Reputation and the Accountability Gap

In her recent article on the reputation of international organizations (IOs), Kristina Daugirdas concludes that reputation's constraining effect has some serious shortcomings in the context of sexual exploitation and abuse (SEA).¹ This essay extends those conclusions to recent mass torts cases against IOs. In particular, it argues that member states and IOs have independent and overlapping concerns that have contributed to devaluing the relevance of a “good reputation,” particularly when it comes to providing compensation for wrongful conduct. IOs, it seems, do not want to develop a reputation for deep pockets. Nonetheless, this essay also demonstrates that when compensation is not at issue, there are instances in which reputation matters to IOs. It concludes by discussing recent cases related to responsibility and organizational immunities and suggests that the trend of narrowing immunities may change the reputational calculus for IOs and member states significantly.

Reputation Matters When There is No Claim for Compensation

As a preliminary matter, reputation has been a driving force behind institutional developments in some situations. For example, the International Criminal Court (ICC) was widely criticized for its “Africa bias” based on its failure, in its first decade of existence, to investigate situations in areas other than Africa.² Facing mounting member state withdrawals from the Court, the ICC eventually expanded its investigations to other regions. Although legal criteria in the ICC's statute determine the scope and admissibility of the new issues under consideration, there was a widespread perception that a change in prosecutorial strategy was necessary in order to protect the Court's legitimacy.³ The various signals of selective justice, neocolonialism, and perhaps even perceptions of racism became deeply problematic for the Court's reputation, even though there was a debate about the accuracy of the allegations.⁴ One would be hard pressed to suggest that the ICC was not motivated to change its geographical focus based on reputational concerns. Significantly, however, it was never alleged that the ICC committed an internationally wrongful act or that there was a basis for a compensation claim.

Notwithstanding the ICC example, reputational concerns do not always prompt IOs to alter their behavior. For instance, Daugirdas's conclusion that reputational sanctions have been generally ineffective in the SEA context can be extended to the UN's handling of recent torts cases.⁵ Although the UN's policy has long been to compensate victims for wrongful acts,⁶ it has recently failed to do so in the context of large torts cases.⁷ In fact, the UN's recent responses to claims make clear that both it and a number of influential states have sought to avoid creating a reputation for compensating victims.⁸

For example, in March 2017 UN Secretary General António Guterres expressed his “profound regret” for the suffering of Roma populations in Kosovo, who the UN resettled in camps contaminated with lead.⁹ However, he stopped short of making a formal apology, admitting responsibility, or offering compensation.¹⁰ His language and approach were almost identical to that used by former Secretary General Ban Ki Moon in addressing the UN's handling of the cholera crisis in Haiti in 2016.¹¹ Despite Ban Ki Moon's statement on Haiti that “[t]he United Nations should seize this opportunity to address a tragedy that also has damaged our reputation and global mission,” reputational concerns did not then lead to a different outcome.¹² The UN has not yet implemented a 2016 decision by the Human Rights Advisory Panel that the UN does owe compensation and should make an apology to the Kosovar victims.¹³ Nor has it ever admitted responsibility or paid compensation to victims of cholera in Haiti. At the time of writing, the Haiti Cholera Fund has only 10 percent of requested contributions and the Kosovo trust fund remains unfunded.¹⁴ As José Alvarez observed, the UN behaves as though it is immune not only in national courts, but also in the court of public opinion.¹⁵

In contrast, the UN has addressed corruption quite differently. In the Oil-For-Food scandal, UN officials, companies, and politicians from a number of countries were implicated in taking kickbacks while implementing the sanctions regime in Iraq.¹⁶ The revelations of corruption triggered an independent inquiry led by Paul Volcker, and produced significant reforms to the UN's internal procurement process and oversight mechanisms.¹⁷ In a subsequent U.S. prosecution of a UN staff member on fraud and corruption charges, the United Nations waived the immunity of its employee, one of the few instances in which it has done so.¹⁸ It is important to note, however, that this waiver did not implicate the organization's responsibility itself, but only one of its personnel. Nonetheless, one must distinguish the different outcomes with regards to corruption and SEA. The UN does not exercise general criminal jurisdiction over individuals or states, and has the discretion to forward both types of cases to national jurisdictions for prosecution. The most likely explanation for the stronger reaction to corruption is that SEA usually occurs in the field by peacekeepers, whereas corruption stems from financing or procurement problems at headquarters, where the UN exercises a higher degree of control.

Responsibility Rules Create Incentives for States and IOs to Ignore Reputational Costs

It is well established that the UN has independent legal personality and the capacity to bear rights and obligations.¹⁹ Despite the development of the Articles on Responsibility of International Organizations (ARIO), there is still a considerable lack of clarity about what constitutes an internationally wrongful act by an IO, and how domestic tort or contract law might apply.²⁰ Current rules create incentives for states to push responsibility to IOs where both are potentially implicated in wrongful activities due to what has been called the “fiction” of exclusive attribution.²¹ The rules on organizational immunity have often protected the IO when this has “push” been successful. Because IOs do not have taxing powers, they remain highly—sometimes exclusively—financially dependent on member state contributions.²² Indeed, control (including budgetary control) of member states over IOs has both led to member states hiding behind the immunities of IOs, and IOs attempting to avoid responsibility because they lack independent income streams.²³

The case of Behrami & Saramati illustrates this push-and-pull dynamic, in which member states shift responsibility to the IO, and the IO then invokes jurisdictional defenses to avoid suit. In Behrami, France and Norway used the doctrines of attribution to argue that the UN was responsible for wrongful acts that French and Norwegian troops had committee in Kosovo.²⁴ The European Court of Human Rights (ECtHR) agreed, and in an admissibility decision, determined it had no jurisdiction and dismissed the case.²⁵ While the ECtHR wrongly applied an “ultimate authority and control” test²⁶, and excluded the possibility of dual attribution, the successful allocation of responsibility to the UN met the states’ goal of avoiding it. However, there were few reputational consequences for either the states themselves or the UN in this strategy.²⁷

Some academics and courts have tried to work around these push-and-pull forces by generating concepts of shared responsibility. The general principle in the law of international responsibility is independent responsibility, such that each state or IO is only responsible for its own wrongful conduct. However, both the Articles on Responsibility of States for Internationally Wrongful Acts and ARIO leave open the possibility for concurrent or secondary attribution of responsibility to multiple actors.²⁸ While the current rules acknowledge (but do not develop) concepts of shared responsibility, there are an increasing number of situations in which states act interdependently.²⁹

In the Mothers of Srebrenica case, which challenged the failure of Dutch troops under UN command to protect the Muslim enclave of Srebrenica against attack by the Bosnian Serb army, a Dutch district court found that although the UN was immune, the Dutch state might nonetheless be responsible. On appeal, a court ordered the Dutch government to compensate the victims.³⁰ Here, the UN avoided gaining a reputation for making compensation through immunity rules, although the state was not so protected in its own courts. Importantly, in Al-Jedda v. United Kingdom, the UK Supreme Court found that the acts of soldiers within the Multinational Force in Iraq did not cease to be attributable to troop-contributing countries even if they were also attributable to the UN, laying further judicial grounds for shared responsibility.³¹ It is perhaps not surprising, however, that neither states nor IOs have embraced concepts of shared responsibility to date in any binding form, as it serves to increase their exposure to suit.

How Immunities Can Change the Reputational Calculus

It is therefore significant that courts are highly engaged with issues of organizational immunity at present. The Second Circuit's decision in Georges v. United Nations involving the cholera case follows the trend of upholding absolute UN immunity.³² Nonetheless, it is notable that the U.S. Supreme Court recently chose to limit the immunity of a different IO—the International Financial Corporation, which is part of the World Bank Group. Indeed, the Supreme Court's decision in Jam v. IFC interprets the immunities conveyed in the International Organizations Immunities Act as tracking those in the Foreign Sovereign Immunities Act, such that exceptions applicable to states, including the commercial activities exception, would apply to IOs designated under the Act.³³ This case demonstrates that the rules of immunity may be starting to change: some courts have read down IO immunity where third parties are denied access to justice within an IO's internal accountability mechanism.³⁴ Although there are only a few cases to suggest this direction at present, a narrowing of IO immunity may in turn change the calculus on reputational costs. If IOs assess that they may eventually have to pay, they will be more careful in their actions in order to protect their reputations.

Like legitimacy, reputation is not a zero-sum game: when one entity loses it, the other does not automatically benefit. Nonetheless, because the current rules on responsibility remain incomplete,³⁵ states and IOs have much to gain by using the rules of attribution to shift responsibility to IOs and then assert immunity. This strategy is further rewarded because there is a widespread understanding that one cannot find member states residually responsible for the wrongful acts of an IO.³⁶ Thus, current rules of responsibility and immunity contribute to the reputational impasse, particularly where a claim for compensation is the end goal.