Can International Organisations Become “Autonomous Sites of Authority”?


Isabella Lowenthal-Isaacs– E –International  Relation

This content was written by a student and assessed as part of a university degree. E-IR publishes student essays & dissertations to allow our readers to broaden their understanding of what is possible when answering similar questions in their own studies.

The recent crop of International Relations (IR) literature dedicated to International Organisations (IOs) is reflective of the evolution of their importance as actors in the international community.  The last fifteen years have witnessed seminal publications that offer different interpretations of the role of IOs from entirely new vantage points. But despite the growing size of the literature, it still suffers from incoherence and fragmentation. The lines of arguments are not as clear cut as in other areas of IR. Most importantly, autonomy is defined too variedly. Whilst neoliberalists incorrectly measure autonomy as the extent to which an IO behaves independently, constructivists define it through the creation of cultural norms and meanings.

For that reason, it is imperative to point out that IOs differ immensely in size, structure and purpose. To argue that all IOs can be deemed ‘autonomous sites of authority’ – or not is too simplistic.  To do so would be to fall victim to the same intellectual rigidity in the earlier literature, that relies solely on theoretical assumptions. Through the analysis of the World Bank (Bank) and the United Nations Security Council (UNSC) this essay demonstrates the variation in IO independence, arguing that IOs do have the tendency to ‘run amok’[1]. But critically, this is not the source of their autonomy. It is rather a symptom of the structural problems with the delegation of the Principal-agent (PA) paradigm. The neoliberalist tendency to equate agency slack with autonomy poses dangerous consequences for academics’ understanding of the role of IOs. The constructivist approach is more fruitful, but still has its limits. IOs can become ‘autonomous sites of authority’, shaping state behaviour through their intellectual and technocratic power. Nevertheless, touching on rationalist thinking, their independence will never exceed that or their creators, and must always sit, at least broadly, within the scope of their principals’ interests.

Realism and its successive schools of neorealism and even neoliberalism approach international relations through a state-centric ontology. Realist scholars have traditionally paid little attention to IOs and, where they have, IOs are viewed as as conduit through which powerful states can achieve their geo-political goals. Mearsheimer maintains that ‘institutions have minimal influence on state behaviour’.[2] In the context of IOs, the realist creed is outdated in that it is unable to explain growing trends: increasing multilateral cooperation that takes place in IOs, instances where they they have deviated from the delegated authority and most importantly their increasing legitimacy as creators of global norms. If not outdated, then it is certainly in need of some empirical validation. Most realist analyses of IOs rely on ‘analytic assumption’[3], and lacks the apparatus to allow us to rigorously engage with Barnett and Finnemore’s assertion.

The most restricting feature of the realist school is its tendency to view IOs as dutiful servants.[4] Given the centrality of ‘self-interest’ in the realist philosophy, somewhat paradoxically, it is unable to concede that IOs might have interests independent of the member states. One main reason for the creation of IOs is their ability to offer superior intellectuality on specialised areas. Thus, huge organisations such as the Bank concentrate a huge amount of experts – ‘and concentrations of people with the same expertise…can create an organizational worldview distinct from the larger environment.’[5] The behaviour of IO individual staff has seldom been looked at. But it is important to note that staff will have their own career goals. Even if we maintain that the realist assumption that IOs are not ‘actors’ in their own right, they can still behave erratically. Armies can defect. Unions can strike. Bureaucracies too, can behave unfavourably to their creators. Thus IOs can possess both a political personality and self-interests at an individual level. The problem with overlooking such basic human tendencies comes into acute focus once we understand the structural influences of IO behaviour.

Nevertheless, the statist assertions that institutions have no effect on state behaviour in the realist school run contrary to the historical truth. The Bank exhibited unruly behaviour for almost the entirety of the 1980s in a series of detrimental environmental policy choices in Brazil and Indonesia that prompted changes in policies of its member states and eventually institutional reform of the organisation.[6] The neoliberal institutionalist use of the (PA) theory [7] is a more useful tool in helping us understand such erratic behaviour.

Rooted in neoliberal microeconomics the PA theory states that principals (states) empower agents (IOs) with ‘real decision making authority’.[8] Since there will always be some disjuncture between the objectives of the principal and agent, the principal looks to limit the extent to which the agent can pervert its instructions. At the crux of the PA logic: sometimes it is more costly for states to apply mechanisms of control than it is to allow agents more discretion.

The Bank offers services that states are unwilling and sometimes unable to provide themselves. States need the Bank and on an ongoing basis. Thus neoliberalists argue the costs of monitoring Bank slippage was for a long time greater than awarding them a wide scope of delegation. However, the scope of delegation can be narrowed, if IOs are perceived to be acting to broadly outside of their principals’ interests, evidenced by the eventual reforms of the Bank. The fact that the Bank were able to run amok for so long is a symptom of the structural inadequacies in their delegation – but not of their independence.

Tierney and Nielsen show that the Bank is subject to both collective and multiple principals.[9] Because of the preference heterogeneity of those principals, the ability to reign in errant behaviour proved difficult for many years. Through empirical analysis, it is convincingly shown that it was only once the Bank’s most powerful members increased convergence that the organisational and behavioural reforms at the Bank eventually stuck.[10] However, there are three key flaws in neoliberalist conclusions. Firstly, the fact that reforms necessitated a coalition of its most powerful principals serves to uphold the realist tradition. In the end, the Bank was subject to the will of the G-7 states. Secondly, independence is not conditional and in fact the Bank’s deviance eventually resulted in more robust control mechanisms of the delegation relationship. Importantly, the Bank increased environmentalist professionals to prevent future mishaps. Finally, and most importantly, the fact that the Bank misused their delegated authority does not represent autonomy, it represents autonomous behaviour. Should we really equate rogue behaviour with independence? There are other, more compelling, sources of independence, put forward by the constructivist school.

The primary source of Bank independence is from its expertise and knowledge. Barnett and Finnemore correctly describe the Bank as a ‘magnet for the ‘best and brightest among development experts’[11]It is their expertise that allows IOs to create new social norms and thus independently influence other actors. This is most clearly witnessed in IOs’ achievements in integrating social aspects into the world’s definition of security. Such autonomy has even led to counter-productive IO policies. The concept of ‘structural adjustment’ legitimated invasive intervention of IOs in developing countries. SAP programs saw the institutional reorganisation of the political economies of developing countries. In the case of Africa at least, the imposition of neoliberal policies had devastating effects.[12] Owing to its specialisation, the Bank’s purported ‘solution’ to the African economy, during the 1990’s ‘neo-liberalism had become a secular religious doctrine’ – ferociously applied to the developing world.[13] We can therefore see that specialisation is a source of independence for IOs, in their ability to inform global norms and significantly alter foreign policies.

By contrast to the Bank, the UNSC is viewed on the world stage as a more transparent tool for which powerful states can achieve their interests. The UNSC is an extremely important organ of the United Nations. Its actions are highly visible and determine the outcome of conflict and warfare.[14] Conventional wisdom would have us believe that the UNSC is ineffective in restraining the geo-political will of the United States among other members of the P5. The failure to prevent the war in Bosnia or the US intervention in Iraq is often alluded to by realists to push the idea of our ‘false belief’ in IOs.[15]  It is true, the UNSC is inherently flawed allowing for overwhelming influence from the hegemonic powers of the P5 and thus compromising its ability to appear independent. The council consists of five permanent members, the US, the United Kingdom, France, Russia and China and ten additional elected members, who serve a strict 2-year term.[16] The P5’s ability to weaponise their veto considerably compromises the UN’s independence. The extensive use of the P5 veto in the Israel-Palestine conflict for example has left a somewhat ambivalent official stance towards the conflict.[17] Mahmood shows that even as the use of the veto has decreased, the mere threat of its use has not diminished.[18]

Further to the tangibility of these structural limits to the UNSC’s independence, there exists an element of ‘bribery’ in the seats of the council. A recent study outlined the ways in which the principals of the International Monetary Fund (IMF) use their influence to gain leverage in the UNSC. Using a robust research design, Dreher et al provide compelling evidence to show that between 1951 and 2004 there is a ‘positive relationship between temporary UNSC membership and participation in IMF programs.’[19]On average, a developing country serving on the security council  can anticipate an additional $16 million from the United States and $1 million from the United Nations.[20] Essentially, this means that votes on the UNSC can be bought. Although we cannot predict the voting patters without such aid, the existence of informal means of political manipulation renders the UNSC, and other IOs, a political trading accessory. The constructivist school cannot account for this.

Nevertheless, as maintained earlier, there are other sources of independence. Thompson’s understanding of the 2003 Iraq invasion is skewed. He argues understanding of UNSC failures rely on a ‘literal interpretation of the Security Council’s role… envisioned in the Charter…as responsible for identifying threats and applying military force against them if necessary.’[21] How else should we interpret the SC role? This essay takes issue with Thompson’s assertion that this ‘sets too high a bar for judging success.’[22] If not constrain the presence of ‘global horse trading’, the UNSC should be able to, at the very least, effectively stop illegitimate ‘interventions’.

The neoliberalist approach sees the preference heterogeneity of the UN as a source of power in the case of Iraq – bringing credibility to the screening function.[23] But the fact that some countries are willing to trade their UNSC vote for IMF participation largely exposes the poverty of this logic. To a lesser extent than the Bank, the UNSC enjoys a level of neutrality – derived from their technocratic expertise. Voeten provides qualitative research to show that states look to the UNSC to ascertain the level of opposition they might face from other actors, including at home.[24] In this sense, the UNSC serves as both a valuable indicator and judge of appropriate behaviour. Thompson shows that with an SC mandate, the IO can screen a ‘desirable intervention policy from an undesirable one.’[25] For example, public support in regards to the first  Gulf War was significantly higher than support for the second;[26] UN authorisation appearing to be an important variable. But unless we dissect the underlying geopolitical reasons for the UN’s stance on such matters we cannot categorically say that this represents independence. Thus, the case of Iraq shows us that the UNSC possessed power from perceived neutrality as opposed to genuine political independence.

To conclude, IOs can become autonomous sites of authority but they are never wholly independent actors. Their influence derives from their bureaucratic capacity. Namely, the IOs concentration of expertise gives them a neutral quality that states cannot compete with. IO power should be understood as ‘soft’. The Bank has been played a pivotal role in global development policy and the UNSC – albeit to a lesser extent – has set standards in the security realm. IOs are also pluralistic in nature and therefore can behave pathologically. The Bank’s persistence in fuelling money into unsavoury environmental projects is an irrefutable example of this. But this should not be confused with independence. There is no one school of thought to adequately explain the variation in IO independence. Whilst this essay borrows the realist rationale, it shows that its deficiencies are unable to explain recent trends. Subscribers of neoliberalism have contributed valuable quantitative findings to show why principals delegate and when that delegation is exploited. However, its understanding of the role of IOs is distorted by its flawed causal logic – measuring independence as slack. Constructivist arguments are the most relevant – showing how IOs influence the world as a yardstick of norms. Yet even constructivism cannot account for all IOs. IOs can and continue to be independent influencers of state behaviour. But changes can only be made through mutual policy adjustment – they will never have the institutional capacity to relentlessly pursue their own agenda. This hypothesis should sufficiently explain the varying roles of the World Bank and the United Nations Security Council in the geopolitical landscape.


Barnett, M. and Finnemore, M. (1999). The Politics, Power, and Pathologies of International Organizations. International Organization, 53(04), pp.699-732.

Dreher, A., Sturm, J. and Vreeland, J. (2009). Global horse trading: IMF loans for votes in the United Nations Security Council. European Economic Review, 53(7), pp.742-757.

Caffentzis, George, (2002), Neoliberalism in Africa, Apocalyptic Failures and Business as Usual Practices, Turkish Journal of International Relations 1 (3), pp. 80-104.

Finnemore, M. (1993). International organizations as teachers of norms: the United Nations Educational, Scientific, and Cutural Organization and science policy. International Organization, 47(04), p.565.

George, Susan and Fabrizio Sabelli 1994. Faith and Credit: The World Bank’s Secular Empire. London: Penguin.

Haftel, Y. and Thompson, A. (2006). The Independence of International Organizations. Journal of Conflict Resolution, 50(2), pp.253-275.

Hawkins, D. (2008). Delegation and agency in international organizations. Cambridge: Cambridge University Press.

Johnson, T. (2013). Institutional Design and Bureaucrats’ Impact on Political Control. The Journal of Politics, 75(1), pp.183-197.

Kuziemko, I. and Werker, E. (2006). How Much Is a Seat on the Security Council Worth? Foreign Aid and Bribery at the United Nations. Journal of Political Economy, 114(5), pp.905-930.

Mahmood, F. (2013). Power Versus the Sovereign Equality of States: The Veto, the P-5 and United Nations Security Council Reforms. Perceptions, 18(4), pp.117-138.

Mearsheimer, J. (1994). The False Promise of International Institutions. International Security, 19(3), p.5.

Nelson, P. (1995). The World Bank and Non-Governmental Organizations. Palgrave Macmillan.

Nielson, D. and Tierney, M. (2003). Delegation to International Organizations: Agency Theory and World Bank Environmental Reform. International Organization, 57(02).

Thompson, A. (2006). Coercion Through IOs: The Security Council and the Logic of Information Transmission. International Organization, 60(01).

Voeten, E. (2005). The Political Origins of the UN Security Council’s Ability to Legitimize the Use of Force. International Organization, 59(03).

Weller, P. and Yi-Chong, X. (2010). AGENTS OF INFLUENCE: COUNTRY DIRECTORS AT THE WORLD BANK. Public Administration, 88(1), pp.211-231.


[1] Daniel L. Nielson and Michael J. Tierney, “Delegation To International Organizations: Agency Theory And World Bank Environmental Reform”,(2003) <>.

[2] John J. Mearsheimer, “The False Promise Of International Institutions”, International Security, 19.3 (1994), 5 <>

[3]  Michael N. Barnett and Martha Finnemore, “The Politics, Power, And Pathologies Of International Organizations”, International Organization, 53.04 (1999), 699-732

[4] Patrick Weller and Xu Yi-Chong, “Agents of influence: country directors at the World Bank”, Public Administration, 88.1 (2010), 211-231 <>.

[5] Michael N. Barnett and Martha Finnemore, (1999), 699-732

[6] Daniel L. Nielson and Michael J. Tierney, (2003)

[7] Daniel L. Nielson and Michael J. Tierney, (2003)

[8] Patrick Weller and Xu Yi-Chong (2010),

[9] Daniel L. Nielson and Michael J. Tierney, (2003)

[10] Daniel L. Nielson and Michael J. Tierney, (2003)

[11] Michael N. Barnett and Martha Finnemore, (1999), 699-732

[12] Neoliberalism in Africa, Apocalyptic Failures and Business as Usual Practices George Caffentzis, (2002).

[13] George, Susan and Fabrizio Sabelli 1994. Faith and Credit: The World Bank’s Secular Empire. London: Penguin

[14] Axel Dreher, Jan-Egbert Sturm and James Raymond Vreeland, “Global Horse Trading: IMF Loans For Votes In The United Nations Security Council”, European Economic Review, 53.7 (2009), 742-757 <>.

[15] John J. Mearsheimer, (1994)

[16] Axel Dreher, Jan-Egbert Sturm and James Raymond Vreeland, (2009)

[17] Fakiha Mahmood, “Power Versus The Sovereign Equality Of States: The Veto, The P-5 And United Nations Security Council Reforms”, Perceptions, 18.4 (2013), 117-138.

[18] Fakiha Mahmood, (2013).

[19] Axel Dreher, Jan-Egbert Sturm and James Raymond Vreeland, (2009).

[20] Ilyana Kuziemko and Eric Werker, “How Much Is A Seat On The Security Council Worth? Foreign Aid And Bribery At The United Nations”, Journal Of Political Economy, 114.5 (2006), 905-930 <>.

[21] Alexander Thompson, “Screening Power: International Organizations As Informative Agents”, in Delegation and Agency in International Organizations, 1st edn (Cambridge: Cambridge University Press, 2006), pp. 229-255.

[22] Alexander Thompson, (2006), p.230.

[23] Alexander Thompson, (2006).

[24] Voeten, E. (2005). The Political Origins of the UN Security Council’s Ability to Legitimize the Use of Force. International Organization, 59(03).

[25] Alexander Thompson, (2006), p.250.

[26] Alexander Thompson, (2006), p.250.

Written by: Isabella Lowenthal-Isaacs
Written at: IBEI
Written for: Charles Roger, Robert Kissack and Frank Borge Weitzke
Date written: December 2018



Please enter your comment!
Please enter your name here