sg-crest A Singapore Government Agency Website
Official website links end with
Secure websites use HTTPS
Look for a lock () or https:// as an added precaution. Share sensitive information only on official, secure websites.

Justice Philip Jeyaretnam: Speech given at the 25th Annual IBA Arbitration Day – Mediation Breakfast

25th Annual IBA Arbitration Day – Mediation Breakfast 
Investor State Disputes: How Arbitration and Mediation can intertwine to provide more resonant solutions
Friday, 23 February 2024
The Honourable Justice Philip Jeyaretnam*
Judge of the High Court
Supreme Court of Singapore
President, Singapore International Commercial Court

I.     Introduction 

1.    I am honoured to have this opportunity to address this distinguished gathering of mediation professionals about mediation in relation to investor–state disputes. My specific topic concerns how mediation supports other modes of dispute resolution, such as arbitration and court adjudication. I know that many of you have travelled some distance to attend the 2024 IBA Arbitration Day. Singapore is the right place to discuss mediation. It was here in Singapore that the United Nations Convention on International Settlement Agreements Resulting from Mediation was signed in 2019. It is known as the Singapore Convention and provides a legal framework for the recognition and enforcement of settlement agreements. The Singapore Convention applies to settlements in the international commercial context.(1) This includes ISDS, although states are entitled to make a reservation to the effect that it will not apply the convention to settlement agreements to which it is a party.(2) Thus far, 5 states (Belarus, Georgia, Iran, Kazakhstan, and Saudi Arabia) have made this reservation.(3) 

2.     Mediated settlements are often performed and complied with voluntarily. However, sometimes they are not, and enforcement becomes important. At present, the Singapore Convention has 57 signatories. It is in force for 11 states (and will be in force for 13 states on 27 May 2024).(4) This may seem limited coverage if contrasted with the New York Convention that provides for the mutual recognition and enforcement of arbitral awards, which is in force in 172 states,(5) or with the ICSID Convention, which has 158 contracting states to date,(6) and which applies in over half of ISDS cases.(7) The ICSID Convention permits pecuniary obligations in ICSID awards to be enforced as if they were no different than a judgment debt ordered by contracting states’ domestic courts.(8) Further, ICSID awards may not be set aside or refused enforcement by individual national courts, and are susceptible only to annulment by an ICSID ad hoc annulment committee.(9) 

3.     However, there are other ways to give teeth to an ISDS settlement agreement. One is where it is recorded as a consent award of an arbitral tribunal. If it is an ICSID tribunal, the parties can then rely on the enforcement provisions of the ICSID Convention. This is increasingly common and is expressly provided for in the ICSID Arbitration Rules.(10) Recording a settlement in a consent award by a non-ICSID Tribunal would result in its enforceability under the New York Convention. Sometimes, parties may mediate before an arbitration is even filed. This leads me to note the alternative route for enforcing a settlement agreement, which is to convert it into a judgment in a state that is signatory to the Singapore Convention and whose judgments are widely enforceable. This would be the case for Singapore Courts including Singapore International Commercial Court (“SICC”). Judgments of the SICC are widely enforceable. This happens under one of four routes: the 2005 Hague Convention on Choice of Court Agreements; reciprocal statutory registration and enforcement; common law enforcement and enforcement by civil law procedures.(11) Collectively, this covers almost all major jurisdictions around the globe.

4.     We can all agree that mediation is an increasingly important option for parties to business disputes. A successful mediation can save time and money for parties. But it is important to keep in mind that mediation does not stand alone. It only operates properly against a background where there are expeditious, cost–effective and fair methods ultimately available for the determination of legal rights and wrongs. In the absence of this – for example, where the time taken to arbitrate a matter and then have it enforced is much too long – mediation becomes a poor substitute where the party seeking a just result must accept whatever is offered because otherwise it will take too long to obtain adjudicated justice. On the other hand, respondents faced with a persistent but unmeritorious claimant may also pay them off in nuisance money because the time and costs of defending a claim are just too great. For these reasons, promoting mediation is not an alternative to keeping the primary modes of dispute resolution clear and unclogged. Keeping justice efficient and accessible is a never–ending task that every generation must renew in the light of technological and social change.

5.     Let me now turn to mediation in the context of ISDS: there are certainly people who are skeptics about its potential. Some skeptics may be here in this audience. I will seek to answer their concerns. First, I outline some of those reasons for skepticism and offer some counterpoints. Then, I turn to the changing sentiments, especially among states, which has led to the adoption of mandatory cooling off periods that themselves offer an opportunity for mediation. I will then describe a recent innovation here in Singapore that may offer a helpful approach, if only by analogy, certainly as compared to the rigidity of multi–tiered or temporally sequential approaches.

II.     Potential difficulties in applying mediation in the ISDS context

6.     There are three principal difficulties that mediation poses in the ISDS context. One relates to how states who are the respondents to such claims may operate in terms of decision–making. Someone must decide to settle the claim. Any agreed settlement of claims, including mediated settlement, may raise governance issues for state agencies or officers. They may be reluctant to settle a claim without a formal legal determination – for fear of potential criticism of their conduct in so doing. This is the phenomenon of bureaucratic indecision: where the path of least resistance for individuals in a large organisation is not to make a decision if it can possibly be avoided. It is safer to leave the decision to determination by the arbitral tribunal.

7.     A second difficulty is that many investor–state disputes have broader social implications. Civil society has often been described as the third stakeholder in investor–state disputes because the outcome of the arbitration may have a social or environmental impact. Civil society in the state being sued may be concerned about the opacity of mediation: that it is all happening behind closed doors in secret.

8.     A third difficulty, this time relating to how investors make decisions about the conduct of their claim, is the growing involvement of third–party funders in financing investor–state claims. Naturally, such funders are in it for the financial return. This has led to an emphasis on the financial payout involved in any outcome versus other non–financial reasons for settlement such as preservation of a potential long–term relationship. The presence of non–financial factors favouring settlement is often cited as a key reason why mediation is often effective. If such factors carry less weight for the funder, then this may make it harder for claims to settle.

9.     These difficulties however can be overstated. In relation to governance, states can establish a decision–making body with proper lines of authority. An example of this is how Egypt set up an inter–ministerial committee to deal with ISDS, with settlement agreements with investors only becoming effective once approved by the Cabinet of Minister. (12) The argument also needs to be made more generally in favour of active anticipation of potential claims under investment protection regimes as well as the benefits of mediated or negotiated settlements as a way of managing and reducing costs. These are policy and operational matters for states to address, and this topic was recently discussed and recommendations made by UNCITRAL Working Group III.(13)

10.     As for civil society’s concerns, these are inherent in investor–state arbitration. Mediating the dispute is itself no more opaque than the arbitrations themselves. Certainly, the original processes for ISDS hindered transparency and public accountability. Over the past decade, however, reforms have been implemented. The 2013 UNCITRAL Rules on Transparency in Treaty–based Investor–State Arbitration and Chapter X of the 2022 ICSID Arbitration Rules (on Publication, Access to Proceedings and Non–Disputing Party Submissions) have been positive steps toward ameliorating this concern about secrecy and allowing interested non-parties to be heard. Moreover, mediation may offer an opportunity to civil society to be heard and participate in the mediation, although as mediation is a confidential process this would generally still have to be subject to the consent of parties.(14)

11.     I turn to a related but distinct concern often expressed by civil society actors: to the extent that disputes impinge on citizens’ rights, it is thought that these do not lend themselves to mediation. Nonetheless, it is noteworthy that the Hague Rules on Business and Human Rights Arbitration, which were launched in December 2019, expressly provides by Article 56 for mediation and other forms of collaborative settlement.(15) Likewise, ICSID recently adopted the ICSID Mediation Rules in 2022, representing the first institutional mediation rules designed specifically for ISDS disputes, which can be used either independently of, or to complement, ongoing ICSID arbitral proceedings.(16) This suggests that with appropriate participation of civil society and states’ active engagement, this problem can be ameliorated.

12.     Turning to the impact of the rise in third–party funding, while this is a real concern, it may be addressed in a number of ways, for example:

(a) Mediation should be considered at an early stage of the dispute, before funding is in place;

(b) Lawyers’ codes of conduct should ensure that lawyers keep their duty to the client paramount, with instructions for settlement being ultimately a matter for the client even though the funder may give input.

III.     The opportunity presented by ‘cooling off periods’ in investment treaties.

13.     This offers a segue into the next stage of this talk, namely the opportunity: the opportunity that is offered by the increasing inclusion of ‘cooling off periods’ in investment agreements prior to the initiation of an arbitration. For example, there are cooling–off periods in the EU–Singapore IPA (which has not yet been fully ratified by all EU countries) and the Singapore–Sri Lanka FTA. And Singapore’s BIT with Indonesia, which recently came into force on 9 March 2021, lays down a 1–year ‘cooling off period’ for consultations and negotiations, with a structured but voluntary mediation process which parties may opt–in to.(17) Such a cooling off period may indeed be an opportunity for the investor and the state to explore mediation before positions harden and costs are expended.

IV.     Potential resonance of INTEGRAF

14.    I now turn to the model of a new scheme being offered in Singapore, a scheme that helps disputants work out the best path through the thicket of options available to them. Before I introduce it, let me make two points. One concerns the utility of unbundling disputes, while the second concerns the importance of flexibility.

         A.     Unbundling complex disputes

15.     First, an investor–state dispute may entail many different aspects. For example, there may be a jurisdictional issue which might concern either the status of the investor or the nature of the investment. This is a binary issue – either there is jurisdiction or there is not – and it may be hard to mediate on its own. However, other issues may be narrowed or resolved through mediation as discrete issues: for example, the breaches of specific clauses or the question of quantum of damages. It is useful to consider how large disputes can be ‘unbundled’ into smaller points that can be addressed through alternatives – whether mediation or evaluation.

16.     For jurisdictional issues, it may be worth considering how these can be most effectively dealt with. Ultimately, they may end up in the courts at the seat of the arbitration or in courts at the place of enforcement. In the setting aside application in Swissbourgh Diamond Mines (Pty) Ltd and others v Kingdom of Lesotho [2019] 1 SLR 263 (“Swissbourgh Diamond v Lesotho”), our Court of Appeal had to determine whether the PCA tribunal had jurisdiction over the ISDS dispute and whether the claimants’ mining rights qualified as an ‘investment’ pursuant to the Protocol on Finance and Investment of the Southern African Development Community.(18) This involved questions of the legal interpretation of the Protocol but also factual questions, such as whether the mining leases had been ‘admitted’ into Lesotho by the Kingdom in order to qualify as a legally valid ‘investment’ within that nation.(19) Courts at the post–award stage, including our own, typically consider the question of jurisdiction afresh, de novo as the Latinism goes. Would it make sense for parties to avail themselves of opportunities to put the question of jurisdiction to an appropriate court, such as an international commercial court, at an earlier stage of the claim? I ask this only to provoke thought and consideration.

         B.     Flexibility

17.     The second point is that it can be difficult for parties to agree what is the best method for different aspects of the dispute once the case is underway and battle lines are drawn. For this reason, multi–tiered dispute resolution clauses have become popular. These may provide for negotiation, followed by mediation and only after that litigation or arbitration. Such provisions can be helpful. But they also suffer from rigidity. They provide a framework in advance that may not be appropriate to the dispute that in fact arises. Moreover, there is a sweet spot for mediation. Too early may be ineffective because parties at that point have insufficient knowledge about the other side’s case. Mediation may be more effective later on in the life of the dispute, including after some preliminary points are adjudicated. But the moment for mediation may pass, and it may become too late for mediation to successfully occur because parties are just too invested in the fight to the death. In short, stipulating in advance tiered steps removes flexibility from the process.

18.     Even more problematic however is that such tiering can itself give rise to disputes over whether steps are mandatory or whether they have indeed been taken or taken in good faith. As a result, such clauses may complicate disputes rather than aid their resolution.

19.     The uncertainty created by such clauses is demonstrated by the different approaches taken by arbitral tribunals in ISDS. Some have characterised steps to trigger negotiation or mediation contractually contemplated to be taken prior to arbitration as preconditions to jurisdiction. Others have regarded them as formalities that do not affect the tribunal’s ability to hear a case. This divergence of opinion was surveyed by the tribunal in Giovanni Alemanni and others v The Argentine Republic, ICSID Case No. ARB/07/8.(20) 

20.     Let me give two examples of differing approaches: the tribunal in Wintershall Aktiengesellschaft v Argentine Republic, ICSID Case No. ARB/04/14 (“Wintershall v Argentina”) took the view that, where the BIT provided that the investor had to first submit its disputes to the local courts for an 18–month period, followed by a further 3–month waiting period, before submitting the dispute to arbitration thereafter, these were mandatory preconditions to the host State’s consent to jurisdiction, and so upheld Argentina’s jurisdictional objection on that basis.(21)

21.     By contrast, in Société Générale de Surveillance S.A. v Islamic Republic of Pakistan, ICSID Case No. ARB/01/13 (“SGS v Pakistan”), the tribunal treated a requirement for parties to observe a 12–month period for negotiations and consultations before proceeding to arbitration as “directory and procedural rather than as mandatory and jurisdictional in nature. Compliance with such a requirement is, accordingly, not seen as amounting to a condition precedent for the vesting of jurisdiction”, thus affirming its jurisdiction over the dispute.(22) 

22.     The differing conclusions of the tribunals in SGS v Pakistan and Wintershall v Argentina no doubt depended on the wording of the provisions. Nevertheless, this demonstrates how adding procedural steps prior to the commencement of arbitration into an arbitration clause may give rise to uncertainty concerning whether the tribunal has jurisdiction over the dispute.

V.     Drawing an analogy to SICC and SMC’s INTEGRAF scheme

23.     So with all of this in mind, let me now turn to Singapore Mediation Centre (“SMC”), who have recently implemented the INTEGRAF scheme in collaboration with the SICC.(23) INTEGRAF stands for Integrated Appropriate Dispute Resolution Framework.  This service offered by SMC provides parties with guidance on how to unbundle disputes and apply the most appropriate dispute resolution mode to the various aspects of a dispute. It is all about containing and managing disputes. There are two ways for parties to access INTEGRAF. One is by agreeing in the contract to establish a ‘Conflict Avoidance Board’ under the auspices of SMC.(24) The other is to seek ad hoc conflict management from a Conflict Management Consultant appointed by SMC after a dispute has arisen. Compared to a multi–tiered dispute–resolution clause, INTEGRAF offers flexibility. Moreover, it is envisaged that the assistance of a professional neutral third party whose duty is to help parties resolve their dispute in the most effective, expeditious and economical manner will make it easier for parties to find a mutually beneficial path forward, notwithstanding the natural antagonism and distrust between disputants.

24.     INTEGRAF is principally aimed at business parties entering into commercial relationships with each other or facing commercial disputes. Nonetheless, the principles underlying it are relevant to ISDS. The utility of a neutral third–party professional becoming involved in helping parties to channel their disputes appropriately is clear. I have likened this to the ‘signal person’ who would make sure that trains proceeded on the right tracks for their destinations.(25) 

25.     Schemes such as INTEGRAF that are aimed at neutral and efficient management of disputes can complement the availability of dispute prevention or anti-escalation schemes, such as Brazil’s Direct Investment Ombudsman.(26) This Ombudsman assesses and responds to complaints as they arise, so that incipient issues may be resolved before they become full-blown disputes.

VI.     Conclusion

26.     Ultimately, parties and their lawyers should keep front–and–centre the question of how to resolve the dispute in the fastest, cheapest and most effective way (or ways) possible. Otherwise, the fate of far too many litigants in big complex matters is to be dragged into multiple cases, stretching over many years, with the procedures adopted only adding to the complexity of what was in the first place not a simple dispute anyway. This can be illustrated by reference to the dispute between two investors, Lao Holdings N.V. and Sanum Investments Limited on the one hand, and the Government of Laos on the other. This dispute, which first arose on 14 August 2012, has led to at least ten arbitration awards from tribunals constituted under the SIAC, ICSID (AF) and the PCA.(27) There have been three judgments of the Singapore International Commercial Court,(28) two Singapore High Court judgments,(29) multiple appeals to our Court of Appeal, as well as decisions from the federal US District Courts in Idaho and Delaware. As proceedings took place before different adjudicators, questions of jurisdiction and issue estoppel multiplied.

27.     I will stop there. I will be happy to take questions and would be interested to hear delegates’ views and concerns.

* I am grateful to my law clerk, Gavin Ezra Goh, for his assistance in the research for and preparation of this paper. 
(1)      United Nations Convention on International Settlement Agreements Resulting from Mediation (adopted on 20 December 2018, opened for signature on 7 August 2019), Preamble and Art 1.
(2)      United Nations Convention on International Settlement Agreements Resulting from Mediation (adopted on 20 December 2018, opened for signature on 7 August 2019), Art 8(1)(a).
(3)      United Nations Commission on International Trade Law, Status: United Nations Convention on International Settlement Agreements Resulting from Mediation (accessed on 19 February 2024).
(4)      United Nations Commission on International Trade Law, Status: United Nations Convention on International Settlement Agreements Resulting from Mediation (accessed on 19 February 2024).
(5)      United Nations Commission on International Trade Law, Status: Convention on the Recognition and Enforcement of Foreign Arbitral Awards (New York, 1958) (accessed on 19 February 2024).
International Centre for Settlement of Investment Disputes, ICSID Convention, Regulations and Rules (accessed on 19 February 2024).
(7)      United Nations Conference on Trade and Development, Investment Dispute Settlement Navigator – Arbitral rules and administering institution (accessed on 19 February 2024).
(8)      Convention on the Settlement of Investment Disputes Between States and Nationals of Other States (entered into force on 14 October 1966), Art 54(1); Kingdom of Spain v Infrastructure Services Luxembourg S.a.r.l. (No. 3) [2021] FCAFC 112, Reasons for Judgment of the Federal Court of Australia dated 25 June 2021 at [9]–[10]; Sodexo Pass International SAS v Hungary, Judgment of the High Court of New Zealand dated 10 December 2021 at [24]–[25].
(9)      Convention on the Settlement of Investment Disputes Between States and Nationals of Other States (entered into force on 14 October 1966), Art 52.
ICSID Arbitration Rules (2022 version), r 55(2)(b); Mobil Investments Canada Inc. v Government of Canada, ICSID Case No. ARB/15/6, Award dated 4 February 2020 at [17]–[20].
Singapore International Commercial Court (SICC) User Guides Note 7: Enforcement of SICC Judgments (2024) at [2]–[3] (URL:
(12)      Fatma Salah, Kluwer Arbitration Blog | Egypt: New Investment Law – ADR for Investor–State Disputes, Wolters Kluwer (14 April 2015).
(13)      United Nations Commission on International Trade Law (UNCITRAL), Report of Working Group III (Investor-State Dispute Settlement Reform) on the work of its forty-fifth session (New York, 27–31 March 2023) at para 14.
(14)      ICSID Mediation Rules (2022), rr 10–11; UNCITRAL Mediation Rules (2021), Arts 5(3) & 6–7.
(15)      International Centre for Settlement of Investment Disputes: World Bank Group, Rules and Regulations | Mediation (accessed on 20 February 2024).
(16)      Centre for International Legal Cooperation (CILC), Hague Rules on Business and Human Rights Arbitration (December 2019), Art 56(1).
(17)      Agreement Between the Government of the Republic of Indonesia and the Government of the Republic of Singapore on the Promotion and Protection of Investments (concluded on 11 October 2018, entered into force on 9 March 2021), Arts 15–16 & 17(1).
(18)      Swissbourgh Diamond Mines (Pty) Ltd and others v Kingdom of Lesotho [2019] 1 SLR 263 at [1]–[6].
(19)      Swissbourgh Diamond Mines (Pty) Ltd and others v Kingdom of Lesotho [2019] 1 SLR 263 at [177]–[181].
(20)      Giovanni Alemanni and others v The Argentine Republic, ICSID Case No. ARB/07/8, Decision on Jurisdiction and Admissibility dated 17 November 2014 at [303]–[304].
(21)      Wintershall Aktiengesellschaft v Argentine Republic, ICSID Case No. ARB/04/14, Award dated 8 December 2008 at [114]–[127] and [198(1)].
(22)      SGS Société Générale de Surveillance S.A. v Islamic Republic of Pakistan, ICSID Case No. ARB/01/13, Decision of the Tribunal on Objections to Jurisdiction dated 6 August 2003 at [183]–[184] and [190(a)].
(23)      Singapore Mediation Centre (SMC), Our Services | INTEGRAF (accessed on 20 February 2024); Singapore International Commercial Court (SICC), Guide to the SICC | SICC and INTEGRAF (accessed on 20 February 2024).
(24)      Rules of the Integrated Appropriate Dispute Resolution Framework (“INTEGRAF”) 2024, First Edition of the INTEGRAF Rules 2024, rr 6.1–6.2.
(25)      Keynote address on Appropriate Dispute Resolution for Transnational Projects in the Asian Context, paragraph 17, 3rd Singapore-China International Commercial Dispute Resolution Conference 2023, co-organised by Ministry of Law, CCPIT and ICDPASO, 20 October 2023
(26)      Established by Brazil in 2016 by Decree No. 8863.
(27)      Sanum Investments Limited v The Government of the Lao People’s Democratic Republic, PCA Case No. 2013–13, Award on Jurisdiction dated 13 December 2013; Lao Holdings NV v The Lao People’s Democratic Republic, ICSID Case No. ARB(AF)/12/6, Decision on Jurisdiction dated 21 February 2014; Lao Holdings NV v The Lao People’s Democratic Republic, ICSID Case No. ARB(AF)/12/6, Interim Ruling on Issues Arising Under the Deed of Settlement dated 19 December 2014; Lao Holdings NV v The Government of the Lao People’s Democratic Republic, ICSID Case No. ARB(AF)/12/6, Decision on the Merits dated 10 June 2015; Lao Holdings NV v The Government of the Lao People’s Democratic Republic, ICSID Case No. ARB(AF)/12/6, Decision on Costs dated 5 November 2015; The Government of the Lao People’s Democratic Republic v Lao Holdings NV and Sanum Investments Limited, SIAC ARB No. 143/14/MV, Final Award dated 29 June 2017; Lao Holdings NV v The Government of the Lao People’s Democratic Republic, ICSID Case No. ARB(AF)/12/6, Decision on the Merits of the Claimants’ Second Material Breach Application dated 15 December 2017; Sanum Investments Limited v The Government of the Lao People’s Democratic Republic, PCA Case No. 2013–13, Award dated 6 August 2019; Lao Holdings NV v The Lao People’s Democratic Republic, ICSID Case No. ARB(AF)/12/6, Award dated 6 August 2019; Sanum Investments Limited and Lao Holdings NV v San Marco Capital Partners LLC, Kelly Gass and the Government of the Lao People’s Democratic Republic, SIAC ARB No. 414/17/QW, Award dated 11 August 2021.
Lao Holdings NV v Government of the Lao People’s Democratic Republic and another matter [2021] SGHC(I) 10; Lao Holdings NV v Government of the Lao People’s Democratic Republic and another matter [2022] SGHC(I) 6; Sanum Investments Ltd and another v Government of the Lao People’s Democratic Republic and others and another matter [2022] SGHC(I) 9.
The Lao People’s Democratic Republic v Sanum Investments Ltd and another and another matter [2013] SGHC 183; Government of the Lao People’s Democratic Republic v Sanum Investments Ltd [2015] SGHC 15.


Share this page: