After the three in-depth case studies conducted in Chapters 4–6, it is time to zoom out. In the following I present the main findings across the case studies in section 7.1. Based on those findings, I then advance a couple of hypotheses that call for future research in section 7.2. Finally, in section 7.3, I conclude by considering the broader implications of this study – including the issue of how IO accountability mechanisms should be reformed.
7.1 Key Findings and Overall Conclusion
I will first discuss the findings across the three cases using the schema of access, participation, neutrality, and outcome. Thereafter, I offer an overall conclusion to the research question set out in the introduction.
7.1.1 Access
The normative yardsticks for access, which are chiefly derived from the right to remedy, center on jurisdiction and admissibility. As explained in Chapter 3, section 3.3.2.1, an accountability mechanism is only accessible to the extent that it has jurisdiction and does not impose excessively restrictive admissibility criteria. Access must be direct, meaning that an aggrieved individuals must be able to submit complaints in their own name. The accountability mechanism must have jurisdiction ratione personae over the IO as such, and aggrieved individuals must be able to put their complaints directly to the accountability mechanism. The system of accountability mechanisms applicable in a particular context must also have jurisdiction ratione materiae and loci over all plausible human rights violations.
There is pronounced variation between the case studies with regard to access. The UNHCR case distinguishes itself by the total lack of access to any accountability mechanism with jurisdiction over the organization itself. Domestic courts are inaccessible because of UNHCR’s jurisdictional immunity, and the two other accountability mechanisms – the Inspector General’s Office and the Officer of Internal Oversight Services – only have jurisdiction ratione personae over UNHCR staff, not the organization as such.
Conversely, in the EU and ICC cases, individuals do have access to multiple accountability mechanisms for grievances against the organizations as such. With regard to the ICC case, the principal accountability mechanism is the administrative appeals procedure. Detained accused may also bring their complaints before the (Pre-)Trial or Appeals Chamber presiding over their case. Both of these have jurisdiction ratione personae over the ICC. A third accountability mechanism, the Independent Oversight Mechanism, is accessible, but lacks jurisdiction ratione personae over the ICC itself.
In the EU case, domestic courts are accessible. In the specific sub-case of the EULEX Kosovo mission aggrieved individuals additionally have access to the Human Rights Review Panel. The European Ombudsman also plays a role, due to the informal practice of complaints-based own-initiative inquiries. All these accountability mechanisms have jurisdiction ratione personae over the EU.
Looking at the findings from a different angle, by type of accountability mechanism, a cross-case pattern emerges: none of the accountability mechanisms of the internal oversight type provide sufficient access. The reason is structural. The right to remedy requires that the accountability mechanism has jurisdiction over the human rights violator, that is the IO itself. But internal oversight mechanisms may at best hold an IO to account indirectly, in the sense that staff misconduct implies organizational responsibility.
In all three cases there are examples of inaccessible accountability mechanisms, such as domestic courts in the UNHCR and ICC cases, and the CJEU in the EU case. When accountability mechanisms are inaccessible to aggrieved individuals, it is unnecessary to assess them with regard to the participation, neutrality, and outcome.
7.1.2 Participation
The normative yardsticks for participation, laid down in Chapter 3, section 3.3.2.2, are derived from procedural justice research. In brief, procedural justice research shows that people “feel more fairly treated if they are allowed to participate in the resolution of their problems or conflicts by presenting their suggestions about what should be done.”Footnote 1 An IO accountability mechanism should therefore give aggrieved individuals the opportunity to tell their side of the story. Decisions should also be reasoned, so that the aggrieved individuals can tell whether their view has been taken into account – regardless of whether the decision is positive or negative.
The degree of participation varied more by type of accountability mechanism than by case. Most types of accountability mechanisms – ombudspersons, inspection and review panels, administrative appeals procedures, and courts – offer high levels of participation. All accountability mechanisms of these types assessed in the case studies provide aggrieved individuals with opportunities to make their voice heard. The decisions of those accountability mechanisms are also reasoned, so that it is possible for said individuals to verify that their views were taken into account.
At the other end of the spectrum, internal oversight mechanisms again distinguish themselves negatively. An individual alleging to be a victim of human rights violations is not party to the case before the internal oversight mechanism. They are on the sidelines, at best having the status of witnesses to the alleged misconduct. Moreover, individuals are usually given little information about the progress and outcome of the case – making it difficult for them to verify whether their views have been taken into account. The lack of participation was roughly equivalent for all three internal oversight mechanisms.Footnote 2
7.1.3 Neutrality
As explained in Chapter 3, section 3.3.2.3, there are two prerequisites for neutrality: independence and impartiality. Both aspects are emphasized in procedural justice research, and integral to the right to remedy. An IO accountability mechanism should thus be institutionally independent, and have procedures for appointment and dismissal of its members that sufficiently protect the independence of the mechanism. Additionally, the accountability mechanism must be impartial, meaning that it cannot be or appear biased from an objective, external point of view.
With regard to independence, the findings across all three case studies are broadly positive. Virtually all the accountability mechanisms assessed are, legally speaking, institutionally independent. The one exception is the ICC’s Independent Oversight Mechanism, which is required to involve the relevant Head of Organ before taking the decision to investigate one of that organ’s staff members.Footnote 3
The procedures for appointment and dismissal of members of the accountability mechanisms – while variable – generally appear to protect the independence of the accountability mechanisms. The only exception is the EULEX Human Rights Review Panel, whose short-term lengths and rules on reappointment arguably encroach upon its independence.Footnote 4
However, although most accountability mechanisms are, formally speaking, independent, some suffer from a lack of actual impartiality. Serious allegations of bias have been made against both the internal oversight mechanisms applicable to the UNHCR case; the Inspector General’s Office and the Office of Internal Oversight Services.Footnote 5 The third internal oversight mechanism assessed in the context of the three case studies, the ICC’s Independent Oversight Mechanism, has not been subject to similar allegations of bias. But any conclusions drawn from the Independent Oversight Mechanism’s practice may be premature since it only became fully functional in late 2015 and has yet to complete more than a handful of investigations.Footnote 6
In sum, the findings with regard to neutrality are mixed. On the one hand, they suggest that the UNHCR case is an outlier since its accountability mechanisms suffer from a lack of neutrality. On the other hand, it might well be the case that internal oversight mechanisms are, due to their nature or structure, particularly prone to improper interference from the organization they serve. If so, the lack of impartiality identified in the UNHCR case is only a symptom of a broader problem, rather than the result of a negative culture within that particular organization. Further research is needed to fully answer these questions.
7.1.4 Outcome
The normative yardsticks for outcome, laid down in Chapter 3, section 3.3.2.4, are primarily derived from the right to remedy. A fundamental requirement is that, within a system of IO accountability mechanisms, there must be at least one mechanism empowered to issue binding decisions. Moreover, the system of accountability mechanisms must be able to prevent a violation, stop a continuing violation, or offer other forms of redress to the victims of human rights violations. Redress can come in different forms. Declaratory relief is the most basic but it may not be sufficient redress. Restitution, if possible in the circumstances, may be sufficient redress. Compensation, monetary or otherwise, is sometimes required – at least in cases involving violations of the right to life or the right to freedom from torture and other inhuman treatment.
The internal oversight mechanisms again distinguish themselves negatively as their only outcomes are nonbinding investigation reports determining whether particular staff members have committed misconduct, without pronouncing on the responsibility of the IO. The European Ombudsman and the EULEX Human Rights Review Panel – mechanisms of the closely related types of ombudspersons and inspection and review panels – also offer merely nonbinding outcomes. But although their decisions are nonbinding, they determine whether the IO is responsible, thus offering declaratory relief. Additionally, both may recommend restitution, and the European Ombudsman can even recommend compensation.
Nonbinding outcomes are, however, not sufficient under the right to remedy. Accountability mechanisms with the power to issue binding decisions are completely lacking in the UNHCR case. Moreover, the accountability mechanisms that are accessible lack the power to recommend restitution and compensation. In the EU case, the domestic courts of EU member states may render judgments. But the bindingness of such judgments is based on domestic law, thus rendering the binding nature of such judgments more theoretical than practical, given the weak position of domestic courts vis-à-vis the Union and the Union’s immunity from enforcement.Footnote 7
Among the three case studies it is only the system of accountability mechanisms applicable in the ICC case that offers individuals binding outcomes. Two of the accountability mechanisms available to the detainees – the administrative appeals procedure and the (Pre-)Trial and Appeals Chambers – have the power to issue binding decisions. Moreover, in addition to thus offering declaratory relief, they may also order restitution. None of them may offer monetary compensation, though but the (Pre-)Trial and Appeals Chambers may in principle offer sentence reduction to detainees that are later convicted.
7.1.5 Overall Conclusion
The sum of the findings that emerge from the case studies is discouraging. In none of the three case studies do the existing accountability mechanisms – whether assessed separately or collectively – fulfill all the normative requirements derived from the right to remedy and procedural justice research. The answer to the research question presented in the introduction is therefore that the existing accountability mechanisms are insufficient in all of the three cases studied in Chapters 4–6.
7.2 Hypotheses Generated by the Case Studies
It is not possible to draw inferences from three case studies that are statistically generalizable to the universe of IOs and IO accountability mechanisms. But what case studies do excel at is hypothesis generation.Footnote 8
In the following sections I present a few hypotheses generated by the three case studies, and which may hopefully inspire further studies. These hypotheses revolve around the great variations observed, both between the systems of accountability mechanisms applicable to the three case studies and between the different types of accountability mechanisms.
7.2.1 Variations between Cases
As the findings above show, there is significant variation between the three cases. The most deficient system of accountability mechanisms is that applicable to the UNHCR case. Apart from internal oversight investigations against individual UNHCR staff members, there are no accountability mechanisms to speak of.
At the other end of the scale, we find the ICC case, where there are three types of accessible accountability mechanisms; an administrative appeals procedure, an internal oversight mechanism, and the court itself (Chambers). The administrative appeals procedure is key as it is easily accessible to all detainees, offers strong rights of participation, sufficiently independent and impartial, and can provide binding outcomes in the form of declaratory relief or restitution orders.
Somewhere in between we find the accountability mechanisms applicable to the EU case. While several accountability mechanisms are accessible in the context of the EU case, the potential outcomes are generally weak. That being said, this case also demonstrates that nonbinding outcomes can have an impact. Although the European Ombudsman and the EULEX Human Rights Review Panel are only empowered to issue nonbinding recommendations, they are generally complied with, likely due to the legitimacy of the ombudsman and the panel in the eyes of the Union and its organs.
The potential victims of human rights violations in the three cases are three distinct groups of persons: populations in conflict-ridden areas, refugees in camps, and ICC detainees. From this perspective one can observe that those provided with the most complete system of accountability mechanisms are the handful of detainees (and in particular those accused of international crimes)Footnote 9 in the ICC Detention Centre. In contrast, vulnerable refugees, who are stowed away in camps by the tens of thousands, have least access to human rights accountability mechanisms.
What may explain such variations? There is no obvious and unified pattern. One possible explanation – at least for the chasm observed between refugees in UNHCR-administered camps and ICC detainees – can be summarized with the dichotomy aid vs. punishment. When establishing an IO meant to punish offenders, states readily appreciate the need for human rights accountability mechanisms. On the other hand, when an IO is tasked with providing aid – and thus has the protection of human rights as its direct aim – the risk of violating (other) human rights may not be as obvious. There may also be a particularly strong normative bias in favor of humanitarian IOs like the UNHCR, which lends them an aura of being inherently good that blinds us from perceiving them as harm-doers.Footnote 10
Alternatively, the explanatory variable may simply be time. The three case studies appear to suggest that the issue of establishing accountability mechanisms is only put on the agenda when IOs are established, or when they are thoroughly reformed. The least sufficient system of accountability mechanisms is that applicable to the UNHCR case.Footnote 11 The UN, which the UNHCR forms part of, has not been significantly reformed since its establishment in 1945.Footnote 12 In contrast, the two other case studies – where the insufficiencies were minor in comparison – concern one IO established less than twenty years ago (the ICC, in 1998) and one that was recently reformed (the EU, in 2007).Footnote 13 Further research is needed to test these hypotheses.
7.2.2 Variations between Different Types of Accountability Mechanisms
The findings from the three case studies show that, at least in some respects, the key dividing lines are not between the different cases, but between the different types of accountability mechanisms. In particular, some types appear to be inherently incapable of providing sufficient accountability.
For example, all three internal oversight mechanisms fail to measure up to the normative yardsticks established above in Chapter 3, section 3.3.2. They are not sufficiently accessible, due to their lack of jurisdiction ratione personae over the organization itself, nor are they capable of offering appropriate relief to victims of human rights violations. As already mentioned, these deficiencies appear to be structural, and thus endemic to internal oversight mechanisms.
The European Ombudsman and the EULEX Human Rights Review Panel lack the power to issue binding decisions. This is a typical attribute of the two closely related types of accountability mechanism that they represent: ombudspersons and inspection and review panels.Footnote 14 At the same time, this is a deficiency; a critical flaw preventing them from providing adequate redress for human rights violations.
The other types of accountability mechanisms applicable to the three case studies also fail to provide sufficient accountability – even as a collective. Yet the reasons do not appear to be related to the structure or nature of those accountability mechanisms, and thus endemic to them, but rather to the particulars of how they are implemented in each case. In contrast, even a best-in-class internal oversight mechanism, ombudsperson, or inspection and review panel would fail to fulfill the normative requirements set out in Chapter 3, section 3.3.2, due to, for example, the lack of binding outcomes.
One may therefore posit the following hypothesis: systems of accountability mechanisms consisting exclusively of internal oversight mechanisms, ombudspersons, and/or inspection and review panels are never sufficient; additional types are needed. That said, I am not suggesting they should always be replaced by other types of accountability mechanism. It is likely better to supplement or strengthen them. For example, an ombudsperson may be a useful as part of a system of accountability mechanisms, by providing a low-threshold avenue for complaints with a conciliatory ethos. But, if an ombudsperson’s recommendations are not followed, aggrieved individuals should be able to turn to a different accountability mechanism that can provide appropriate redress.
7.3 Broader Implications: Sketching an Agenda for Reform
The above conclusions and hypotheses signal a need for reform. While the findings from the three case studies are obviously not statistically generalizable to the universe of IOs, I submit that they nevertheless indicate a need for reforming the human rights accountability mechanisms of IOs generally.Footnote 15 It is particularly the nature of the three cases that support this view. In addition to meeting the case selection parameters set out in the introduction,Footnote 16 they constitute examples of situations where there are good reasons to expect a sufficient level of accountability. But, as we have seen, these expectations are not met in any of the cases.
In the context of UNHCR refugee camp management, where some of the world’s most vulnerable people are under the de facto authority of the organization, one should expect a high level of accountability. However, there is virtually no human rights accountability mechanisms. In the EU case expectations of accountability follow, in particular, from the general perception of the Union having the most advanced system of accountability mechanisms of any IO. But, while this may be true, even the Union’s accountability mechanisms are insufficient in certain contexts – such as this study demonstrates with regard to CSDP missions. At the ICC Detention Centre, where detainees are kept on European soil under the ICC’s exclusive authority, one would expect European levels of human rights accountability. But also in that case the human rights accountability mechanisms are lacking.
The human rights accountability mechanisms applicable in these three cases are thus insufficient, despite the fact that each of them represent a situation where there are strong reasons to expect a sufficient level of human rights accountability. Given that, it seems fair to assume that the cases are indicative of a broader pattern of subpar accountability – and consequently that reform is needed.
That there is a more general need for reform of IO accountability mechanisms is also increasingly being recognized in the literature.Footnote 17 In the words of Klein:
In view of the fact that [individuals] are much more frequently affected than states or international organizations by situations where the international responsibility of an organization is at stake, the creation of mechanisms allowing for the invocation of such responsibility by [individuals] definitely emerges as one of the most significant challenges in this area for the years to come.Footnote 18
The debate has indeed shifted significantly since the second half of the twentieth century, when IOs were near-universally perceived as forces for good. True, there were admittedly some cracks in this flattering portrait – in the form of criticism against organizations operating in the field of international trade and finance, particularly from the 1980s onward.Footnote 19 But the bulk of IOs were perceived as inherently good.
Particularly the 1990s – the euphoric decade after the end of the cold war – were characterized by an optimistic attitude to the establishment of a true international community. Multilateralism and a rules-based world order were essential components of this project. That this implied that the prominence of states had to fade, and that IOs would assume some of their tasks and powers, was seen as both natural and beneficial. If one was to build a world order based on law and justice rather than naked power, there was no other option.
The contrast to the present day – the eve of the second decade of the twentieth-first century – is stark. Attitudes toward a multilateralism, in the sense of a rules-based international community, have changed dramatically since the turn of the millennium. Key proponents and benefactors of multilateralism during the 1990s are now instead emphasizing sovereignty and bilateralism. It is not difficult to find examples demonstrating this shift: The Doha round of World Trade Organization negotiations, which begun in 2001, have been stalled for over a decade. The current US president has proclaimed “America first” as his foreign policy strategy. The United Kingdom is in the process of withdrawing from the EU, purportedly to reassert its sovereignty. Poland and Hungary, two states that in the 1990s liberated themselves from the iron grip of the USSR and embraced multilateralism, are now among its chief critics. At the same time, new forms of IOs are emerging, which challenge rather than encourage multilateralism. The Shanghai Cooperation Organization is a case in point. It is composed of six nondemocratic Eurasian states that have joined forces to, in particular, promote the principles of noninterference and state sovereignty.Footnote 20
These changes have happened fast, and appear to be accelerating. When I first began researching IO accountability back in 2012, conducting a critical study of IOs still appeared somewhat contrarian. Now in 2019, however, a critical study of the accountability of IOs appears perfectly aligned with the new wave of skepticism toward multilateralism.
But I do not think the findings that emerge from this study entails a challenge to IOs as such or multilateralism more generally. These findings should rather be seen as constructive criticism, which will hopefully contribute toward triggering a reform of the current system of IO accountability mechanisms.
Reform is needed to enhance the legitimacy of IOs and multilateralism more broadly. One way of building legitimacy is to ensure that IOs are sufficiently accountable – also in the field of human rights. As procedural justice research shows, reforming the mechanisms of accountability is key to increasing legitimacy. That is because people determine the legitimacy of power-wielders “primarily by assessing the fairness of their decision-making procedures. Hence, using fair decision-making procedures is the key to developing, maintaining, and enhancing the legitimacy of rules and authorities.”Footnote 21
The difficult question is not whether reform is needed, but what it should look like.Footnote 22 This study’s methodology is better suited to offering a diagnosis than a cure. Further research, as well as involvement of stakeholders, is needed to iron out more concrete reform proposals. In the following I will nevertheless attempt to sketch out some contours of reform, based on the experience gained from this study. First, in section 7.3.1, I argue against the idea that the jurisdictional immunity of IOs should be curtailed, so that domestic courts may act as IO accountability mechanisms. Second, in section 7.3.2, I end with some preliminary observations on what reform of IO accountability mechanisms at the international level should look like.
7.3.1 Enhancing Accountability by Curtailing Jurisdictional Immunity?
I stand with those opposing radical changes to the current regime of IO immunities. The jurisdictional immunity of IOs is a necessity in order to “preserve and ensure the independence of the organization, and to enable it to fulfil its functions,” which could otherwise be compromised by unwarranted interference from states.Footnote 23
Although immunities remain essential to protect IOs from unwanted interference, it is true that interference is not a binary matter but rather a spectrum.Footnote 24 As Reinisch argues, allowing an individual to bring a claim for compensation against an IO for human rights violations appears to interfere less with the organization’s functioning than, for example, allowing an injunction that seeks to prevent an organization from carrying out a particular peacekeeping operation.Footnote 25
This and similar arguments have been advanced in support of minor adjustments to the current regime of absolute immunity. One possible adjustment, suggested by Reinisch, is to subject IO immunities to a functional test, whereby the domestic courts should consider “the (anticipated) consequences of denying immunity” and that only lawsuits that “threaten [the] proper functioning” of the IO in question should be barred by immunity.Footnote 26 But it is doubtful whether this could be a workable test in practice, given its vagueness – partially due to the conceptual fuzziness of functionalism itselfFootnote 27 – and consequent susceptibility to abuse.
Another adjustment to IO immunities that has been suggested is to adopt a restrictive doctrine – for example by recognizing a distinction between acts jure imperii and acts jure gestionis – as has emerged in the law of state immunity.Footnote 28 However, there are both conceptual and practical difficulties with such a test. Conceptually, it is not clear how the quite state-centric concept of jure imperii conduct could be applied to IOs. And, from a practical perspective, human rights violations would probably often arise as a consequence of jure imperii conduct. A restrictive doctrine of IO immunities along these lines is therefore clearly incapable of solving the problem.
A third possible way of amending the current regime is to allow domestic courts to set aside the jurisdictional immunity of IOs if, on balance, upholding the immunity would entail a too significant impairment of the right of access to justice in that specific case.Footnote 29 But also this approach would be open to exploitation, since the application of such a balancing exercise would invariably lead to a large margin of discretion for domestic courts. Again there is obvious risks of abuse, as well as arbitrariness.
More importantly, even if a technically workable way of curtailing IO immunities was identified, a fundamental question remains to be asked: does the advantages of restricting IO immunities outweigh the disadvantages. The answer to this question appears to be no.
The advantages of restricting IO immunities are not as clear-cut as they seem at first glance. Importantly, one must keep in mind that immunity is only one of several impediments to access. As I explain in Chapter 3, section 3.2.8, domestic courts may still decline to exercise jurisdiction for other reasons. Difficult issues may arise with regard to jurisdictional competence, and domestic courts tend to employ various avoidance techniques to weasel out of cases involving IOs.Footnote 30 Moreover, if a domestic court does indeed exercise jurisdiction, it is quite possible that the organization would simply refuse to comply with the judgment rendered. This would in turn raise the issue of whether judgments could be enforced against the organization – both in law and in practice. There is also the risk of variations in practice among domestic courts, which could lead to legal uncertainty both for victims and IOs.
On the other hand, the disadvantages of restricting IO immunities appear quite clear and tangible. Restrictions upon the immunities of IOs appear to be negatively correlated with their independence. Unlike states, they have no territory and no population. Their tasks can only be carried out on the territory of states, by nationals of states.Footnote 31 Immunities are their only protection against improper state interference. It is thus difficult, if not impossible, to design a restrictive doctrine of IO jurisdictional immunity that allows the right suits and protects against the improper ones. Moreover, the jurisdictional immunities of IOs are, with few exceptions, firmly entrenched in treaties that provide unequivocally for absolute jurisdictional immunity. Restricting IO immunities is therefore no quick fix either.
Looking beyond these legal-practical arguments, limiting or removing the jurisdictional immunities of IOs implies a strengthening of state sovereignty, to the detriment of IOs. Such a trade-off would perhaps be acceptable if there was no better alternative. But I submit that it is possible to both ensure that IOs are not exposed to further risks of improper meddling by states, while simultaneously protecting the rights of individuals. That is by conducting reform at the international level.
7.3.2 Enhancing Accountability through Reform at the International Level
The main advantages of pursuing reform at the international level are clear: IOs retain their independence, while the accountability toward individual victims is ensured – provided that sufficient accountability mechanisms are established. The main challenges for reform at the international level are practical. As observed by Klein, IOs “have not proven keen to even consider the creation of such mechanisms and display a considerable degree of resistance toward such evolutions.”Footnote 32 Amendments to constituent instruments may even be needed, which would require agreement among all member states. But, as already alluded to, comparable difficulties are associated with restricting the immunities of IOs.
There is no single recipe for how such reform could be carried out, and I certainly do not purport to have a comprehensive reform plan. That said, it seems necessary to have at least some court-like mechanism that can issue binding decisions – it is in particular that which is lacking. This could be achieved by establishing internal courts, for example modeled after the international administrative tribunals that deal with disputes between IOs and their staff.Footnote 33
Another possible avenue of reform is the accession of IOs to international human rights courts and treaty bodies. This would have the dual advantage of not needing to set up new mechanisms and ensure an element of external accountability. The EU is currently the only IO which has taken steps in this direction. As mentioned in Chapter 3, the EU is already party to the CRPD and may join its optional protocol in the near future. Moreover, the EU is obliged to accede to the ECHR. Although the accession process was derailed by the CJEU in Opinion 2/13,Footnote 34 negotiations are set to resume in March 2020.Footnote 35 When completed the EU will set a new standard, one that other IOs could aspire to.
Court-like mechanisms should preferably be coupled with other accountability mechanisms. The exact configuration would have to be determined on a case-by-case basis. In some cases an administrative appeals procedure would be suitable, while in others an ombudsperson may be an appropriate accountability mechanism. What is important is that the system of accountability mechanisms lives up to the demands set by the right to remedy and people’s perceptions of procedural justice.
This may sound like utopian visions. But bold ideas are necessary precursors to bold action. Retaining the status quo will further damage the legitimacy that IOs are so dependent on. With time this is also likely to encourage domestic courts to disrespect the jurisdictional immunities of IOs, which in turn could enable undue interference. Given the unattractiveness of these alternatives, time is ripe for substantial reform at the international level.
IOs need to enhance their legitimacy. Reforming their system of human rights accountability mechanisms is not the only – or perhaps even the most effective – way that IOs can increase their legitimacy. But it would certainly be a step in the right direction.
