. (53) As a result, the United States is the only member of the collective principal that can unilaterally veto an amendment to the basic rules or to the membership of those organizations (because it is the only member with more than fifteen percent of the voting shares). Every other member requires at least one other member government in a coalition to do so. Although the United States has delegated authority to the World Bank on the issue of aid allocation, it has not delegated authority to the Bank or even pooled its sovereignty on the issue of changing the fundamental rules governing the organization.
The second example of differential delegation offered by Bradley and Kelley is also illustrative. If a state maintains a veto over the authoritative decisions of a collective body of which it is a member, then that state has delegated less authority to the collective body than states that lack a veto. (54) As Bradley and Kelley explain, such states are less likely to pay sovereignty costs associated with such delegations. The obvious example is the UN Security Council, in which the five permanent members can veto any resolution and prevent any (legal) change from the status quo on issues that come before it. Although this example makes any initial delegation of authority comparatively difficult, it also insulates agents that have previously been delegated authority by the Security Council, since any new instructions from the Council to its agent are subject to veto by any one of the five permanent members. Once authority has been granted to an agent of the Security Council, no individual member of the Security Council will be able to easily control that agent ex post. (55) This effect of formal rules has substantial implications for outcomes in international relations that are illustrated in the next Part of this paper.
DELEGATING AUTHORITY TO WEAPONS INSPECTORS IN IRAQ
A. Creating and Hiring Agents: Formal Rules, Marching Orders, and IO Autonomy
At the end of the 1990-1991 Persian Gulf War, the United States and Iraq faced a bargaining problem. Neither country wanted to continue fighting, but a temporary ceasefire threatened to collapse over the issue of Iraqi WMD. Iraq agreed in principle to eliminate its WMD capability in return for a cessation of hostilities and a commitment from coalition forces that they would not penetrate further into Iraqi territory. However, the United States insisted on verifying the destruction of all WMD in Iraq and on inspection of facilities used to create them. Iraq refused to permit U.S. troops on its soil, claiming that such inspections would violate its sovereignty and provide an opportunity for espionage. The United States would not agree to a permanent ceasefire without some assurance that Iraq had actually complied with its commitments. Ultimately, Iraq and the U.S. military commanders, who were negotiating terms, agreed that UN weapons inspectors would verify Iraqi compliance. The delegation of authority to an international body helped to prevent a resumption of hostilities that neither side wanted. (56) This welfare-improving deal assumed, however, that there was an international body available to perform the specific tasks required--an assumption that was only partially justified.
The core elements of the inspection regime were spelled out and codified in UN Security Council Resolution 687 on April 3, 1991. (57) Acting under Chapter VII of the UN Charter, (58) the Security Council ordered Iraq to document and then eliminate its entire stockpile of chemical, biological, and nuclear weapons and their associated production and research programs, as well as all missiles with a range of 150 kilometers or more. The UN Security Council tasked the IAEA to verify the destruction of Iraq's nuclear weapons program, and it created a new organization, UNSCOM, to conduct inspections and verify compliance with the other three elements of the WMD program. (59)
Equally important, the UN Security Council granted the IAEA and UNSCOM the authority to determine "any additional locations" within Iraq that should be subject to onsite inspections. (60) Hence, these IOs had the authority (and they used it) to identify sites in Iraq that were not named by the Iraqi government, by the Security Council, or by any member state of the Security Council. This meant that the inspectors were given substantial autonomy by their principal in terms of where to look and how to conduct inspections. Further, UNSCOM was given the authority to request intelligence information from member states, and since these resolutions were passed under Article VII, all UN member states were legally obligated to provide such assistance. (61)
In order to utilize this intelligence information effectively, UNSCOM itself created an Information Assessment Unit, which was essentially an independent intelligence-analysis shop. This unit was created over the objections of the U.S. representative on the Security Council, but in the absence of support from other Council members, there was little the United States could do to stop it, for altering the delegation contract once it has been authorized required a new voting coalition within the collective principal that included all of the Permanent Five (P5) members. A few years later in 1995, UNSCOM unilaterally interpreted its own mandate once again, this time in the face of opposition by Russia, to create a Concealment Unit designed to discover prohibited activities. (62) Again, since the members of the collective principal could not come to an agreement about the desirability of this policy, they could not alter the decision of their agent. In sum, at the outset of the inspection regime, the Security Council was unified about what to do, and delegated substantial authority to its agents to monitor Iraqi compliance. (63) As individual members or even larger subgroups diverged in their preferences, they were unable to redirect their agents, despite repeated efforts by individual governments. (64) New orders required unanimity among the five permanent members; however, after 1995, unanimity was in short supply.
Of course, it was not UNSCOM's collective principal that objected most strenuously to the manner and location of inspections in the early- to mid-1990s, but rather Iraq's government, which was occasionally successful in persuading individual members of the UN Security Council to pressure inspectors into showing more deference to Iraqi sovereignty. For example, in 1998 both Russia and France questioned Richard Butler, UNSCOM's Executive Chairman, and Scott Ritter, the head of UNSCOM's Concealment Unit. Russia and France claimed that inspectors were being "overly-aggressive" in their tactics and unnecessarily trampling Iraqi sovereignty. (65) But the complaints and demands of individual members of the Security Council were not authoritative instructions, and the UNSCOM inspectors found refuge in their mandate and in previous marching orders. They continued to follow the instructions of their principal--previous winning coalitions within the Security Council. (66) In every instance of Russian and French complaint, UNSCOM refused to moderate its behavior until it received an authoritative instruction from the Security Council as a whole. (67) Such orders could come in the form of new resolutions or in the form of official statements from the Council. But these were few and far between, since the permanent members of the Security Council disagreed on the conjoined issues of whether the inspections regime should be softened and whether economic sanctions should be lifted. Deadlock on the Council allowed Richard Butler and Scott Ritter to appear defiant from the perspective of the French and Russian representatives on the Security Council. In fact, these agents were actively and effectively (68) pursuing the mandate that they had been given back in 1991.
In 1991, the Security Council explicitly linked the suspension of sanctions to the formal certification that Iraq had fully complied with all the provisions of Resolution 687. The resolution explicitly identified UNSCOM and the IAEA as the agents that could certify Iraqi compliance, but the UN Security Council retained the right to lift or alter the sanctions regime that had been in place since September 1990. Therefore, individual members of the Security Council could still block any changes to the sanctions regime, but no individual member had the authority to certify Iraqi compliance after April 1991.
B. Mapping Delegation Contracts and Their Consequences
The formal delegation contracts in this regime were varied and complex. (69) The Security Council maintained direct control and oversight over UNSCOM during its eight-year existence. (70) The Executive Chairman and the entire staff of UNSCOM were appointed directly by the Security Council and were not subject to the standard civil-service rules in place for most UN employees. (71) This exception meant that the staff did not have to reflect the geographic balance of the member states; its hires were not subject to ratification or oversight by the UN General Assembly, nor did the staff report to the UN Secretary-General. Moreover, the staff did not have to sign the standard UN personnel documents forbidding staff members from representing the interests of their home countries, and UNSCOM was not subject to oversight by the Budget Committee of the General Assembly. (72)
After years of Iraqi obstructions and significant success on the part of IO inspectors in verifying the destruction of declared weapon systems and even finding many hidden ones, the politics within the collective principal changed, and there was increasing pressure to relax the sanctions against Iraq. Three issues converged that would alter the inspections regime. First, France and China had strong commercial incentives to see the sanctions brought to a conclusion. Similarly, Russia knew that it could not collect on billions in sovereign debt that Iraq owed while the sanctions were in place. All three of these governments were increasingly interested in declaring the inspections regime a success and in ending the sanctions. (73) Second, for eight years the Iraqis had been complaining that the U.S. government was using the UNSCOM inspections to conceal an espionage program against Iraq. In 1998 and 1999, credible evidence emerged that Iraq was correct and the United States had been sending intelligence officers to Iraq with UNSCOM inspectors, where they planted listening devices that transmitted information back to Langley, Virginia. (74) Third, in the face of continued Iraqi noncooperation with inspectors on the ground, Butler issued a stinging report to the Security Council. The United States and United Kingdom pursued a new resolution sanctioning the use of force, but none of the other permanent members was inclined to approve it. On December 16, 1998, all UN inspectors left Iraq and shortly thereafter the United States and the United Kingdom launched operation Desert Fox, a bombing campaign that punished Iraq for noncompliance, but did little to damage Iraq's purported WMD facilities. In the aftermath, Iraq refused to permit inspectors back into the country, and the Clinton Administration was nowhere near a decision to implement inspections by military force.
By January 1999, the United States and the United Kingdom had no chance of getting either Iraqi cooperation under the old regime or cooperation from the other members of the Security Council to enforce prior resolutions. (75) UNSCOM was effectively dissolved in January 1999 when Butler announced that UNSCOM had flown its last aerial surveillance mission, and the following year saw the inspections regime drift with little new multilateral activity. In this context, the United States and the United Kingdom essentially agreed to trade a loosening of the sanctions regime in return for a renewed commitment by Iraq to allow weapons inspectors back in the country. This agreement produced UN Security Council Resolution 1284 in December 1999, which officially disbanded UNSCOM and created a new organizational emanation, the United Nations Monitoring, Verification, and Inspection Commission (UNMOVIC). (76)
However, Iraq insisted that any renewed cooperation would require protection of its sovereignty. These protections came in the form of staffing changes within the inspection agency. First, the delegation contract became more complex and included a role for the Secretary-General, General Assembly Oversight Committee, and the standard set of UN civil-service guidelines that ensure "neutrality." Second, inspection teams were now required to have diplomats, in addition to technical specialists, as members of their teams. Third, UNMOVIC would be overseen by a College of Commissioners made up of twenty representatives of UN member states. These individuals would be appointed by the Secretary-General in consultation with the Security Council. Fourth, the post of "Deputy Executive Chairman," which had always been filled by an American, was abolished. Fifth, UNMOVIC was now headed by an Executive Chairman who was nominated by the Secretary-General of the UN and subsequently approved by the Security Council. Importantly, the new Chairman would report "through the Secretary-General, to the Council, following consultation with the Commissioners, every three months on the work of UNMOVIC." (77) The first Executive Chairman of UNMOVIC was the former Director of the IAEA, Hans Blix. Like his predecessors, Blix knew that his marching orders came from the Security Council rather than from any of its individual members, and in practice, he was even more conscious of this fact than his predecessors. David Malone concludes his assessment of Blix's tenure thus: "In the crucible of fierce international pressure and media attention from November 2002 to March 2003 he proved committed to his mandate, independent in his views (over time infuriating several camps within the international community)...." (78)
The key differences (and similarities) between UNSCOM and UNMOVIC are illustrated in Figure 2. All the members of the UN (permanent, elected, and nonmembers of the Council) have individually delegated authority to the UN Security Council. Permanent members appear within the Council in italic font and rotating members in smaller fonts; nonmembers are not represented within the Council by a representative from their government. Since all members of the UN have equal representation in the General Assembly, all displayed members have identical solid lines.
[FIGURE 2 OMITTED] (79)
The case of weapons inspections in Iraq focuses attention on the second link in the delegation chain--what Bradley and Kelley refer to as the "collective redelegation" stage. (80) Note that in the case of UNSCOM, the Security Council is the sole principal of its agent. In the UNMOVIC era, the Security Council was also the most important principal, since that body had the authority to create the organization, eliminate it, approve leadership nominees, and issue authoritative instructions to the agent. However, in the case of UNMOVIC, the General Assembly delegates some authority through its agent (the Secretary General) and through its existing budgetary oversight mechanisms and previously passed personnel rules that apply to all UN employees, including those working within UNMOVIC. More importantly, the Secretary-General has some direct control over UNMOVIC through appointment rules, reporting procedures, and other rules that restrict UNMOVIC staff from reporting information to member states outside of authorized channels. Naturally, the Secretary-General is also responsible to the Security Council both through the specific resolution authorizing the Secretary-General's role in UNMOVIC inspections, which could be rewritten assuming unanimity among the five permanent members of the Security Council, but also because the Secretary-General is approved by the UN Security Council and can be reappointed only with a positive vote of the Council. (81) Although the IAEA is not pictured here, its relationship to the UN and its member states looks more like UNMOVIC than UNSCOM. (82) These changes to the formal rules defining the delegation contract between the new agent and its principals had substantial (and predictable) consequences for the behavior of that agent and the outcome that resulted--war.
C. Type of Delegated Authority
The actual authority delegated to UNSCOM, UNMOVIC, and the IAEA was limited in scope to include monitoring and reporting on Iraqi compliance with Security Council resolutions. The organizations were granted the authority to determine and declare whether a state is in compliance, (83) and this authority became very important politically in 2002 and 2003. However, none of the inspecting agents had the authority to "enforce" any resolutions. The Bradley and Kelley typology combines "monitoring and enforcement" into a single analytic category, and while these activities may go together in many cases, they do not seem to go together in this case. (84)
The authority to enforce the resolutions was retained by member states. Some states, such as the United States and the United Kingdom, interpreted Resolution 687 as granting them authority to enforce the resolution if Iraq failed to comply. Other states, such as France and Russia, argued that enforcement authority was retained by the Security Council and required an additional resolution by that body to authorize any use of force in 2003. All of the five permanent members of the Security Council agreed that UNMOVIC and IAEA were the agents with the authority to determine and declare that Iraq had fully complied with Resolution 687. Whereas both agents declared that substantial stockpiles of WMD had been destroyed and that they had not discovered additional banned weapons during their inspections in 2002 and 2003, neither would assert any conclusions that Iraq had fully complied with Resolution 687 nor with subsequent resolutions. (85)
Bradley and Kelley's discussion of "agenda-setting" authority makes clear that none of the IO agents authorized by the Security Council were granted such authority. They certainly did have what Bradley and Kelley call "informal agenda-setting power," since the information that they chose to report (and withhold) had a major influence on "the substantive agenda of an international body." This influence was also clearly "the consequence of other forms of delegation." (86) The only reason the world waited with great anticipation and trepidation for the official reports of Blix and Mohamed El Baradei, the Director General of International Atomic Energy Agency, stemmed from their having been granted authority to monitor and give advice to the UN Security Council. The substance of their claims shaped the voting behavior of states on the Security Council, and it shaped the willingness of other governments and individuals around the world to support or oppose the subsequent U.S.-led war against Iraq. (87)
D. Costs of Delegation
Bradley and Kelley suggest a number of interesting hypotheses on the varying level of costs paid by states that participate in any multilateral international delegation of authority. One of their central claims is that different states can pay different costs as the result of the same episode of international delegation. The case of weapons inspections in Iraq certainly illustrates this point.
The most extreme costs associated with international delegations of authority imply interference with "domestic authority" structures, or situations in which international bodies interfere in "relations between the state and its citizens." (88) UNSCOM, the IAEA, and UNMOVIC did not interfere in the domestic authority structures of any members of the UN Security Council, but they repeatedly intervened in Iraq by traveling to sites within Iraq against the expressed wishes of the Iraqi government, by revealing information that the Iraqi government sought to keep secret, by interviewing Iraqi citizens inside Iraq against the explicit instructions of Iraqi authorities, and, in the case of UNSCOM, by acting outside its legal mandate by engaging in espionage against the state of Iraq. United States Vice President Richard Cheney was certainly correct to characterize this inspection regime as "the most intrusive system of arms control in history." (89) Although it is common to think of the inspections regime as a one-way street of externally imposed policies, Iraq actually "authorized" such behavior by ratifying the UN Charter (thus granting the UN Security Council the right to impose its will under specific circumstances) and by subsequently making a series of explicit agreements about inspections between 1991 and 2002 with the UN Security Council in an effort to avoid war, get economic sanctions lifted, or both.
The permanent members of the UN Security Council also paid varying levels of costs as a result of delegating to weapons inspectors. Throughout the 1990s, Russia and France were severely constrained by the presence of UN inspectors. Both governments sought a relaxation of economic sanctions so that they could export restricted products to Iraq, and both governments sought modifications to restrictions on energy-service contracts. Such changes to prior UN Security Council resolutions would require UNSCOM and IAEA inspectors to certify that Iraq was in compliance with relevant resolutions. That never happened, and both the Russian and French governments continued to pay asymmetrical economic costs relative to the United States, China, and Britain. From 2001 until March 2003, it was the United States that was constrained by prior delegations to UNMOVIC and the IAEA. In addition to influencing the timing of the invasion, the inspectors shaped the coalition politics on the Security Council in ways that infuriated the U.S. government and prevented it from getting a "final" resolution authorizing the use of military force. The failure to get a final resolution meant that the United States was primarily responsible for funding and fighting the war and was stuck with the job of postwar occupation and reconstruction. To quote Colin Powell's version of the "Pottery Barn rule," "You break it, you own it." (90) As of July 2007, Iraq seems to have been an exceedingly expensive piece of pottery. Although the delegation of authority to weapons inspectors did not cause the United States to go to war and pay all these costs, the inspectors' independent behavior made it much more difficult for the United States to assemble the type of multilateral coalition that would share the costs of war as it had in 1990-1991. (91) Hence, the case of UN-sanctioned weapons inspections in Iraq clearly illustrates Bradley and Kelley's point about variation in the costs of delegation paid by different state parties to any delegation contract.
PUBLIC OPINION OF GOING IT ALONE: A SURVEY
Although the Security Council's refusal to authorize the use of military force in early 2003 had little apparent impact on the behavior of the U.S. government or on the opinions of U.S. citizens, the immediate and perhaps long-term implications of the inspection regime and the failure to get a UN mandate are in dispute. Some politicians and analysts conclude that the Iraq case has led to a fundamental shift in the U.S. approach to the use of force--a shift in a unilateral direction. For example, a postwar assessment of the inspections regime by British columnist James Bone argues:
Dr. Blix's recent performance has not only discredited himself but has betrayed the trust of all those many millions around the world who put their faith in the United Nations. Worse, his disingenuousness has guaranteed that the world's sole remaining superpower will never put its security in the hands of a multilateral inspection agency because he has proved it is unreliable. (92)
However, there is growing evidence that the opposite is true--not only is the U.S. government currently contemplating the expanded use of the IAEA and other UN inspectors in separate nonproliferation deals with Iran and North Korea, (93) but new survey evidence suggests the U.S. public is now less inclined to support analogous military interventions if the U.S. government lacks UN authorization.
In October of 2006, researchers at William and Mary designed a survey which was carried out by a polling company (Polimetrix). (94) The survey asked U.S. registered voters to answer twenty-one questions about contemporary foreign-policy issues. Four of the questions were designed to test hypotheses about the domestic political relevance of delegation to the UN Security Council to approve or disapprove the use of military force to inhibit WMD proliferation. These questions were asked for two reasons. First, a number of political scientists have recently theorized about the "legitimation effect" of UN approval for a particular use of military force. (95) Second, in the run-up to the 2003 war in Iraq, a number of surveys asked respondents in different countries whether they would support a war against Iraq with and without a UN Security Council resolution. Canadian and Western European respondents were much more likely to support the use of force if the Security Council approved such action. For example, in Canada support jumped from ten percent to fifty-six percent, in France from seven percent to thirty-four percent, and in Germany from nine percent to forty-eight percent. So, on average, support for war jumped four-fold with a UN resolution, and very few people were willing to support a war without a UN resolution. For U.S. respondents, the level of support was much higher for unilateral military action; it did not increase as much proportionally with a UN resolution, but there was still a substantial gap. (96)
One might speculate that since the Iraq War has been more costly than expected, and since no WMD were discovered in Iraq ex post, Americans might be more affected by a UN resolution now than they were in 2003. Therefore, the TRIP/Polimetrix survey from 2006 asked each U.S. respondent to answer a pair of questions about North Korea and a pair about Iran. The survey provided no information to the respondents other than the questions. In each pair of questions, the survey asked whether the respondent would support military action by the United States and whether she would support military action by the international community after a UN Security Council resolution.
The survey results revealed large gaps between support for unilateral military action and military action approved by the UN Security Council. In brief, support for military force jumped twenty-nine percent for an attack on Iran and thirty-two percent for an attack on North Korea after a UN Security Council resolution. Although the questions are not identical, the gaps are similar in magnitude to the gaps found in Canada and Western Europe in the 2003 Gallup survey prior to the Iraq War. The effect on opposition to the use of force is even more remarkable, especially considering that the United States is currently involved in a very unpopular and costly war. If the UN Security Council were to approve the use of force against Iran, only sixteen percent of Americans said they would oppose it. Only thirteen percent reported that they would oppose the use of such force against North Korea. The results of the survey are displayed in Figures 3 and 4 below. (97)
[FIGURE 3 OMITTED]
[FIGURE 4 OMITTED]
Alex Thompson sums up the relationship between weapons inspectors and their principals in this way: "In these difficult circumstances, UNSCOM sought the truth about Iraq's weapons even more diligently than could have been hoped for; it was a faithful agent." (98) Thompson's conclusions about UNSCOM are accurate and can be generalized to the other inspection agencies that worked for the Security Council in Iraq. Whereas individual member governments of the UN Security Council can now identify behavior on the part of the inspectors that ran counter to their interests, with few exceptions these IO agents pursued their mandates and implemented official marching orders with exceptional energy and competence.
The case of the Iraqi WMD inspections regime demonstrates the importance of distinguishing between agency models. Scholars that incorrectly labeled the delegation to UNSCOM and UNMOVIC as one from multiple principals determined that the inspections regime largely "failed." The inspectors were "bad agents" and the United States, France, and Russia paid significant sovereignty costs by delegating to an agent whom they were unable to control ex post. Efforts by France and Russia to limit discretion of UNSCOM by restricting its ability to inspect undeclared sites failed, allowing inspectors to substantively monitor Iraqi disarmament. Moreover, the United States was unable to bully UNMOVIC to declare Iraq in material breach of Council resolutions in order to justify a war that the international community largely saw as unnecessary. The agents did not have to respond to attempts by individual states to renegotiate the terms of the delegation contract. Characterizing the source of authority for the inspections regime as a collective principal provides the explanation for this lack of responsiveness. From this perspective, inspectors fulfilled their mandates successfully, and the United States and the rest of the P5 benefited from the autonomy and discretion delegated to UNSCOM and UNMOVIC. Inspectors were not influenced by individual states or coerced to pursue objectives that the Security Council did not approve, but rather implemented orders from the Council with competence.
The case also demonstrates the importance of reviewing legal mandates and formal delegation contracts. Scholars who inaccurately described the inspection regime as failing because it did not represent the interests of its principals failed to examine the formal relationship between principal and agent. The case of the inspections regime confirms that analyzing these documents is an important task of researchers studying delegation and P-A relationships. The legal mandates and delegation contracts that establish a P-A relationship matter and shape the behavior of both principals and agents. In fact, UNSCOM, UNMOVIC, and IAEA inspectors did represent the interests of their principal, but scholars incorrectly asserted that individual states were the principals, rather than the collective institution (UN Security Council) that held formal authority over the inspections regime.
Bradley and Kelley argue that delegation can be quite powerful when the actions of these bodies "circumscribe policy autonomy by creating international or domestic pressures on governments." (99) Delegation to UNSCOM and UNMOVIC certainly demonstrates how such IOs can result in "international or domestic pressure on ... governments." (100) As the previous case study demonstrates, this statement captures much of the story behind the collapse of the Iraq inspections regime. Further, such substantial evidence as the William and Mary survey reveals that the UN Security Council is increasingly perceived as a stamp of approval for military action (or disapproval in the absence of a resolution supporting a coercing state), (101) and may be increasingly important to citizens in democratic societies. Hence, the decisions of the UN Security Council may increasingly shape the behavior of political leaders who want to maintain the political support of their people within democratic polities, including the United States.
Hence, this case presents a clear example of delegation success and policy failure. All the major goals established by the Security Council in 1991 that were designed to prevent the resumption of hostilities and to disarm Iraq had been achieved by 2002. Despite this fact, the outcome was an interstate war in Iraq that produced thousands of deaths, racked up billions in debt, produced an ongoing civil war in Iraq, and led to growing doubts about the efficacy of multilateral solutions in international security affairs. This outcome contradicts an implicit assumption of much principal-agent work: that when delegation "succeeds" and the agent accomplishes the assigned task at a cost lower than the principal would pay through direct implementation, policy outcomes will improve and cooperation will be sustained. (102) The case of weapons inspections in Iraq appears to be a case of delegation success and policy failure. The agents did their jobs effectively and efficiently, but cooperation broke down, and war--policy failure--was the result. The costs of that war, fought without the mandate of a resolution explicitly authorizing the use of force, have helped to shape the perception of U.S. voters regarding the utility of unilateral versus multilateral military intervention to prevent WMD proliferation and will likely shape the behavior of U.S. foreign policy in the future.
This Article is also available at http://www.law.duke.edu/journals/lcp.
(1.) John King & Suzanne Malveaux, Bush: U.S. Will Move on Iraq If U.N. Won't, CNN, Sept. 13, 2002, available at http://archives.cnn.com/2002/US/09/12/bush.speech.un/(last visited Feb. 3, 2008).
(2.) UN Charter art. 23, 27. The formal rules for passing resolutions at the UN Security Council specify that a resolution must receive at least nine of the fifteen votes and that none of the five permanent members votes against the resolution.
(3.) See Curtis Bradley & Judith Kelley, The Concept of International Delegation, 71 LAW & CONTEMP. PROBS. 1, 6, fig. 1 (Winter 2008). This article explores a case of "international delegation" under the Bradley and Kelley typology, discussed as "collective redelegation," using UNMOVIC as an example of an "other international body." For a comparable illustration with proper nouns that correspond to the case of UN weapons inspectors, see Figure 2, infra Part IV.
(4.) S.C. Res. 1284, 4084, UN Doc. S/RES/1284 (Dec. 17, 1999). In terms of the Bradley and Kelley typology of delegation, the UN Security Council had delegated the authority to monitor, do research, and give advice. Whereas Bradley and Kelley suggest that "monitoring and enforcement" go together as one of the eight forms of delegation they identify, in this case the IAEA, UNMOVIC, and (previously) UNSCOM only had the authority to monitor. Enforcement of the disarmament resolution was left to the member states.
(5.) BOB WOODWARD, PLAN OF ATTACK 240 (2d ed. 2004). On the view of the U.S. Administration in late 2002, Bob Woodward explains, "The intelligence indicated that Blix was not reporting everything and not doing all the things he maintained he was doing. Some of the Principals believed that Blix was a liar. In any case, it looked like the inspections effort was not sufficiently aggressive, would take months or longer and was likely doomed to fail." For other exemplary criticisms suggesting UNMOVIC and the IAEA were unresponsive to U.S. demands, see David Albright, Persistent or Pushover: Views of Blix's Record Vary, WASH. POST, Dec. 5, 2002, at A01; Gregory Copley, Hans Blix and the Politics of Smug Myopia, DEFENSE AND FOREIGN AFFAIRS STRATEGIC POLICY, Mar. 2004, at 3 ("I will be blunt: Blix is a liar through distortion, omission and misrepresentation. He has shown his political colors, and they are anti-US."); Helena Smith & Ewan MacAskill, As Arms Inspectors Arrive, Row Erupts over US Smears, GUARDIAN, Nov. 19, 2002, at 1; Steven R. Weisman, To White House, Inspector Is Now More a Dead End than a Guidepost, N.Y. TIMES, Mar. 2, 2003, at A13; Byron York, Blix-krieg, NATIONAL REVIEW, Oct. 1, 2002, at 31.
(6.) Mona Lyne et al., Who Delegates: Alternative Models of Principals in Development Aid, in DELEGATION AND AGENCY IN INTERNATIONAL ORGANIZATIONS 44 (Hawkins et al. eds., 2006); Darren Hawkins et al., Delegation Under Anarchy: States, International Organizations, and Principal-Agent Theory, in DELEGATION AND AGENCY IN INTERNATIONAL ORGANIZATIONS 11 (Hawkins et al. eds., 2006).
(7.) See Barbara Koremenos, When, What, and Why Do States Choose to Delegate? 71 LAW & CONTEMP. PROBS. 151 (Winter 2008). See also GOLDSTEIN ET AL., LEGALIZATION AND WORLD POLITICS 1-8 (2001).
(8.) See Bradley & Kelley, supra note 3, at 7.
(9.) See George Downs et al., Is the Good News about Compliance Good News for Cooperation?, 50 INT'L ORG. 379, 380-81 (1996). Although legal scholars have a long tradition of taking treaty language and rules seriously, such a practice is the exception among international relations (IR) scholars, to whom rules are either irrelevant, see John Mearsheimer, The False Promise of International Institutions, 19 INT'L SECURITY 7 (1994), or reflect underlying interests. The concept of "regime" was preferred to "organization" in part because the former approach could accommodate the informal rules and norms that seemed to account for much of the behavior that scholars observed but could not explain as a result of organizational voting power and formal decision rules. See generally KEITH MIDDLEMAS, ORCHESTRATING EUROPE: THE INFORMAL POLITICS OF EUROPEAN UNION 1973-1995 (1995) (arguing that a focus on codified treaties and formal rules within IOs actually obscures more than it reveals and "make[s] it difficult for these theories to offer accurate explanations"). The growth of formal IOs and the concomitant increase in the legalization of world politics provide incentives to revisit the impact of formal rules on international outcomes and state behavior. See John Pevehouse et al., International Governmental Organizations, in THE POLITICS OF GLOBAL GOVERNANCE 9-24 (Paul F. Diehl ed., 2005) (tracing growth of IOs); JUDITH L. GOLDSTEIN ET AL., LEGALIZATION AND WORLD POLITICS (2001) (considering legalization of world politics). Recent empirical work in this tradition suggests that formal rules may do a reasonable job accounting for patterns of outcomes in and around IOs. See, e.g., Mona Lyne et al., Who Delegates: Alternative Models of Principals in Development Aid, in DELEGATION AND AGENCY IN INTERNATIONAL ORGANIZATIONS 41-76 (Hawkins et al. eds., 2006); Daniel Nielson & Michael J. Tierney, Delegation to International Organizations: Agency Theory and World Bank Environmental Reform, 57 INT'L ORG. 241 (2003); Daniel Nielson & Michael J. Tierney, Theory, Data, and Hypothesis Testing: World Bank Environmental Reform Redux, 59 INTERNATIONAL ORGANIZATION 785 (2005); Mark Pollack, Learning from the Americanists (Again): Theory and Method in the Study of Delegation, WEST EUROPEAN POLITICS, Jan. 2002, at 200; David P. Rapkin & Jonathan R. Strand, Reforming the IMF's Weighted Voting System, 29 WORLD ECONOMY 305 (2006); Roger Congleton, Agency Problems and the Allocation of International Environmental Grants: The Return to Rio, 20 ECONOMIA DELLE SCELTE PUBBLICHE 125 (2003).
(10.) See generally Nielson & Tierney, Theory, Data, and Hypothesis Testing: World Bank Environmental Reform Redux, supra note 9; Mona Lyne & Michael J. Tierney, The Politics of Common Agency: Implications for Agent Control with Complex Principals (Aug. 28-31, 2003) (unpublished paper presented at Am. Pol. Sci. Ass'n Meeting).
(11.) See generally GARY W. COX, MAKING VOTES COUNT: STRATEGIC COORDINATION IN THE WORLD'S ELECTORAL SYSTEMS (1997); AREND LIJPHART, PATTERNS OF DEMOCRACY: GOVERNMENT FORMS AND PERFORMANCE IN 36 COUNTRIES (1999); MATTHEW SOBERG SHUGART & JOHN M. CAREY, PRESIDENTS AND ASSEMBLIES: CONSTITUTIONAL DESIGN AND ELECTORAL DYNAMICS (1992).
(12.) See ARTHUR LUPIA & MATHEW D. MCCUBBINS, THE DEMOCRATIC DILEMMA: CAN CITIZENS LEARN WHAT THEY NEED TO KNOW? 4-13 (1998).
(13.) A few positive results reveal little because of problems of observational equivalence, but if large numbers of studies are done within a given research tradition, then such accumulation of positive findings provides greater confidence in the utility of the approach. On accumulation of knowledge, see generally Gary KING ET AL., DESIGNING SOCIAL INQUIRY: SCIENTIFIC INFERENCE IN QUALITATIVE RESEARCH (1994).
(14.) This research strategy is distinct from the predominant approach to framing research in international relations, which has been to set up a "three-cornered fight" in which theory X and theory Y occupy two corners and the evidence occupies the third corner. The two theories make contested conjectures about the same evidence, which is used to choose between them. See Colin Elman & Miriam Elman, Lessons from Lakatos, in PROGRESS IN INTERNATIONAL RELATIONS THEORY: APPRAISING THE FIELD 21-68 (Colin Elman & Miriam Elman eds., 2004) (describing such "three-cornered fights"). This is not the strategy that institutionalists have pursued in American politics, comparative politics, or economics. To paraphrase the editor of a prominent journal in advocating a normal scientific approach:
Stop worrying about the alternative theory so much. Its champions will claim that you have mis-characterized their model no matter how careful you are. Concentrate on making your own argument and then testing it. If lots of people do that for the next ten years, then we will have some sense about whether you are onto something.
Confidential Editorial Comments, provided by a Prominent Social Science Journal Editor, via email exchange (May 2002).
(15.) UNSCOM is the United Nations Special Commission tasked by the Security Council to monitor Iraq's compliance with various UN Security Council resolutions ending the 1990-1991 Gulf War. UNMOVIC is the United Nations Monitoring, Verification, and Inspection Commission established in 1999 to replace UNSCOM. For a clear description and history, see Scan D. Murphy, Efforts to Address Iraqi Compliance with UN Weapons Inspections, 96 AM. J. OF INT'L L. 956 (2002). The International Atomic Energy Agency (IAEA) predated the conflict in Iraq, but due to its technical expertise, it was tasked by the Security Council in 1991 and again in 1999 to monitor Iraqi compliance in the area of nuclear programs.
(16.) See generally BARBARA KOREMENOS ET AL., THE RATIONAL DESIGN OF INTERNATIONAL INSTITUTIONS (2001); MARK POLLACK, THE ENGINES OF EUROPEAN INTEGRATION (2003); Nielson & Tierney, Theory, Data, and Hypothesis Testing: World Bank Environmental Reform Redux, supra note 9; Darren Hawkins et al., Delegation under Anarchy: States, International Organizations, and Principal Agent Theory, in DELEGATION AND AGENCY IN INTERNATIONAL ORGANIZATIONS 3-38 (Hawkins et al. eds., 2006); Alex Thompson, Coercion Through IOs: The Security Council and the Logic of Information Transmission, 60 INT'L ORG. 1 (2006); Michael Lipson, Between Iraq and a Hard Place: U.N. Arms Inspections and the Politics of Security Council Resolution 1441 (Sept. 3, 2006) (unpublished paper presented at APSA Conference).
(17.) In order to develop a general approach to international delegation, our definitions must be broader than "a grant of authority by a state to an international body...." There are two reasons for this, and both are revealed in the cogent discussion by Bradley and Kelley. First, because of the possibility of redelegation, any given delegation contract may not involve a state, but rather come from some international body that is itself acting on the basis of delegated authority. Bradley & Kelley, supra note 3, at 5. Second, and more importantly, most delegations of authority to international bodies or even to other states do not arise from individual decisions by states, but through collective decisions by groups of states that are authorized to delegate, redelegate, or change an existing delegation contract. Bradley & Kelley, supra note 3, at 6-9. In the case study that follows, no individual state had the authority to create UNSCOM, direct it, hire its personnel, or fire its leader. Of course, any discussion of principals and agents could be avoided even while still developing a general framework and using new terms such as "delegator," "delegators," "entity," "delegatee," or "trustee." But, as argued below, any delegation of authority implies "principals" and "agents." Hence, I use these conventional terms in this article.
(18.) For the clearest discussion in the international relations literature, see KENNETH WALTZ, THEORY OF INTERNATIONAL POLITICS (1979). But see the theoretical intuition, developed in OLIVER E. WILLIAMSON, MARKETS AND HIERARCHIES: ANALYSIS AND ANTITRUST IMPLICATIONS (1975), and the formal proofs found in JEAN TIROLE & DREW FUDENBERG, PERFECT BAYESIAN AND SEQUENTIAL EQUILIBRIA: A CLARIFYING NOTE (1988), or any modern textbook of micro- or new-institutional economics.
(19.) You can delegate authority to a plumber to fix your pipes because you own the pipes and thus have the ultimate right to do with them as you please. A plumber who works on your pipes without your approval is called a Good Samaritan, or a vandal. A plumber who works on your pipes after you have authorized him to do so is called an agent.
(20.) In this view the U.S. Constitution is a delegation contract that identifies the ultimate principals, "We the people ..." and their various agents--Congress, President, the courts, and other authoritative political bodies.
(21.) See Bradley & Kelley, supra note 3, at 17 (noting that such authority can be, and often is, redelegated to other bodies, and that the ultimate source of authority stems from the sovereignty of states). In fact, the case study, infra Part IV, illustrates this point clearly: Through the UN Charter, member states of the UN first delegate authority to the Security Council to resolve issues of peace and security, and then the Security Council redelegates parts of that authority to UNSCOM, the IAEA, and UNMOVIC.
(22.) This paper does not suggest that abdications of authority (for example, the shift from the Articles of Confederation to the Constitution in 1789) are impossible empirically, or that one should not explain them: but abdication is qualitatively distinct from delegation and should not be confused with it, especially when such conflation can lead to flawed tests of resulting hypotheses. The distinction also has political weight since politicians and activists often make claims that IOs are increasingly "unaccountable." See generally Ruth Grant & Robert Keohane, Accountability and A buses of Power in World Politics, 99 AM. POL. SCI. REV. 29 (2005); Andrew Moravcsik, Is There a Democratic Deficit in World Politics: A Framework for Analysis, 39 GOV'T AND OPPOSITION 336 (2004). Legal scholars should care about these definitional issues for similar reasons, as Sarooshi explains: "Failure to distinguish between different types of conferrals of powers confuses analysis of the differing legal consequences of these conferrals and obfuscates the domestic policy debates that surround their conferral." DAN SAROOSHI, INTERNATIONAL ORGANIZATIONS AND THEIR EXERCISE OF SOVEREIGN POWERS 1 (2005).
(23.) Ronald H. Coase, The Nature of the Firm, 4 ECONOMICA 387, 387-89 (1937).
(24.) All the major works in political science that attempt to explain the causes or consequences of delegating authority employ a similar logic borrowed from Williamson, and they all use the well-established terms of "principal" and "agent" to describe the strategic actors in a delegation relationship. As Hawkins et al. explain, principals and agents are constituted by the act of delegation. See DARREN HAWKINS ET AL., DELEGATION AND AGENCY IN INTERNATIONAL ORGANIZATIONS 7 (2006). If you have delegation, then you have principals and agents by definition. See generally DAVID EPSTEIN & SHARYN O'HALLORAN, DELEGATING POWERS: A TRANSACTION COST POLITICS APPROACH TO POLICY MAKING UNDER SEPARATE POWERS (1999); JOHN HUBER & CHARLES SHIPAN, DELIBERATE DISCRETION: THE INSTITUTIONAL FOUNDATIONS OF BUREAUCRATIC AUTONOMY (2002); RODERICK KIEWIET & MATHEW D. MCCUBBINS, THE LOGIC OF DELEGATION: CONGRESSIONAL PARTIES AND THE APPROPRIATIONS PROCESS (1991); Terry Moe, The New Economics Organization, 28 AM. J. OF POL. SCI. 739 (1984); Pollack, supra note 9 (suggesting scholars should be "learning from the Americanists, again"). The take-home message from Pollack is that scholars should not reinvent the wheel in studies of delegation within the context of the European Union.
(25.) SAROOSHI, supra note 22, at 55. See especially chapters 3-5, in which Sarooshi develops a typology of conferrals of sovereign power to IOs. Like Bradley & Kelley, supra note 3, he includes the legal possibility of "full transfers" of sovereign authority, but he argues that these are distinct from "delegations" of authority, which are conditional by definition. See SAROOSHI, supra note 22, at 29-30 (noting that, even in the case of "full transfers," "in practice ... it would seem that conferrals of powers are always revocable so long as the State has retained its independent legal personality and not merged its powers and personality into a larger political unit").
(26.) This definition follows from Torbjorn Bergman et al., Introduction: Parliamentary Democracy and the Chain of Delegation, 37 EUR. J. OF POL. RES. 255, 257 (2000): Kiewiet & McCubbins, supra note 24, at 4-5.
(27.) Moe, supra note 24, at 757.
(28.) Although this approach is consistent with Hawkins, supra note 16, at 11, Bradley and Kelley provide a much more comprehensive typology and discussion of the various forms of international delegation--and thus the range of actual strategies that are available to states. Bradley & Kelley, supra note 3, at 10-17.
(29.) E.g., Congleton, supra note 9; Helen Milner, Why Multilateralism? Foreign Aid and Domestic Principal-Agent Problems, in DELEGATION AND AGENCY IN INTERNATIONAL ORGANIZATIONS 107 (Hawkins et al. eds., 2006); Mark Copelovitch, Master or Servant: Agency Slack and the Politics of IMF Lending (Nov. 17-18, 2006) (unpublished paper presented at IPES Meeting).
(30.) See generally BERTIN MARTENS, THE INSTITUTIONAL ECONOMICS OF FOREIGN AID (2002); Alexander Cooley & James Ron, The NGO Scramble: Organizational Insecurity and the Political Economy of Transnational Action, 27 INT'L SECURITY 5 (2002); Steve Radelet, A Primer on Foreign Aid (Center for Global Development Working Paper, No. 92, 2006).
(31.) See generally Darrel Hawkins & Wade Jacoby, How Agents Matter, in DELEGATION AND AGENCY IN INTERNATIONAL ORGANIZATIONS 199 (Hawkins et al. eds., 2006); Geoffrey Garrett, The Politics of Legal Integration in the European Union, 49 INT'L ORG. 171 (1995); Geoffrey Garrett & George Tsebelis, Agenda Setting Power, Power Indices, and Decision Making in the European Union, 16 INT'L REV. OF L. AND ECON. 345 (1996); Pollack, supra note 14, at 165.
(32.) See generally Alex Thompson, Screening Power: International Organizations as Informative Agents, in DELEGATION AND AGENCY IN INTERNATIONAL ORGANIZATIONS 229 (Hawkins et al. eds., 2006); Alex Thompson, Coercion Through IOs: The Security Council and the Logic of Information Transmission, 61 INT'L ORG. 1 (2006); Lyne et al., supra note 9.
(33.) See Lyne & Tierney, supra note 10, at 6 for the example of the Concert of Europe, in which a group of states (great powers in nineteenth-century Europe) collectively granted authority to one of their members (France) to put down a rebellion in Naples. Political scientists are not alone in recognizing that a single state can be construed as an agent of another state under international law. As former president of the IJC, Eduardo Jimenez de Arechaga, explains, "The 1923 Treaty between Switzerland and Liechtenstein ... seemed to constitute a case of agency, in which one State entrusted another with the power to represent it not only for the purpose of concluding certain treaties, but also for the purpose of claiming rights under those treaties." Sarooshi, supra note 22, at 33 (illustrating the point that "principal" and "agent" have a specific meaning in international law and that meaning is consistent with the conventional use of these terms in political science and economics).
(34.) See generally Lipson, supra note 16.
(35.) Leslie Johns, A Servant of Two Masters: Communication and the Selection of International Bureaucrats, 61 INT'L ORG. 245 (2007).
(36.) See, e.g., Bradley & Kelley, supra note 3, at 9.
(37.) Kiewiet & McCubbins, supra note 24; KEITH KREHBIEL, INFORMATION AND LEGISLATIVE ORGANIZATION (1991); Thomas Gilligan & Keith Krehbiel, Specialization Decisions Within a Committee, 13 J. L. ECON. & ORG. 366 (1989).
(38.) Thompson, supra note 32, at 229.
(39.) Often this intermediary takes the form of an executive board that contains representatives from some or all of the member states. For a graphic illustration of such a delegation within the Asian Development Bank, see Lyne et al., supra note 9, at 45, Fig. 1.
(40.) See discussion infra Part IV for analysis of one example. For a broader discussion of the frequency of such delegation in international relations, see Koremenos, supra note 7.
(41.) Once a state is a member of a collective body, council, or board, it can certainly quit and thus remove itself from the membership of a collective principal, but it cannot act individually to fire or provide authoritative instructions (redelegation) to the ultimate agent.
(42.) See supra note 24.
(43.) See Randall Calvert et al., A Theory of Political Control and Agency Discretion, 33 AM. J. OF POL. SCI. 588 (1989); Epstein & O'Halloran, supra note 24; Thomas Hammond & Jack Knott, Who Controls the Bureaucracy? Presidential Power, Congressional Dominance, Legal Constraints, and Bureaucratic Autonomy in a Model of Multi-Institutional Policy-Making, 12 J. L. ECON. & ORG. 119 (1996). Although the control of bureaucratic agents may be enhanced by inter-branch cooperation, it is simply not the case that agent recontracting requires cooperation between the President and Congress. Both of these principals have their own contract with the bureaucratic agent, and both have the authority to alter that contract unilaterally.
(44.) Pollack, supra note 16, at 75-154.
(45.) See, e.g., Bradley & Kelley, supra note 3, at 7 (discussing delegation to a "collective body or subgroup," as illustrated in Fig. 1).
(46.) Kiewiet & McCubbins, supra note 24.
(47.) David Lake, Delegating Divisible Sovereignty: Some Conceptual Issues 7 (Mar. 3-4, 2006) (unpublished paper for Workshop on Delegating Sovereignty), offers a telling example when he explains, "As a liberal democrat in a heavily republican Congressional district, I am typically very unhappy with the way my representative votes on legislation, but it would be inappropriate to say that he is shirking his responsibility when he follows the wishes of the majority of my neighbors." Similarly, and obviously, Lake cannot independently vote his representative out of office. Therefore, he is not a principal of his representative.
(48.) Note that most of the literature refers to the first type of principal as a "single principal." To avoid confusion, this article uses the same terminology. However, a more accurate description would be "single-unitary principal." Strictly speaking a "collective principal" is a single principal composed of more than one individual--its corporate nature is the distinguishing feature.
(49.) This extensive literature is reviewed in Hammond & Knott, supra note 43.
(50.) Fearon and Lohman both characterize voters as multiple principals of elected officials. James D. Fearon, Electoral Accountability and the Control of Politicians: Selecting Good Types Versus Sanctioning Poor Performance, in DEMOCRACY, ACCOUNTABILITY, AND REPRESENTATION 55-97 (Adam Przeworksi et al. eds., 1999); Suzanne Lohmann, An Information Rationale for the Power of Special Interests, 92 AM. POL. SCI. REV. 809 (1998). For a more recent example focusing on IOs, see the modeling assumptions in Johns, supra note 35, at 248-58.
(51.) See Xu-yi Chong & Patrick Weller, International Civil Servants: The Forgotten Power 6 (2004) (unpublished paper presented at the 2004 meeting of the SGIR's Sixth Pan-European International Relations Conference); see also Grant & Keohane, supra note 22, at 29-30.
(52.) On the Consensus Rule in NATO, see generally Michael Leo, NATO Decision-Making: Au Revoir to the Consensus Rule?, 202 STRATEGIC FORUM 1 (2003). For complications in the context of the intervention in Kosovo, see generally David Auerswald et al., Double Hats, Triple the Trouble: Agency Problems in NATO Interventions (Sept. 2, 2006) (unpublished paper delivered at the American Political Science Association Conference).
(53.) Leo Van Houtven, Governance of the IMF: Decision Making, Institutional Oversight, Transparency, and Accountability, IMF PAMPHLET SERIES 53, Aug. 12, 2002, at 1, 73-74.
(54.) Moravcsik, supra note 23, at 336 concurs and explains the concept clearly: "Sovereignty is delegated when supranational actors are permitted to take certain autonomous decisions, without an intervening interstate vote or unilateral veto."
(55.) See, e.g., Karen J. Alter, Delegation to International Courts and the Limits of Re-Contracting Political Power, in DELEGATION AND AGENCY IN INTERNATIONAL ORGANIZATIONS 312-38 (Hawkins et al. eds., 2006); Andrew Cortell & Susan Peterson, Dutiful Agents, Rogue Actors, or Both? Staffing, Voting Rules, and Slack in the WHO and WTO, in DELEGATION AND AGENCY IN INTERNATIONAL ORGANIZATIONS 255-80 (Hawkins et al. eds., 2006). Both make similar points about the political effects of international decision rules in other empirical contexts.
(56.) COLIN POWELL, MY AMERICAN JOURNEY 507-42 (1996); NORMAN SCHWARTZKOPF, IT DOESN'T TAKE A HERO 451-72 (1993).
(57.) S.C. Res. 687, UN Doc S/RES/687 (Apr. 3, 1991). Resolution 687 stated that "Iraq shall unconditionally accept the destruction, removal, or rendering harmless" of all named categories of weapons and their related research and production facilities. The regime was modified in August 1991 by Resolution 707, which gave inspectors the authority to fly unfettered throughout Iraqi airspace and required Iraq to provide assistance in this regard.
(58.) See UN Charter art. 42 (making the resolution binding on all UN member states and implying that military force could be used to enforce the resolution if Iraq did not comply).
(59.) Whereas the IAEA was a large standing organization with its own employees and budget, UNSCOM was created anew for the purpose of monitoring Iraqi compliance. The total number of staff working for UNSCOM at any given time ranged between 100 and 120.
(60.) Id. [paragraph] 8-10, 12-13.
(61.) Usually, this meant intelligence and analysis from the United States or the United Kingdom, but UNSCOM also asked for and received raw intelligence information other UN members, including France, Kuwait, Saudi Arabia, and Israel.
(62.) See generally DAVID M. MALONE, THE INTERNATIONAL STRUGGLE OVER IRAQ: POLITICS IN THE U.N. SECURITY COUNCIL 1980-2005 (2006) (see especially chapter 6); SCOTT RITTER, ENDGAME: SOLVING THE IRAQ PROBLEM--ONCE AND FOR ALL (1999).
(63.) G.S. Pearson, THE UNSCOM SAGA: CHEMICAL AND BIOLOGICAL WEAPONS NONPROLIFERATION 28-32 (1999); Alex Thompson, Principal Problems: The Rise and Fall of U.N. Weapons Inspections in Iraq (Mar. 22-25, 2006) (unpublished paper presented at the International Studies Association meeting).
(64.) In an excellent analysis of the Iraq inspections regime using a P-A framework and formal methods, Johns, supra note 35, at 269, adopts conventional concepts from a multiple principals model to analyze the case. She repeatedly refers to the members of the Security Council as if they were principals in their own right. For example, "Blix's reports during this time period often induced intense conflict amongst his political principals." As the collective-principal model suggests, as UN rules clearly state, and as the case study illustrates, Blix had only one principal--an authorized supermajority on the Security Council. This made changing the status quo very difficult, and it gave Blix and El Baradei great discretion.
(65.) Ritter, supra note 62, at 194; Thompson, supra note 63, at 8-18.
(66.) As always, a winning coalition on the UN Security Council requires a supermajority plus the acquiescence of all five of the permanent members. In order to redirect the inspectors, Russia and France would, at minimum, need to convince the U.S. and Britain to go along.
(67.) JEAN E. KRASNO & JAMES S. SUTTERLIN, THE UNITED NATIONS AND IRAQ: DEFANGING THE VIPER 41-77 (2003); Pearson, supra note 63, at 41; Ritter, supra note 62, at 194-95.
(68.) See generally Charles Duelfer, Comprehensive Report of the Special Advisor to the DCI on Iraq's WMD (2004) (unpublished report delivered to the CIA). Researchers now know that the inspection regime was extremely effective, since the U.S. military has subsequently occupied the entire country of Iraq and the Director of Central Intelligence tasked the Iraq Survey Group (ISG) to conduct a detailed investigation into Iraqi WMD programs. The ISG spent more money in one year than UNSCOM, UNMOVIC, and the IAEA had spent in twelve years combined; employed thousands of personnel compared to hundreds of IO inspectors; and enjoyed unfettered access to all sites within Iraq. Yet it found no evidence that Iraq had any WMD at the time of the U.S. invasion. Hence, it is reasonable to conclude that the UN weapons inspectors were highly effective at their job. Iraq had been completely disarmed of WMD by the spring of 2003 and probably much earlier.
(69.) See Fig. 2, infra Part IV.B.
(70.) See S.C. Res 715, UN Doc S/RES/715 (Oct. 11, 1991). These lines of authority were strengthened and clarified in Resolution 715 in October 1991 when UNSCOM was named as a "subsidiary organ of the Security Council."
(71.) See Lipson, supra note 16, at 11. The original Executive Chairman of UNSCOM was Rolf Ekeus, a Swede who served from 1991 to 1996. Ekeus was followed by Richard Butler, an Australian diplomat who served from 1997 until UNSCOM ceased its operations in January 1999.
(72.) See Cortell & Peterson, supra note 55, at 255-80, for a discussion of how staffing rules within IOs affect the independence of different IO agents. As Krasno & Sutterlin, supra note 67, at 25, explain, "UNSCOM was structured in such a way as to preclude any possibility of General Assembly involvement, even in its financing." This contract was designed to ensure that UNSCOM staff would be insulated from any demands other than those issued by an authorized supermajority coalition on the UN Security Council (all five permanent members plus at least four elected members). For a similar argument about how principals at time T1 design contracts to insulate their agents from political meddling by others at time T2, see David E. Lewis, The Adverse Consequences of the Politics of Agency Design for Presidential Management in the United States, 34 BRIT. J. POL. SCI. 377, 395-402 (2004); Nielson & Tierney, Delegation to International Organizations, supra note 9.
(73.) As one UN official explained in November 1998, "You can never have 100 percent proof of disarmament ... so at some point technical exercise gives way to political judgment.... At some point it becomes impossible to prove the negative." Barton Gellman, U.N. Team Downcast About Iraq Mission; Inspectors Note Eroded Authority, WASH. POST, Nov. 22, 1998, at A01.
(74.) See, e.g., Amin Saikal, Iraq, UNSCOM and the US: A UN Debacle?, 53 AUSTL. J. INT'L AFF. 283, 287 (1999); Ritter, supra note 62, at 143-44. See also an official rebuke in a letter from Kofi Anan to the President of the Security Council, S.C. Res 1172, UN Doc S/1998/1172 (Dec. 18, 1998). Ironically, this example of one member of the UN Security Council's using clandestine means to infiltrate a UN body and perform intelligence operations not authorized by the UN Security Council reveals the limit of any approach that focuses exclusively on formal lines of authority. Of course, once this activity was revealed, the collective principal fired the agent and created an entirely new one (UNMOVIC) with checks and balances designed to prevent this type of agency slack.
(75.) The U.S. government had now confirmed in official policy what Iraq had been claiming for years--that no amount of compliance would convince Washington to vote to lift the sanctions. In late 1998, Secretary of State Madeline Albright suggested that sanctions might continue indefinitely. Later that year, the U.S. Congress passed a resolution calling for regime change in Iraq. Any incentive for cooperation from Saddam Hussein was then gone, and UNSCOM inspectors knew it.
(76.) S/RES/1284, supra note 4.
(78.) Malone, supra note 62, at 167.
(79.) Figure 2 is very similar to Bradley & Kelley, supra note 3, at 6, except that Figure 2 has proper nouns inserted in place of the general concepts represented by Bradley and Kelley. Lines with arrows represent delegations of authority. Consistent with Bradley and Kelley's conceptions of international delegation at Stage 1, all individual states that have signed and ratified the UN Charter have conferred on the Security Council the "primary responsibility for the maintenance of peace and security." They also agree that in doing so, the UN Security Council "acts on their behalf." See UN Charter art. 42, supra note 58.
(80.) Supra note 3, at 6.
(81.) As always, nine of fifteen, with the caveat of unanimity among the P5.
(82.) For a detailed comparison and a compelling argument about the role of organizational culture within different IOs, see generally Lipson, supra note 16.
(83.) Robert L. Brown, Non-Proliferation Through Delegation, Ch. 7 (2007) (unpublished Ph.D. Dissertation, Univ. of Cal.-San Diego) (on file with author).
(84.) Bradley and Kelley have five ordinally ranked categories displayed along a single dimension, suggesting that some types of delegation are associated with higher sovereignty costs than others. Of the nine distinct types of delegation, the inspectors appear to have been granted two types: "monitoring" and "research and advice." Supra note 3, at 21.
(85.) HANS BLIX, DISARMING IRAQ 241-50 (2004).
(86.) Bradley & Kelley, supra note 3, at 14-15.
(87.) See William Horsley, Polls Find Europeans Oppose Iraq War, BBC, Feb. 11, 2003, http://news.bbc.co.uk/2/hi/europe/2747175.stm (last visited Feb. 9, 2008); Gallup Int'l, Iraq Poll 2003, http://www.gallup-international.com (follow "Survey Archive" hyperlink; then follow "Survey Overview" hyperlink; then follow Iraq Poll 2003 hyperlink) (last visited Feb. 3, 2008); TRIP survey data reported infra Part VI. For extensive results from the TRIP survey, see Maliniak et al., The View from the Ivory Tower: TRIP Survey of International Relations Faculty in the United States and Canada, Program on the Theory and Practice of International Relations (2007). For analysis of the earlier survey data, see Alex Thompson, Understanding IO Legitimation, in THE POLITICS OF INTERNATIONAL ORGANIZATIONS: BRIDGING THE RATIONALIST-CONSTRUCTIVIST DIVIDE (Tierney & Weaver eds., 2008).
(88.) Bradley & Kelley, supra note 3, at 30.
(89.) Richard Cheney, Vice President Honors Veterans of Korean War, WHITE HOUSE OFFICE OF THE PRESS SECRETARY (Aug. 29, 2002), http://www.whitehouse.gov/news/releases/2002/08/200208295.html (last visited Feb. 3. 2008).
(90.) BOB WOODWARD, PLAN OF ATTACK 125 (2004).
(91.) See generally ANDREW BENNET ET AL., FRIENDS IN NEED: BURDEN SHARING IN THE PERSIAN GULF WAR (1997).
(92.) James Bone, Blix Should Turn the 'Smoking Gun' On His Own Head, LONDON TIMES, Mar. 12, 2003, available at http://www.timesonline.co.uk/tol/news/world/iraq/article1118964.ece (last visited Feb. 10, 2008).
(93.) Rebecca Johnson, Looking Towards 2010: What Does the Nonproliferation Regime Need?, 84 DISARMAMENT & DIPL. (2007), available at http://www.acronym.org.uk/dd/dd84/84npt.htm (last visited Feb. 3, 2008).
(94.) For more on the survey and the broader project, see Daniel Maliniak et al., Inside the Ivory Tower, FOR. POL'Y 62 (Mar./Apr. 2007); Maliniak et al., The View from the Ivory Tower: TRIP Survey of International Relations Faculty in the United States and Canada, Program on the Theory and Practice of International Relations (2007).
(95.) See generally Ian Hurd, Legitimacy and Authority in International Politics, 53 INT'L ORG. 379 (1999); Thompson, supra note 32; Erik Voeten, The Political Origins of the UN Security Council's Ability to Legitimize the Use of Force, 59 INT'L ORG. 527 (2005).
(96.) Gallup Int'l Iraq Poll 2003, http://www.gallup-international.com/download/GIA%20press%20 release%20Iraq%20Survey%202003.pdf; Thompson, infra note 101.
(97.) The first question was, "If Iran continues to produce material that can be used to develop nuclear weapons, would you support or oppose the U.S. taking military action against Iran?" Answer options were "Support, Oppose, and Unsure." The second question was, "If Iran continues to produce material that can be used to develop nuclear weapons and the U.N. Security Council votes to use military force against Iran, would you support or oppose the international community taking military action against Iran?" Again, answer options included "Support, Oppose, and Unsure." The survey included a similar pair of questions about North Korea. Maliniak et al., supra note 93, at 44.
(98.) Thompson, supra note 63, at 24.
(99.) Bradley & Kelley, supra note 3, at 18.
(100.) Id. at 33.
(101.) See generally Alex Thompson, Understanding IO Legitimation (Mar. 5, 2005) (unpublished paper presented at the annual meeting of the International Studies Association).
(102.) Arthur Lupia & Matthew D. McCubbins, Representation or Abdication? How Citizens Use Institutions to Help Delegation Succeed, 37 EURO. J. POL. RES. 291, 299-301 (2000).
MICHAEL J. TIERNEY, Assistant Professor of Government, College of William and Mary.
|Printer friendly Cite/link Email Feedback|
|Author:||Tierney, Michael J.|
|Publication:||Law and Contemporary Problems|
|Date:||Mar 22, 2008|
|Previous Article:||Negotiate or litigate? Effects of WTO judicial delegation on U.S. trade politics.|