It is not business as usual in investment dispute resolution these days. In late April 2018 in New York, governments and experts met under the auspices of UNCITRAL Working Group IIIto continue vigorously debating how investor-state dispute settlement (ISDS) should be reformed or replacedby an investment court. This is not the first investment court proposal, however.

In the 1950s and 1960s, eminent international lawyers from around the world — Martin Domke, George Haight, F A Mann, Gunnar Lagergren, Elihu Lauterpacht, Raisa Khalfina, and Ignaz Seidl-Hohenveldern, to name a few — discussed an international investment court, notably at International Law Association conferences in 1958, 1960, and 1962. In 1960, ILA participants compared a court and arbitration directly, discussing “Draft Statutes of the Arbitral Tribunal for Foreign Investment” and “Draft Statutes of the Foreign Investments Court.”

Views in 1960, like today, varied sharply. No expert consensus existed that arbitration was better than a court for resolving investor-state disputes. This lack of consensus echoed even earlier debates: in 1905, when ASIL was founded, it was directed “exclusively to the interests of international law as distinct from international arbitration” (as Mark Mazower notes, page 92) because arbitration involved a compromise between interests rather than fidelity to the law. For ASIL founder Elihu Root, arbitration was an advance toward peace, but “the next advance to be urged along this line is to pass on from an arbitral tribunal…to a permanent court composed of judges who devote their entire time to the performance of judicial duties.” (Root might smile if he could see European officials coming to ASIL to discuss why an investment court should replace arbitration.) Unlike ASIL’s founders, arbitration’s supporters in 1905 praised the modesty of its procedures and goals—it was imperfect but feasible.

Feasibility was emphasized again in the 1960s, by attendees at the 1960 ILA conference and at the ICSID Convention’s drafting a few years later. This perception of feasibility stemmed in large part from assumptions made about arbitration in the 1960s, including that arbitral tribunals would not be agents of legal development, that appointing arbitrators was simple, and that arbitration was low cost. One can no longer make these assumptions about investor-state arbitration.

Today, reforming arbitration and creating a court are not mutually exclusive, nor are they only options under consideration at UNCITRAL. They are “elephants in the room” at UNCITRAL, however, so it’s interesting to compare how these two dispute resolution mechanisms look to participants today with how they looked in 1960. In this post, we cover three issues: legal development, appointments, and costs.

Courts and Tribunals as Agents of Legal Development

Permanent international courts can be influential agents of legal development. When a permanent court has jurisdiction over an issue, it can use its discretion to push policy in a particular direction.

This was recognized in the 1960s—many ILA participants expressed concerns about an investment court having an outsize ability to shape the law, given the unsettled status of international law regarding foreign investment in the 1960s. (In 1964, the US Supreme Court observed there “are few if any issues in international law today on which opinion seems to be so divided as the limitations on a state’s power to expropriate the property of aliens.”) Other ILA attendees believed a court was the right actor to find a middle ground between states, through the gradual development of legal reasoning. An ILA attendee from Denmark, Dr. Foighel, noted:

During the discussion there has been some fear as to the ability and the possibility of an international court to solve the different interests between the capital-importing and the capital-exporting states…
This proves as a matter of fact that there exist some common interests between the capital importing and exporting states. To clarify this common interest is the most important task for international lawyers today, and an international court will be a very useful organ to assist in this task.

What wasn’t recognized during the 1960s was that arbitral tribunals could become agents of legal development. Yet this is what has happened. Today arbitral decisions are often characterized as jurisprudence constante, even though contradictory decisionsoccur and calls for a review mechanismare common. Citing precedent is extensive, and increasing transparency makes legal development through arbitral awards more likely to occur.

Today, arbitration tribunals and courts are both agents of legal development. The choice facing participants at UNCITRAL is not court=legal development or arbitration=no legal development. Instead, participants must decide if in the long run, they prefer legal development to occur in a permanent body where precedent operates formally or in a more decentralized manner where precedent operates informally and somewhat inconsistently. And in the meantime, through what means (joint interpretative committeesor more specific substantive law, for instance) would they like to limit adjudicator discretion or encourage consistency?

Designing appointment systems for international courts is a delicate diplomatic task, and appointments to international courts can become political struggles, as seen in the 2017 ICJ elections.

This was recognized in the 1960s—several ILA attendees believed that governments were unlikely to agree on a method of appointing judges to an international investment court. Others thought an effective procedure could be created. The “Draft Statutes for the Foreign Investment Court” set out a detailed procedure for electing judges, and specified standards that judges should meet. (Many of these standards are relevant today, for instance Article XIV (1) addressing double hatting: “No member of the Court may act as agent, counsel, or advocate in any case.”) In the 1960s, the view that appointing arbitrators was simpler and therefore more feasible won out; it was a view shared by the father of the ICSID Convention, Aron Broches, who believed the “political complications” of appointing judges with fixed terms would prove an “insurmountable obstacle” to the creation of an investment court.

What Broches and others did not anticipate was party appointment becoming so controversial that addressing public perceptions of arbitrators as beholden to the party that appointed them would becomea pressing concern for some governments. The perception of party-appointed arbitrators as potentially biasedand disproportionately stale, male, and paleis unlikely to go away without reform. If anything, it is likely to intensify as challenges and other developmentsbecome more common and as increased transparency enables empirical scholarship on arbitrator behavior (from just last year, see Charlotin;Franck et alLangford, Behn, and Lie; and Puig and Strezhnev).

The question facing officials now is if they wish to focus on minor reforms to arbitral appointment procedures (perhaps limiting the number of times a law firm can appoint the same arbitrator), on major reforms to arbitral appointment procedures (perhaps getting rid of party appointment), or on designing a procedure for fixed-term appointments. As Anthea Roberts has written, it makes sense for states to think about generating a suite of reform options: designing a procedure for fixed-term appointments does not exclude or replace reforming arbitral appointments.

Permanent international courts can be costly, and negotiating financing arrangements for them is a challenging task.

This was recognized in the 1960s—even ILA attendees who saw merit in a court recognized the low cost of arbitration as an important advantage. Ignaz Seidl-Hohenveldern, for instance, who believed a court had many theoretical advantages, observed, “An Arbitral Tribunal would be preferable to a Permanent Foreign Investments Court, especially in view of the costs involved in the latter case.” Many suggestions for reducing costs, such as fewer judges initially or employing fact-finding officials (so judges would deal only with legal questions) were presented. Despite these suggestions, it was hard to beat arbitration when it came to the cost issue in the 1960s: as Aron Broches observed, arbitration “wouldn’t require money.”

Consenting to arbitration still doesn’t require money, and governments do not fund arbitral secretariats. While this looks like a good deal for governments in the short run, in the long run it has led governments to exercise less oversight of what these secretariats do, and arguably gives secretariats an incentive to be more entrepreneurial, which tends to help generate more cases against states in the long run.

Cases do cost governments money, of course, and dramatic increases in costs are an urgent concern for governments. Costs were the main substantive topic discussed in the initial UNCITRAL Working Group meeting last fall. This discussion identified factors that contribute to lengthy, costly ISDS proceedings and highlighted ways to reduce costs, such as early dismissal mechanisms or consolidating cases. The Report of the Working Group makes only an oblique reference to a court, noting, “the systematic nature of the concerns identified indicated a need for systemic solutions, which would bring with them the reduction of the overall costs through enhanced predictability and a greater ability to control proceedings themselves.” This subtly asserts that while a court has higher standing costs, it may generate lower costs per case, because counsel will not spend time on arbitrator selection and because, if precedent operates, counsel will be able to consider certain questions of law settled. While many questions about costs remain unanswered, assertions like these show the debate has moved beyond the 1960s binary in which a court required money while arbitration required no money.

There are many broader questions that need careful consideration in ISDS reform debates, including whether any legal dispute mechanism is warranted and whether the justifications for such a mechanism (many of which are inherited from the 1960s) still stand. The reform debate is not arbitration versus a court. If we do look only at these two elephants and how they have changed since they were last compared in the 1960s, however, we observe that many of the challenges that accompany creating a court are the same, while investor-state arbitration has been transformed, invalidating traditional assumptions and generating a range of new reform challenges.