The legitimacy of the World Bank's dispute resolution body - The International Centre for the Settlement of Investment Disputes (ICSID) - is a matter of heated debate. Some states have alleged that ICSID is biased, withdrawn from the ICSID Convention, and advocated creating alternative arbitration systems. Using pre-2007 archival data of the population of then- known arbitration awards, this Article quantitatively assesses whether ICSID arbitration awards were substantially different from arbitration awards rendered in other forums. The Article examines variation in the amounts claimed and outcomes reached to evaluate indicators of bias. The results indicated that there was no reliable statistical relationship between ICSID arbitrations and either amounts claimed or ultimate outcomes. The results generally did not show a statistical difference when controlling for (1) the presence of an Energy dispute, (2) the presence of a Latin American respondent, or (3) the respondent's Development Status. Nevertheless, although outcomes were not statistically different for Latin American and non-Latin American respondents, amounts claimed against Latin American states were higher - but only for non-ICSID arbitration. While the arguably higher initial arbitration risk may contribute to concerns related to perception of bias, the results provide initial evidence that those criticisms may have been misattributed to ICSID. Results suggested, on the whole, that ICSID arbitration awards were not statistically different from other arbitral processes, which is preliminary evidence that ICSID arbitration was not necessarily biased or that investment arbitration operated in reasonably equivalent ways across forums. Caution about this finding is appropriate given the size of the pre-2007 population and as one analysis suggested that for the subset comprised only of ICSID Convention awards as compared to all other awards (including ICSID Additional Facility awards), awards against Low Income respondents were statistically higher than awards against High Income respondents. Qualitative commonalities in that small subset of awards revealed the presence of certain types of law firms (or the lack thereof) or recent civil war in African states. In light of the initial quantitative findings for a pre-2007 population of arbitration awards, but recognizing the need for replication and methods to facilitate qualitative and normative assessments of ICSID, this Article concludes by suggesting that there may be value in implementing tailored reforms and structural safeguards to address arguable concerns of bias, improve the management of international economic conflict, and minimize a potential backlash to the international investment system.
The ICSID Effect? Considering Potential Variations in Arbitration Awards,
Virginia Journal of International Law
Available at: https://digitalcommons.wcl.american.edu/facsch_lawrev/964