Legitimacy, evolution, and growth in investment treaty arbitration: empirically evaluating the state-of-the-art.

Author:Behn, Daniel
Position:III. An Empirical Evaluation of Recent Decisions D. Dispute Subject Matter through IV. Future Evidence-Based Research in Investment Treaty Arbitration, with footnotes, p. 389-415
  1. Dispute Subject Matter

    As to the subject matter of disputes, investment treaty arbitration has historically been used primarily in cases involving the extractive industries. These types of projects are typically large, long-term, expensive, and tied to the state where the project occurs. This means that these types of investments frequently have large sunk costs that diminish the bargaining power of the foreign investor as the project progresses. In some cases, this shift in bargaining position has allowed host states to make midstream changes to the regulatory and legislative regime governing a particular foreign project. These types of cases (in the extractive industries sector) still represent a large percentage of the caseload in the data set. However, there is also a significant increase in cases arising out of the manufacturing and service sectors. Diversification into these types of disputes can assist in contributing to the legitimacy of investment treaty arbitration by signaling that it is not a system of adjudication designed for use by particular economic sectors (especially such sovereignty-sensitive sectors as the extractive industries).

    Using the same coding as the ICSID Caseload, the data set shows the following distribution of dispute types: of the 147 cases, there are four cases in agriculture, fishing, and forestry; eight cases in construction; twenty cases in electric power; nine cases in finance; five cases in information and communication; thirty-seven cases in the extractive industries; twenty-one cases in other industries; twenty-three cases in services and trade; five cases in tourism; seven cases in transportation; and eight cases in water and waste.

    According to the most recent ICSID Caseload, (72) the percentage of different subject-matter disputes in all ICSID cases initiated through mid-2014 equals (with a comparison of the percentages in the data set in parentheses): 4% (3%) in agriculture, fishing, and forestry; 7% (7%) in construction; 13% (12%) in electric power; 7% (11%) in finance; 6% (4%) in information and communication; 26% (23%) in the extractive industries; 13% (12%) in other industries; 4% (11%) in services and trade; 4% (2%) in tourism; 10% (7%) in transportation; and 6% (6%) in water and waste.

    The percentages of different subject-matter types of disputes in the data set are remarkably close to those in the ICSID Caseload. However, there is one key exception. There is a noticeable increase in disputes in the services and trade sectors; and it is this area that may indicate some diversification of the types of disputes being adjudicated in recent investment treaty arbitration cases. This category of investments is not 72 one that has featured historically in prototypical foreign direct investment disputes. The rise in service and trade sector disputes over the past few years could be attributable to macro factors like increased interdependence and globalization, or it could be attributable to factors associated with investment treaty arbitration specifically. (73)

    Regardless of the exact reasons for this increased diversification, there are both negative and positive legitimacy issues that could arise. On the positive side, dispute type diversification may signal that there is less evidence of structural bias in favor of particular types of disputes (e.g., those in the extractive industries). On the negative side, dispute type diversification could signal an expansion of investment treaty arbitration at a time when states may be attempting to reduce this type of dispute resolution in newly negotiated IIAs. (74) While there does appear to be an increase in service and trade sector disputes in the data set, there continues to be a large percentage of cases in the extractive industries sector and in the electric power sector. Combined, these two sectors constitute fifty-seven of the 147 (39%) disputes in the data set, which is identical to the historical percentages indicated in the latest ICSID Caseload. (75) In addition to the frequency of cases in these two sectors, it may also be relevant to note that thirty-two out of the fifty-seven disputes (56%) in these two sectors were brought by extra-large MNEs; and that seventeen of the twenty-four disputes (71%) involving companies on the Global 500 list are in the extractive industries or electric power sectors.

    To summarize, a majority of the largest companies to bring disputes in the data set are energy and extractive industry companies, and the majority of disputes in these two sectors are being initiated by some of the largest companies in the world. While this empirical evidence is likely to be viewed as unsurprising, it does indicate that there continues to be a diversity issue in investment treaty arbitration in regard to both the size of claimants and the economic sector in which these companies operate.

    In regard to outcomes, there are a few identifiable deviations from the average success rates for all resolved cases in the data set. While tourism and transportation cases only account for eight cases in the data set, their combined success rate is 75%. These two sectors appear to have been very successful in litigating investment treaty cases (at least among the eight cases in the data set). It also appears that cases in the extractive industries are more successful (42% success rate) than the overall win rate among all cases in the data set (30%). In terms of less successful claimant sectors, the construction sector did not win a single case in the data set and cases arising out of the finance and banking sectors were successful in only one out of nine cases in the data set.

    Finally, it is interesting to note the sectors that may be more or less prone to discontinuing or settling cases. In cases arising out of the communication and information sector, three out of the four cases settled (75%); and in cases arising out of the finance and banking sectors, 45% of the cases were either discontinued or settled, which is higher than the overall average in the data set (30%). Overall, it does appear that cases arising out of distinct economic sectors may be more or less likely to be successful. However, the main caveat here is the limited size of the data set and that it is difficult to make sweeping conclusions about some sectors that were only represented by a few cases in the data set.

  2. Geographic Distribution

    Discussions about the geographic distribution of investment treaty arbitration disputes have often acted as a proxy for issues relating to the development status of states having to respond to IIA claims. It is a proxy because, for the most part, most early (pre-2000) investment treaty arbitrations were brought by companies or individuals in the developed world against states in the developing world. (76) Due to the fact that there are few developed states outside of Western Europe and North America, most of the claims originate in these two regions; and while there is a greater opportunity for diversity in the geographic distribution of respondent states, there is a perception that investment treaty claims are targeting developing states, but only developing states in certain parts of the world. (77) From a legitimacy perspective, the targeting of particular regions or states by claimant-investors can be problematic. In the case of South America, perceived targeting can result in exit from the regime altogether; and in the case of Ecuador, Argentina, Bolivia, and Venezuela, this is exactly what appears to be happening. (78)

    While a diversification in the geographic distribution of states responding to investment treaty arbitration claims is possible, diversification away from claimants in Western Europe and North America may be more difficult. It will be more difficult because of the structure of investment treaty arbitration as it is currently practiced. IIAs have traditionally been signed between one developed state and one developing state. While there are exceptions to this rule, (79) this type of arrangement constitutes the majority of IIAs in force. Due to the fact that the amount of outward investment between these two types of states (developed capital-exporting states to developing capital importing states) is stark, the reciprocal nature of IIAs is largely theoretical. From a legitimacy perspective, these numbers on geographic distribution in the data set indicate that investment treaty arbitration is a system of adjudication that is being used primarily by claimants from a few regions (Western Europe and North America) against states from a few (albeit different) regions. (80) This kind of limited geographic scope in investment treaty arbitration can lead to the perception that this is not a general system of adjudication, but rather is designed in a way that provides remedies only for claimants originating in certain parts of the world (i.e., companies in Western Europe and North America) against certain respondent states (i.e., less developed states).

    Of the 147 cases in the data set, the claimants come from thirty-three different states, and the majority of claimants come from Western Europe and North America: (81) there are eighty-two claimants from Western Europe, thirty-six claimants from North America, thirteen claimants from the Middle East and North Africa, eight claimants from Eastern Europe and Central Asia, five claimants from South and East Asia and the Pacific, two claimants from South America, one claimant from Central America and the Caribbean, and no claimants from Sub-Saharan Africa.

    In terms of outcomes, there are a couple patterns worth noting. Cases involving Western European claimants are almost identical to the overall averages for all resolved cases in the data set. In other words, even though Western European claimants represent a majority of all resolved cases in the data set, they do not seem to be more or less likely to succeed in cases when compared...

To continue reading