Contesting sovereignty: informal governance and the battle over military expenditure at the IMF.

Clegg, Liam

There is a battle over military expenditure at the International Monetary Fund, with consistent pressure from its most powerful member for the Fund to get tough on military expenditure being pitted against lower-order states' invocation of the organization's sovereignty-protecting rules and practices. While the formal victory of the lower-order states has been codified in the Fund's relatively weak Guidelines on Military Expenditure, on a case-by-case basis policy shifts continue to be imposed on borrowers through the application of informal influence by the US Executive Director in the IMF boardroom. By integrating insights from literature exploring the tension between formal rules and informal practices in international organizations, this case study extends the understanding offered in the international relations literature of the foundations of sovereign inequality in international politics. Keywords: conditionality, IMF, military spending, sovereignty, transparency, US Congress.

**********

DESPITE ITS FOUNDATIONAL IMPORTANCE TO THE DISCIPLINE OF INTERNATIONAL relations, there is little agreement over the theory and practice of state sovereignty. Here, through an exploration of the International Monetary Fund (IMF) engagement with military expenditure, I analyze the intersection of international organizations (IOs) and state sovereignty. This study hooks in to the work of Randall Stone and others on the politics of informal global governance, (1) which highlights the need for scholarly attention to be maintained on the full range of mechanisms through which powerful states are able to retain privileged access to power and control in IOs. While the IMF's formal rules served to permit nondisclosure of data on military expenditure on the grounds of protecting state sovereignty, pressure from the United States has over recent years served to compel some members toward enhanced transparency in this area. In this case, governance through informal processes has generated sovereign inequality at the IMF. (2)

While many facets of (global) military expenditure have been subject to analysis in previous scholarship, (3) interventions from the IMF into the issue remain underexplored. The overall finding that the preferences of materially weak states were, at the end of the day, trumped by the preferences of the leading power will confirm many scholars' assumptions about the nature of international politics. What is perhaps more surprising is the willingness of US authorities to contravene formal rules in cases involving minimal strategic interests, (4) and in a context with a relatively low level of support from among other state representatives for this action. Threats to withhold support for loan arrangements unless military spending data was released were deployed against a series of low-income states, and without the existence of an active coalition inside the organization's Executive Board. (5) Given the extreme power asymmetries involved, targeted states were willing to rapidly forgo the previously fought-for protections of national sovereignty.

In developing this analysis of the intersection of sovereignty and military expenditure at the IMF, I begin the article by outlining the conceptual contribution at the heart of the work. The case study findings contribute primarily to ongoing efforts in the international relations literature to explore the foundations of sovereign inequality, in particular by highlighting the manner in which informal influence has been used to challenge the formally established boundaries of nonintervention at the IMF. Next, I review the evolution of the formal Guidelines on Military Expenditure at the IMF, which were framed in accordance with institutional rules that placed significant restrictions on the extent of intervention in members' domestic affairs. Then, I provide an overview of developments through which US domestic political action has served to push the US executive Directors of the IMF to use their voice and vote to unilaterally secure behavioral change on the part of individual borrowing states. Through this action, informal practices by the US representative have served to create differential experiences of sovereignty among IMF member states. I conclude the article by briefly reprising key insights generated into informal governance and sovereign inequality.

International Organizations and the Foundations of Sovereign Inequality

Sovereignty remains a central focus of international relations literature. A consequence of the growth of this literature has been a proliferation of competing definitions of sovereignty, and of contrasting judgements over the extent to which state actors retain sovereignty in various spheres of operation. After introducing the conceptualization that captures core elements of the relationships at the heart of this case study, I explore competing interpretations in the existing literature on the role of IOs in (re-)producing sovereign inequality. The roots of these disagreements can, I suggest, be clarified by differentiating between formal rules and informal practices in the world of IOs. By doing so, the foundations are revealed of the contradictory role played by the IMF in relation to state sovereignty and military expenditure.

The contributions of Stephen Krasner to the international relations literature on sovereignty provide a valuable point of entry into the investigation of the role of IOs in the (reproduction of sovereign inequality. Krasner's four-fold typology identifies the domestic, interdependence, international legal, and Westphalian pillars of sovereignty. (6) These pillars refer respectively to the organization of authority in a given state, the capacity of public authorities to control transborder flows, the formal recognition within key regimes, and the extent to which external actors exercise authority over domestic issues. With its implications over the extent to which an external actor, in the form of the IMF, is able to exercise authority over domestic military expenditure, my case study relates most closely to the Westphalian pillar. Drawing on Krasner's definition, in the discussion below I take sovereign inequality to refer to variations in the extent to which a given external actor (the IMF) is able to exercise authority over a given domestic issue (military expenditure monitoring) across different state spaces.

For Krasner, the nature of sovereignty is contextually determined, with state actors' experiences fluctuating widely across the international system. (7) The different dimensions of sovereignty are subject to independent variation; states can, for example, gain formal recognition while ceding authority in a given issue area to an external agent. Krasner reminds us that throughout the sweep of contemporary history "only a very few states have possessed ... the bundle of properties associated with sovereignty." (8) In place of the rarely realised condition of full sovereignty, levels of autonomy wax and wane across time and issue areas at differing rates for differing state actors. Sovereign inequality is, for Krasner, an ever present feature of the system. And in line with Krasner's broad observation, a vibrant strand of literature has emerged seeking to explore the mechanisms through which sovereign inequality is layered into the international system.

The existing literature suggests that IOs display somewhat Janus-faced characteristics in their interactions with the practice of state sovereignty. On the one hand, international institutions have been found to help reinforce the boundaries of nonintervention surrounding a given policy area. The central channel through which international institutional structures serve to protect state sovereignty concerns the codifications they offer of the limits to legitimate intervention in domestic affairs. Perhaps most obviously, in its role as the ultimate arbiter of statehood, the United Nations remains deeply implicated in the production of sovereign space. (9) As noted in the UN Charter, the organization is "based on the principle of the sovereign equality of all its Members." (10) The vision of sovereignty offered under the Charter does attach certain riders to the broad principle of nonintervention; in particular, with the evolution of the human rights and Responsibility to Protect regimes, core standards of behavior have emerged on which noninterference in domestic affairs is now conditioned. (11) However, a core function of the organization continues to be the upholding of "the principle of equal rights and self-determination of peoples." In this role, the UN serves to embed the norm of Westphalian sovereignty at the heart of the international system. Beyond the UN, other IOs can be seen to play parallel roles in the protection of sovereign spaces through their definition of a realm of nonintervention. The commitment in the World Bank's Articles of Agreement to avoid intervention in the "political affairs" of member states, for example, served throughout much of the institution's history to define a sphere of operational actions that remained under state actors' purview, beyond the reach of Bank staff. (12)

Additional details into the process through which IOs function to support state sovereignty come from the work of Christian Reus-Smit. (13) Through his analysis of nationalist actors' efforts to gain independent statehood, Reus-Smit outlines how the codified principles laid out by the UN came to function as a "weapon of the weak." (14) J. Samuel Barkin and Bruce Cronin advance a similar position, noting that claims to nonintervention in domestic affairs have been bolstered by actors' ability to demonstrate their adherence to standards of appropriate behavior as laid out by international institutions. (15) In these instances, IOs' rule-based frameworks have provided state actors with relatively well-defined yardsticks with which to enhance their claims to be legitimately entitled to sovereign autonomy.

On the other side of ledger, a range of works have highlighted mechanisms through which IOs serve to enhance sovereign inequality. The hierarchical nature of international institutions has been highlighted as a key mechanism through which sovereign inequality is maintained. In this connection, key international agreements have been found to bestow packages of rights and responsibilities that systematically differentiate between more- and less-powerful members of the system. (16) The Nuclear Non-Proliferation Treaty, with its division between first movers who are able to legitimately hold nuclear weapons and latecomers who must commit to the development of related technologies for exclusively peaceful means, has been held up as a case in point in this regard. (17) In addition, formally constituted IOs have been flagged as heavily implicated in the reproduction of sovereign inequality. Given their low levels of influence in governance processes, less-powerful states often act as rule takers in key IOs in a manner that undermines their capacity to act autonomously. (18) Indeed, the IMF itself has been flagged as a generator of sovereign inequality through this mechanism. (19)

The existing literature, then, tends to portray IOs as contributing either to the protection of sovereign space or to the erosion of sovereign space and exacerbation of sovereign inequality. Findings from the exploration of the IMF engagement with military expenditure, however, present evidence of a more nuanced interaction. In this instance, the organization's formal rules have provided a framework that identifies the parameters of nonintervention while also providing informal mechanisms through which interventions into this policy area have been extended on a case-by-case basis. To unpack the foundations of this contradictory role, it is useful to explore the formal and informal aspects of IO operations.

A key innovation offered by Randall Stone has been to identify the existence across a series of arenas of global governance of twin-tracked formal and informal governance processes. Stone develops the argument that, through "normal times," IOs operate according to a set of codified formal rules. Under these structures, voting power is generally distributed relatively widely among 10 members, and standard operating procedures ensure that broadly equitable treatment is received. For Stone, however, these normal times of governance according to formalized rules and regulations is only possible because the most powerful states in a regime are safe in the knowledge that informal processes exist that enable them to bypass standard procedures and exert exceptional control over an 10's operations. The most powerful states are held back from endemic rule breaking by the need to maintain lower-order states' belief in the legitimacy of existing structures; too much drawing on special privileges will eventually lead to the formation of alternative structures by dissatisfied parties. (20)

Through the study of military expenditure at the IMF, we see that sovereign inequality has been generated through practices of informal governance. Whereas codified rules protect nondisclosure of data on military expenditure, US actions have served to informally tie the institution's resources into the promotion of transparency. In this case, minimal strategic interests were at stake, and no active coalition existed in support of this drive. Within the IMF, the US Executive Director's continued advocacy of military expenditure transparency in fact served to isolate the Director from other representatives in the organization, who viewed the dogmatic behavior as going against the grain of the organization's "consensus based" approach. (21) Given their position of weakness, a series of states whose representatives had previously fought hard to secure sovereignty-protecting rules have been forced to rapidly give up on these protections. Below, I outline the processes through which sovereignty-protecting rules were established, before reviewing the conditions surrounding the US creation of sovereign inequality through the application of informal influence.

Protecting Sovereignty: The Formal Governance of Military Expenditure at the IMF

The questions of how far the IMF does and should reach into the domestic affairs of its member states continue to attract much comment. After briefly reviewing the institutional rules that shape IMF interactions with its members, I provide an overview of the evolution of the organization's surveillance activities and outline boardroom discussions surrounding the establishment of the formal Guidelines on Military Expenditure. These Guidelines served to enshrine states' freedom to maintain secrecy in relation to military spending and to ensure that conditionality was not used in relation to military expenditure. Findings from content analysis of bilateral and multilateral surveillance reports and of the Monitoring of Fund Arrangements (MONA) dataset illustrate the high extent to which staff respected these sovereignty-protecting rules.

As is well known, the chief protagonists at the Bretton Woods conference (US lead negotiator Harry Dexter White and UK lead negotiator John Maynard Keynes) held differing views on the extent to which the IMF should be able to intervene in members' domestic affairs. (22) While ambiguity was used as a means of leaving the door open for staffs' operational reach to extend into the realm of domestic economic management, (23) key clauses in the IMF Articles of Agreement served to enshrine state sovereignty by defining a sphere of nonintervention beyond the organization's purview. Article IV Section 3 forbade staff from taking consideration of member states' "domestic social or political policies" in their dealings with governing authorities, with Schedule C paragraph 7 reiterating this restriction. (24) While the precise meaning of these terms remains contested, the sovereignty-protecting clauses remain an active point of reference at the IMF. Indeed, the content of these protections featured explicitly in the discussions over the potential reform to surveillance practices entailed by enhanced military expenditure monitoring and informed the resulting Guidelines.

Surveillance, the practice of routinely monitoring member states' policy programs and the overall health of the global economic system, has grown to become a core activity of the IMF. (25) However, this strand of work represents a comparatively recent innovation in the life of the institution; it was in particular through the late 1970s and early 1980s that these practices began to be codified. The annual process of bilateral Article IV reports was established with the 1978 Amendment to the Articles of Agreement. Through the Article IV process, the Executive Board reviews staff-produced analyses of key economic trends in individual member states, and any comments made are reported back to the country authorities concerned. (26) Article IV reports served to normalize the act of member states "opening their books" to Fund staff for critical appraisal and have been accompanied by innovations in multilateral surveillance. The roots of multilateral surveillance at the IMF run somewhat deeper; since the 1960s, staff have produced an annual World Economic Outlook (WEO) to "keep score" on the state of the international financial system. (27) However, it was with the Executive Board's approval in 1986 of a Manual on Government Finance Statistics that the level of detail with which domestic expenditure dynamics were reported underwent an appreciable expansion. With the manual providing a template for the collection of internationally comparable data, the Fund began in the late 1980s to publish an annual Government Financial Statistics Yearbook (GFSY). In contrast to the broad brushstroke approach at the WEO data tables, the GFSY drills down to reveal a total of over 200 classes of macroeconomic information in nine standardized tables.

It was at the time of the launch of the GFSY that IMF staff began to establish engagements with the issue of military expenditure. The structure of the GFSY, which in line 702 included a designated location for the recording of data on defense expenditure, is itself illustrative of this burgeoning interest. (28) It was in 1987 that a team within the Fiscal Affairs Department (FAD) first investigated the issue. (29) Under the leadership of FAD's Henri Lorie, a research project was established that culminated in the publication of an IMF Staff Paper on "The Resilience of Military Expenditure" in 1989. This paper concluded not only that developing countries' real expenditure levels in this area had risen fivefold since the 1960s, but also that military spending was a "potentially important policy issue" requiring the attention of Fund staff. (30)

Boardroom discussions on IMF surveillance of military expenditure took place in September and October of 1991. On one side, a coalition including the organization's largest quota-holding representatives spoke in favor of enhancing the Fund's surveillance of military expenditure. Four of the five most powerful single-country constituency Executive Directors spoke in favor of enhanced monitoring, (31) and together thirteen out of the twenty-one Directors present provided backing for this course of action. (32) The views advanced by Mr. Peretz, UK Executive Director, are indicative of those expressed by this grouping. In general, supporters conceptualized this reform as a pragmatic fine-tuning of the Fund's already expanding surveillance activities: "I have no difficulty with the concept that public expenditures in this area are a legitimate subject for Fund surveillance.... The Article IV relationship the Fund has with all its members makes the Fund uniquely well placed to collect data in this area." (33)

By asserting an absolute equivalence between military expenditure and other types of government spending activity, Peretz sought to counter claims that military expenditure represented an exceptionally sensitive (and, therefore, legitimately undisclosed) form of activity:

   There seems to be a strong case for seeking better data to enable
   the Fund staff to develop a better understanding of the interaction
   between military spending and economic growth, so that the Fund can
   give advice to members in the context of surveillance discussions.
   This seems to me to be no more or less appropriate ... [than advice
   on] particular areas of public spending, such as subsidies or
   spending on employment and training measures. (34)

The other members of the pro grouping, who together held the majority of the voting power at the IMF, (35) reiterated this general interpretation throughout the course of the Executive Board Meeting. However, the discussion also contained strong dissenting views.

On the dissenting side, a coalition of Directors presented the case that military expenditure was an atypical form of activity, one that government authorities should not be compelled to disclose. Consisting in the main of representatives of emerging market and less-developed member states, Directors that spoke against the prospect of IMF staff proactively seeking out more detailed data on military expenditure held approximately 30 percent of voting power on the board. (36) In addition to highlighting the practical difficulties faced by low-capacity governments in accurately tracking expenditure flows, the organization's Articles of Agreement were repeatedly invoked as providing strict limitations on the Fund's sphere of operations. In the words of Towe and Al-Jasser, representatives of the Canadian and Saudi Arabian constituencies, respectively:

   We are concerned that an emphasis on such analysis could extend the
   Fund beyond its mandate and distort its character as a monetary
   institution.... Indeed, the requirement of Article IV, Section 3(b)
   that Fund surveillance respect members' domestic social and
   political priorities seems to preclude any such insistence [on the
   provision of military expenditure data]. (37)

   In a nutshell, the Fund should not waste its resources in areas
   that are beyond its mandate and the expertise of its staff and that
   could jeopardise the institution's relations with its members. (38)

And beyond these objections based on appeals to the organization's existing formal rules, the prospect of having state actors be pushed to publicly communicate data on military expenditure was challenged on grounds of sovereignty. Through the course of the discussion, military expenditure monitoring was labeled as "having national security and sovereignty implications" and being a "serious derogation of sovereignty," and it was suggested that "partial transparency was a fundamental exercise of sovereignty related to national security." (39)

In the face of an unusually passionate and polarized discussion, the negotiations over the Managing Director's Summing Up of the meeting were a protracted affair. The Summing Up is an institutionally important document in the IMF; records of board decisions are collated in the Selected Decisions and Selected Documents of the IMF compendium, an annual publication from the IMF Legal Department that functions as an "instruction sheet" for IMF staff. (40) During the negotiations over the Summing Up on military expenditure, the UK Director explicitly pulled rank by suggesting to the Managing Director that those in favor of seeing an enhanced role for the Fund in the collection and analysis of military expenditure "probably constituted a majority of the voting power of the Board." (41) However, in line with the organization's consensus-based approach to decisionmaking, the approved version of the Summing Up was a nuanced statement, reflecting opinions from both sides of the boardroom divide. While noting that information about military expenditures "may be necessary to permit a full and consistent assessment of the member's economic position and policies," the Summing Up also noted that staff must "continue to rely on the voluntary cooperation of authorities in the submission of data," and that authorities are permitted to report an aggregated expenditure figure rather than disaggregate according to military and other expenditure types. The Summing Up also noted that conditions relating to military expenditure should not feature in Fund-supported programs. (42)

It is widely acknowledged that, at the IMF, institutional structures exist to ensure that staff "toe the line" on formal rules and regulations. (43) By analyzing the Monitoring of Fund Arrangements dataset and the content of IMF surveillance publications from the mid-1990s to the late 2000s, it is possible to gauge the manner in which the 1991 Guidelines were operationalized by staff. The MONA dataset contains a record of all conditions attached to IMF programs from 1993 to date. Within this time period, formal conditionality relating to military expenditure has been included in only one case: the Greek Stand-By Arrangement of 2010. In this instance, military expenditure was mentioned in a minor subclause of a structural benchmark relating to deficit reduction; (44) to my knowledge, there is no other record of military expenditure featuring in an IMF program. In addition to this widespread upholding of the direction included in the Guidelines to avoid the use of conditionality related to military expenditure, evidence from a variety of sources demonstrates that nondisclosure of military expenditure remained a widespread practice through this period. In line with the formally codified ruling, state sovereignty continued to rule in this policy area. With the subsequent US interventions into this issue focusing exclusively on developing countries rather than on Fund membership overall, it is useful at this stage to pay particular attention to the experiences of this grouping.

Turning first to the GFSY, it is clear that the number of member states reporting on military expenditure levels has remained modest from 1993 to 2010 (Figure 1). Through the mid- to late 1990s a relatively stable group of fifteen governments released disaggregated figures in this area, rising to around thirty-five through the 2000s. (45) However, paralleling the boardroom split on the subject, no data on military expenditure has been published by developing country governments through the GFSY. (46) With developing country members exercising their nondisclosure prerogative, World Economic Outlooks produced between 1993 and 2010 were largely silent on military expenditure. (47) The Autumn 1999 WEO is the sole publication that includes indicative data on military expenditure, presented in the form of a figure outlining trends that have been aggregated by region, and this and three additional publications contain brief written references to particular case studies and methods of combating off-budget military expenditure. Turning to bilateral surveillance (Table 1), we see that through the decade from 1997 an average of 15 percent of Article IV reports with developing countries contained reference to military expenditure. However, these mentions tend to be extremely cursory in nature, and data on military expenditure does not appear in any of these surveillance documents.

[FIGURE 1 OMITTED]

In spite of the wishes of the IMF's most powerful member states, the formally endorsed Guidelines on Military Expenditure served to maintain states' capacity to exercise autonomy in this area. However, from the late 1990s, informal influence was deployed by the United States to ensure that, on a case-by-case basis, this line in the sand was redrawn. The passage of a series of legislative mandates by the US Congress provided the mechanism through which the US Executive Director was encouraged to use his or her informal influence toward this end.

Contesting Sovereignty: The Informal Governance of Military Expenditure at the IMF

The US Executive Director at the IMF was propelled into the promotion of transparency in the realm of military expenditure by a series of legislative mandates from US Congress through the 1990s. Legislative mandates are widely used by members of Congress as a tool for fermenting operational reform within institutions of global economic governance. A recent edition of the US Treasury "black book" on the international financial institutions contains some 171 directives on the use of US voice and vote inside the IMF and World Bank, its Bretton Woods twin. (48) Whether added as riders to appropriations bills or amendments to pieces of legislation that are less directly related to the operations of the IOs, these mandates constitute legally enforceable terms of reference that must be adhered to by the relevant representative. As such, these directives are a potentially potent mechanism for domestic policymakers to influence international decisions. (49) While successive interventions in the realm of military expenditure monitoring have failed to secure revisions to the Fund's formal sovereignty-protecting rules, the mandates have led to individual developing country borrowers being pushed to alter practices and to disclose data to domestic actors. With these outcomes, the US Director has effectively rolled back the boundary of nonintervention codified in the IMF Guidelines on Military Expenditure.

It was during 1992, with the launch of replenishment negotiations as drawings on Fund resources from postcommunist transition countries began to put pressure on the organization's liquidity, that the IMF's relation to military expenditure became a congressional issue. With established Cold War justifications of military spending having been dramatically removed, global leaders had begun to speak of the potential for a "peace dividend" even before the formal dissolution of the USSR had occurred. (50) In this context, actors on Capitol Hill were concerned to ensure that resources released through the IMF were used to support restructuring programs and not funneled toward the maintenance of military-industrial complexes. (51) To this end, as negotiations over the appropriation of the US contribution to the IMF were brought to a conclusion in October 1992, Congress approved the first of its rulings on the IMF and military expenditure. At the same time as releasing some $12 billion to bolster the Fund's reserves, (52) the efforts by US policymakers to redraw the boundaries of nonintervention were under way.

In their call on the US Executive Director of the IMF to push the organization to publicly report on member states' military expenditure levels, US legislators were relatively precise in the presentation of their aims. In addition to directing the US representative to encourage the IMF to produce annual reports on regional trends in military expenditure, (53) the mandate also noted--in tension with the Guidelines' legitimation of the practice of nondisclosure--that "the United States Executive Director of the Fund shall use the voice and vote of the United States to urge the Fund ... to include in every Article IV consultation with a developing country an analysis of the level of military spending by the developing country in the immediately preceding calendar year." (54) Once this initial guidance was set in place, throughout the 1990s a series of additional rulings were made by Congress to ratchet up the pressure on the US Executive Director and the IMF. All directives remained specifically targeted at the Fund's developing country membership.

In 1994, the Director was empowered to withhold support for a loan application from a developing country if the member had either failed to provide "accurate data on the annual expenditures and receipts of the armed forces," or failed to demonstrate a commitment to "mak[ing] substantial reductions in excessive military expenditure." (55) Two years later, the force of the mandate was significantly increased; whereas previously the Executive Director was allowed flexibility, the September 1996 amendment removed the element of discretion by ruling that loans must be opposed if accurate data or a commitment to reduce excessive military spending was not forthcoming. (56) In another round of revisions, approved in October 1998, the imperative force of these directions was increased further, as the US representative was called on to "aggressively" and "vigorously" police this set of policy aims. (57) Throughout the 1990s, then, the congressional desire to ferment operational change regarding IMF surveillance of developing country members' military expenditure was repeatedly demonstrated. However, as a consequence of a grace period introduced by Congress to allow the Treasury to develop an implementation plan, (58) it was only from 1999 that US Directors began to use their informal influence to advance the issue in the IMF boardroom.

In preparation for their 1999 deadline, the Treasury's International Affairs Division (IAD) established an interagency policy group to provide guidance on the implementation of the legislation. To operationalize the directive that action be taken against states that were failing to provide accurate data on military expenditure, the policy group established a definition of acceptable practice in this area: "A country must routinely be conducting a post-expenditure examination, verification of accuracy, and reconciliation of irregularities of receipts that fund the military. Results of the audit must be reported to a nonmilitary entity." (59) By applying this definition, it was decided by the policy group that some twenty-two low-income countries were failing to fulfill the requirements of the legislative mandates on military expenditure. With the activation of the mandates, military expenditure became an established feature of institutionalized links between the US executive directorate and IAD, with the watch list being reviewed at the weekly strategy meetings at which the US positions on upcoming loan applications were decided.

The legislative mandates became active on 1 October 1999. One year later, five of the original group of twenty-two countries that the policy group had deemed to be noncompliant in the provision of data on military expenditure had approached the IMF to begin a loan arrangement. In line with the IMF Guidelines, none of the programs agreed on contained conditions relating to military expenditure. And while the IMF Board approved the applications from Burkina Faso, Guinea-Bissau, Indonesia, Kazakhstan, and Rwanda, in each case US concerns were raised through the use of abstentions and no-votes. Operational shifts occurred in all five of these countries, with IMF staff playing a coordinating role in the process. In addition to channels of communication between officials from the US directorate and borrowing country directorates, IMF staff engaged in lending programs acted as informal promoters of the US agenda. Through these mechanisms, it was confirmed to country officials that US approval of a loan arrangement and release of subsequent tranches of funding would be conditional on satisfactory publication of data on military expenditures. (60)

In the case of Burkina Faso, Treasury staff supplemented the Executive Director's intervention in the boardroom by "working through the IMF Country Director" to raise the issue with country officials following the loan approval. In April 2000 the president of Burkina Faso released a decree ruling that the Defence Ministry produce an annual military audit and, as a consequence of the operationalization of this decree, Treasury judged the country to be in compliance with the legislative mandate in June 2000. Similar dynamics unfolded in each of the other four cases throughout 2000. Indeed, by the close of this period Guinea-Bissau was the only one of the five to be judged to be still noncompliant, and even in this case Treasury judged that sufficient processes were in place such that compliance was "anticipated." (61)

The remaining seventeen states on the policy group's noncompliance watch list provide a useful control sample. These seventeen states did not attempt to borrow from the IMF throughout the first year in which the legislative mandate was in force and, during this time, the Treasury judged that just one of the cohorts moved into a state of compliance. Following this initial activity in the immediate aftermath of the activation of the legislative mandates on military expenditure, interventions from the US director have continued to occur sporadically. In early 2004, the US director abstained from a vote on a loan application from Burundi on the grounds that the government had failed to adequately audit military spending and appeared to direct an excessive level of resources to this area. (62) And through 2008, a further three abstentions occurred on these grounds. Five additional abstentions took place through 2010. (63)

The US Treasury has judged that its successes in the application of the legislative mandates can be attributed to a combination of borrowing countries' fear of losing access to IMF resources should they continue to be in noncompliance, and IMF staffs' belief in the beneficial effects of greater transparency in this area. (64) In the decade since the activation of the legislative mandates, around fourteen lower-income member states have been subject to informal conditionality at the IMF in relation to military expenditure monitoring. Through this process, interventions from the US representative have served to roll back the frontiers of nonintervention enshrined in the formally approved IMF Guidelines on Military Expenditure. Whereas the formal Guidelines permitted nondisclosure and prohibited the use of policy conditionality relating to military expenditure, the US-led actions have tied policy reform in this area to IMF loans, with this mechanism leading to the establishment of regularized military expenditure audits and the disclosure of data collected.

Conclusion

The theory and practice of sovereignty remains the subject of much debate in the existing international relations literature. However, among the significant points of contention, there is a broad consensus that experiences of sovereignty vary widely across the international system and that international organizations can play a significant role in shaping sovereign inequality. Throughout this article, I have drawn on the case study of the IMF engagement with military expenditure to extend insights into this area. In particular, drawing on recent work from Stone and others, I have explored the intersection between informal governance and sovereign inequality.

In the case examined, a coalition of lower-order states was able in the early 1990s to ensure that the organization's formal Guidelines on Military Expenditure served to protect their sovereign capacity in this area. However, since the late 1990s, the US representative has pushed for reform from IMF borrowers whose military expenditure auditing processes were judged to be inadequate. Driven by a series of legislative mandates, the Fund's major shareholder has rolled back the frontiers of nonintervention at the IMF. Sovereign inequality has resulted from the application of this informal influence.

Significant reversals of position on the part of low-income borrowing countries resulted from the US application of informal pressure in these cases. With a clear cleavage having existed between developed and developing country representatives in the IMF boardroom during initial discussions on military expenditure, most borrowers that were pushed to release data on military expenditure were represented by directorates that had initially sought to secure sovereignty-protecting clauses. The level of strategic interest involved for the United States in these interventions was relatively low. This observation challenges existing insights from the literature on informal governance, which suggests that derogation from formal rules tends to be reserved for more important cases. Conventional wisdom understandings of international politics are likely to remain untroubled by the outcome observed in this case, which ultimately saw powerful state preferences trumping materially weak states' preferences. Through further studies, a more developed sense can be established of the prevalence of discriminatory rule breaking in global governance and, consequently, of the extent to which informal governance systematically produces sovereign inequality across the practice of international politics.

Notes

Liam Clegg is lecturer in international relations at the Department of Politics, University of York. He is author of the monograph Controlling the World Bank and IMF: Shareholders, Stakeholders, and the Politics of Concessional Lending (2013), and his work has appeared in journals including New Political Economy, Review of International Organizations, Review of International Political Economy, and Review of International Studies.

(1.) Randall Stone, "The Scope of IMF Conditionality," International Organization 64, no. 4 (2008): 589-620; Randall Stone, Controlling Institutions (Cambridge: Cambridge University Press, 2011); Jeffrey Chwieroth, '"The Silent Revolution': How the Staff Exercise Informal Governance over IMF Lending," Review of International Organizations 1, no. 3 (2013): 265-290; Randall Stone, "Informal Governance in International Organizations," Review of International Organizations 7, no. 3 (2013): 121-236.

(2.) Through this article, case study evidence supports the conceptual narrative that is developed. As is highlighted by Alexander George and Michael Bennett in Case Studies and Theory Development in the Social Sciences (Cambridge: MIT Press, 2005), such a technique can be used to explore potential variables and mechanisms that shape observed outcomes. Overall, I present this article as an exercise in process tracing and thick description, rather than an attempt to establish foundations for generalizable findings.

(3.) Ron Smith, "Military Expenditure and Capitalism," Cambridge Journal of Economics 1, no. 1 (1977): 61-76; Frank Blackarby and Thomas Ohlson, "Military Expenditure and the Arms Trade: The Problems of Data," Security Dialogue 13, no. 4 (1982): 291-305; Saadet Derger and Somnath Sen, Military Expenditure: The Political Economy of International Security (Oxford: Oxford University Press, 1990); Alexandra Homolar, "The Political Economy of National Security," Review of International Political Economy 17, no. 2 (2010): 410-423.

(4.) Whereas legislation pushing the US Executive Director to promote military expenditure transparency was initially passed to try to limit such spending from former Cold War opponents, its application has primarily been targeted toward low-income developing countries.

(5.) Indeed, it is suggested that the pursuit of transparency on military expenditure served to isolate the US Executive Director. General Accounting Office (GAO), International Monetary Fund: Attempts to Advance US Policies at the Fund (Washington, DC: GAO, 2001), p. 7.

(6.) Stephen Krasner, Sovereignty: Organized Hypocrisy (Princeton: Princeton University, 1999), pp. 9-25.

(7.) Stephen Krasner, "Sovereignty: An Institutional Perspective," Comparative Political Studies 2, no. 1 (1988): 66-94.

(8.) Stephen Krasner, "Globalization and Sovereignty," in David Smith, Dorothy Solinger, and Stephen Topik, eds., States and Sovereignty in the Global Economy (London: Routledge, 1999), p. 220.

(9.) J. Samuel Barkin and Bruce Cronin, "The State and the Nation: Changing Norms and the Rules of Sovereignty in International Relations," International Organization 48, no. 1 (1994): 123-124; Janice Thomson, "State Sovereignty in International Relations: Bridging the Gap Between Theory and Empirical Research," International Studies Quarterly 39, no. 2 (1995): 229-230.

(10.) United Nations Charter, Chap. I, Art. 2(i), available at www.un.org/en /documents/charter/chapterl.shtml, accessed 4 July 2013.

(11.) Gareth Evans and Mohamed Sahnoun, "The Responsibility to Protect," Foreign Affairs 81, no. 6 (2002): 99-110; Alex Bellamy, "Realizing the Responsibility to Protect," International Studies Perspectives 10, no. 2 (2009): 111-128.

(12.) Catherine Weaver, Hypocrisy Trap: The Poverty of Reform at the World Bank (Princeton: Princeton University, 2008); Liam Clegg, "The Governance of the World Bank," in Tony Payne and Nicola Phillips, eds., Handbook of the International Political Economy of Governance (London: Edward Elgar), pp. 259-274.

(13.) Christian Reus-Smit, "Human Rights and the Social Construction of Sovereignty," Review of International Studies 27, no. 4 (2001): 519-538.

(14.) The phrase, not used by Christian Reus-Smit, comes from James Scott, Weapons of the Weak: Everyday Forms of Peasant Resistance (New Haven: Yale University, 1985).

(15.) Barkin and Cronin, "The State and the Nation"; see also J. Samuel Barkin, "The Evolution of the Constitution of Sovereignty and the Emergence of Human Rights Norms," Millennium 27, no. 2 (1998): 229-252.

(16.) Jack Donnelly, "Sovereign Inequalities and Hierarchy in Anarchy," European Journal of International Relations 12, no. 2 (2006): 139-170.

(17.) Joseph Nye, "NPT: The Logic of Inequality," Foreign Policy 59, no. 1 (1985): 123-131.

(18.) Mary Tsai, "Globalization and Conditionality: Two Sides of the Same Coin," Law and Policy in International Business 31, no. 4 (2000): 1317-1330.

(19.) Catherine Lee, "To Thine Ownself Be True: IMF Conditionality and the Erosion of Economic Sovereignty in the Asian Financial Crisis," University of Pennsylvania Journal of International Economic Law 24, no. 4 (2003): 875-903; Liam Clegg, "Global Governance Behind Closed Doors: The IMF Boardroom, the Enhanced Structural Adjustment Facility, and the Intersection of Material Power and Norm Stabilisation in Global Politics," Review of International Organizations 7, no. 3 (2012): 285-308.

(20.) Stone, Controlling Institutions, pp. 11-48; Stone, "Informal Governance in International Organizations," pp. 121-126. Additional recent explorations of manifestations of informal governance are provided by Kevin Morrison, "Membership No Longer Has Its Privileges: The Declining Informal Influence of Board Members on IDA Lending," Review of International Organizations 8, no. 2 (2013): 291-312; Marieke Kliene, "Knowing Your Limits: Informal Governance and Judgement in the EU," Review of International Organizations 8, no. 2 (2013): 245-264; Chwieroth, "The Silent Revolution."

(21.) GAO, International Monetary Fund, p. 7.

(22.) For Harry Dexter White, representing the major creditor, the ability to compel policy adjustments from member states represented a necessary means of ensuring that IMF resources secured desirable outcomes; for John Maynard Keynes, representing a likely debtor, the aim was to retain autonomy for domestic authorities and to avoid the organization from "being grandmotherly" by forcing interfering policy advice onto borrowing governments. See Sidney Dell, "On Being Grandmotherly: The Evolution of IMF Conditionality," Princeton University Department of Economics Working Paper Series no. 144 (1981): 1. A detailed overview of this history is provided in Benn Steil, The Battle of Bretton Woods: John Maynard Keynes, Harry Dexter White, and the Making of a New World Order (Princeton: Princeton University Press, 2013).

(23.) Jacqueline Best, "Ambiguity and Uncertainty in International Organizations: A History of Debating IMF Conditionality," International Studies Quarterly 56, no. 4 (2012): 674-688.

(24.) See IMF website at www.imf.org/external/pubs/ft/aa/index.htm, accessed 7 August 2013.

(25.) Manuela Moschella, Governing Risk: The IMF and Global Financial Crises (London: Palgrave, 2010).

(26.) Manuela Moschella, "IMF Surveillance: The Past, the Present, and the Future of the Reform Process," Global Society 26, no. 1 (2012): 43-60.

(27.) James Boughton, Silent Revolution: The International Monetary Fund 1979-1989 (Washington, DC: IMF, 2001), pp. 227-230.

(28.) The inclusion of line 702 appears to have attracted little comment; archival records from committee meetings on the standardization of IMF data collection practices focused largely on technical issues surrounding intercountry comparability. See, for example, IMF, "Management Data System of Principal Economic Indicators (Office Memorandum from W. O. Habermeire to Tun Thin, 7th April)," IMF Archive, Fiscal Affairs Department Fond, Fiscal Affairs Department Immediate Office Sous-Fond, Vito Tanzi Chronological File Series (1982); IMF, "Time Series Data--Status Report (Office Memorandum from Tun Thin to H. Johannes Witteveen, 21st April)," IMF Archive, Fiscal Affairs Department Fond, Fiscal Affairs Department Immediate Office Sous-Fond, Vito Tanzi Chronological File Series (1982).

(29.) Jacques Polak, "The Changing Nature of Conditionality," OECD Working Paper No. 41 (Paris: OECD, 1991), p. 39.

(30.) Paula di Masi and Henri Loire, "How Resilient Are Military Expenditures?" IMF Staff Working Papers 36, no. 1 (1989): 130.

(31.) These were, namely, France, Germany, the United Kingdom, and the United States. At present there are five Executive Directors with single-country constituencies. See IMF website at www.imf.org/external/np/sec/memdir/eds.aspx, accessed 19 March 2012.

(32.) Author's analysis of IMF, "Executive Board Meeting Minutes (27th September, 1991)," IMF Archive Reference EBM/91/134.

(33.) IMF, "Executive Board Meeting Minutes (27 September 1991)," pp. 18-19.

(34.) Ibid., pp. 19-20.

(35.) The four single-country constituency Directors alone held 40 percent of the voting power.

(36.) Author's figure, calculated by matching positions stated in the boardroom discussion with the overview of Directors' voting power contained in Appendix VII of the 1991 IMF Annual Report.

(37.) IMF, "Executive Board Meeting Minutes (27 September 1991)," p. 35.

(38.) Ibid., p. 40.

(39.) Ibid., pp. 20, 48, 70.

(40.) For more information on the processes through which formal decisions made by the IMF Board are enforced, see Bessma Momani, "IMF Staff: Missing Link in Fund Reform Proposals," Review of International Organizations 2, no. 1 (2007): 39-57; Liam Clegg, "Social Spending Targets in IMF Concessional Lending: US Domestic Politics and the Institutional Foundations of Rapid Operational Change," Review of International Political Economy 21, no. 3 (2014): 735-763.

(41.) IMF, "Executive Board Meeting Minutes (2nd October, 1991)," IMF Archive Reference EBM/91/138.

(42.) IMF, Selected Decisions of the IMF, 32nd ed. (Washington, DC: IMF, 2007), pp. 86-87.

(43.) Bessma Momani, "Recruiting and Diversifying IMF Technocrats," Global Society 19, no. 2 (2005): 167-187; Momani, "IMF Staff"; Clegg, "Social Spending Targets."

(44.) The program includes a structural benchmark stipulating that the government must "publish a medium-term budget laying out time-bound plans to address: (i) restructuring plans for large and/or loss-making state enterprises; (ii) the closure of unnecessary public entities; (iii) tax reform; (iv) reforms of public administration; (v) the public wage bill; and (vi) military spending." Author's analysis of MONA dataset. See IMF website at www.imf.org, accessed 13 August 2013.

(45.) Expressed in proportionate terms, these increases saw the percentage of GFSY participants disclosing military expenditure data rising from under 10 percent in the mid-1990s to almost 25 percent in the mid-2000s. Author's analysis of GFSY data tables.

(46.) For the purpose of this analysis I take a broad understanding of the term developing country to include low-income and lower-middle-income countries, as defined by the World Bank. For the full list of countries, see the World Bank website at http://data.worldbank.org/about/country-classifications/country-and-lending-groups #Lower_middle_income, accessed 10 December 2011.

(47.) Author's analysis of World Economic Outlook publications from 1993 to 2010.

(48.) Author's analysis of US Treasury, Department of the Treasury Compilation of Legislative Mandates Applying to US Participation in the IFIs (Washington, DC: US Treasury, 2010).

(49.) Kathryn Lavelle, Legislating International Organization: US Congress, the IMF, and the World Bank (Oxford: Oxford University Press, 2011); Clegg, "Social Spending Targets."

(50.) Evidence of this shift is contained in a 1991 IMF Staff Paper, which makes note of recent moves by the German and Japanese governments to incorporate assessments of military expenditure into allocations of overseas development assistance, and also of G7 calls on developing countries to transfer military expenditure toward productive activities. IMF, Military Expenditures and the IMF (Washington, DC: IMF, 1991).

(51.) Jessica Matthews, "A New String on Third World Loans," Washington Post, 25 October 1991, sec. A2.

(52.) See "Bush Signs Measure to Increase US Share in IMF Aid to Ex-USSR," Journal of Commerce, 27 October 1991, sec. A2.

(53.) United States Code, Title 22, chap. 7, subchap. XV, sec. 286MM(b). Quoted in General Accounting Office, US Policies at the Fund (Washington, DC: GAO, 2001), pp. 23-29. For the full text, see US Houses of Congress website at http://uscodebeta.house.gov/view.xhtml?req=granuleid:USC-title22-section286mm&num=0, accessed 27 January 2012.

(54.) Ibid.

(55.) United States Code, Title 22, chap. 7, subchap. XV, sec. 262(o-l). Quoted in General Accounting Office, US Policies at the Fund. For the full text, see US Houses of Congress website at http://uscodebeta.house.gov/view.xhtml?req=granuleid:USCtitle22-section262o-1 &num=0, accessed 27 January 2012.

(56.) United States Code, Title 22, chap. 7, subchap. XV, sec. 262(k). Quoted in General Accounting Office, US Policies at the Fund. For the full text, see the US Houses of Congress website at http://uscodebeta.house.gov/view.xhtml?req=granuleid:USC-title22-section262k-l&num=0, accessed 27 January 2012.

(57.) United States Code, Title 22, chap. 7, subchap. XV, sec. 262(o-2). Quoted in General Accounting Office, US Policies at the Fund. For the full text, see US Houses of Congress website at http://uscodebeta.house.gov/view.xhtml?req=granuleid:USCtitle22-section262o-2&num=0, accessed 27 January 2012.

(58.) The 1996 Legislative Mandate set 1 October 1999 as the target date. GAO, US Policies at the Fund, p. 57.

(59.) Ibid., pp. 57-58.

(60.) Ibid., p. 7.

(61.) Ibid., p. 61.

(62.) A full list of the US Executive Director's votes can be found at the US Treasury website at www.treasury.gov/resource-center/intemationaFint-monetary-flmd/Pages/imf .aspx, accessed 14 August 2013. I have cross-referenced information gathered from this source with the relevant Executive Board Meeting minutes, which can be found at the IMF website at www.imf.org/extemaFnp/arc/eng/archive.htm, accessed 14 August 2013.

(63.) The rationale behind these 2010 abstentions cannot be readily ascertained, as relevant Executive Board Meeting minutes remain under the IMF Archive's embargo period.

(64.) GAO, US Policies at the Fund, p. 62.