AN INTERNATIONAL CONVENTION ON EXPERT DETERMINATION AND DISPUTE BOARDS?

Abstract This article makes a case for an international convention on expert determination (ED) and Dispute Boards (DBs) that would require its Contracting States to recognise agreements on ED/DBs and enforce ED/DB decisions. Whilst strong, the case for the convention may not be compelling as there are arguments against it. But at least the time has come for the international legal community to start thinking about and debating the need for such an international regime. This article takes the first step towards imagining this international regime by evaluating a number of key issues relating to its scope of application.


I. INTRODUCTION
Dispute resolution methods in commercial contracts are wide ranging and include, in addition to litigation, arbitration and mediation as private (non-State) or alternative dispute resolution (ADR). Underpinned by the New York Convention (NYC) 1 that recognises arbitration agreements and enforces foreign arbitral awards in their own right, arbitration has become a highly prominent ADR mode globally. The international community has recently taken a further step to promote ADR, and its associated benefits, by adopting the Singapore Convention on Mediation (SCM) 2 that recognises and enforces settlement agreements resulting from mediation. In comparison with arbitration and mediation, another set of ADR methods-a third-party expert determination (ED) and Dispute Boards (DBs)-lacks an equivalent international regime.
ED is a process whereby parties agree to refer an issue, difference or dispute, arising from their contract, to be resolved by an expert. DBs are a somewhat similar process typically used in construction and infrastructure projects, whereby contractual issues or disputes are entrusted to what is often a 'standing' body during the life of a project. Despite many benefits of ED/ DBs and their extensive use in numerous contracts and industries, they are not as well known internationally as other ADR mechanisms, such as arbitration. Some jurisdictions have a long-standing experience and welldeveloped law on ED, whilst some others have very little, if any, experience and an unclear position on ED/DBs. On the whole, legal scholarship internationally has not paid sufficient attention to and not kept pace with the world of ED/DBs, 3 whose prominence, use and complexity have increased in recent decades.
This article seeks to contribute to legal scholarship on ED/DBs by evaluating: whether there is a need for an international convention on ED/DBs, a counterpart international regime to those in arbitration and mediation; and, proceeding on the premise that such a convention is needed, a number of key issues concerning its scope of application. Jurisdictions with a rich experience of ED have rules and principles relating to many aspects of ED (and to a lesser extent DBs). But what holds back the reliability and effectiveness of ED/DBs as an ADR mode is the possibility of their not being recognised by some legal systems and the inability of ED/DB decisions to be enforceable in their own right internationally, a critical factor in a commercial party's choice of an ADR method. 4 The truth of this proposition is evidenced by the NYC and SCM focussing on the legal recognition of agreements on arbitration and those resulting from mediation and the enforcement of arbitral awards and settlement agreements resulting from mediation. Therefore, the need for a convention on ED/DBs will be evaluated on the premise that this convention should be concerned with the legal recognition of agreements on ED/DBs and the enforcement of ED/DB decisions. 5 Part I of the article explains the basic features of ED and DBs. Part II explores the current legal framework relating to ED/DBs and demonstrates that there is 3 There is, of course, some excellent research on this topic, including C Freedman and J Farrell, Kendall on Expert Determination (5th edn, Sweet & Maxwell 2015) and F De Ly and PA Gélinas (eds), Dispute Prevention and Settlement through Expert Determination and Dispute Boards (International Chamber of Commerce 2017) and others cited in this article. 4 It is the ability of such a convention to ensure that the agreements on ED/DB are recognised and that ED/DB decisions are enforceable in their own right internationally that explains this article's focus on an international legislative instrument, as opposed to a 'soft' law instrument, such as a model law. The latter, whilst promoting harmonisation of domestic laws in this area amongst jurisdictions, cannot establish a uniform international regime for legal recognition of ED/DB agreements and enforcement of ED/DB decisions (see further (nn 44-47) and the accompanying main text and Section IV). 5 The grounds on which the recognition of agreements on ED/DB and enforcement of an ED/ DB decision can be refused should be this convention's integral part. However, this issue requires detailed treatment, which cannot be done in this article and is left for a separate examination.

698
International and Comparative Law Quarterly considerable legal uncertainty and complexity surrounding legal treatment of ED/DBs across the world, which in international transactions makes it difficult to predict whether an agreement on ED/DB and an ED/DB decision will be recognised and enforced. Part III evaluates the arguments in favour of and against a convention on ED/DB. Part IV assumes that the arguments in favour of this convention prevail and analyses key questions concerning the convention's possible scope of application, namely: whether the convention should apply to both ED and DBs; which ED/DB decisions should fall within the convention; whether the convention should cover ED/DB only as an ADR mode or also as a contractual gap-filling mechanism; whether the convention should apply to agreements on ED/DB being a precondition to litigation or arbitration; the distinction between arbitration and ED/DB; the distinction between mediation and ED/DB; whether the convention should apply to certain forms of ED, such as adjudication, and DBs statutorily regulated in some jurisdictions; 6 and the criteria of an 'international' ED/DB. The conclusions are drawn in the final part.

II. EXPERT DETERMINATION AND DISPUTE BOARDS
A third-party expert determination (ED) is a process whereby parties agree to refer an issue, difference or dispute, arising from their contract, to be resolved by an expert. The issue may relate to a step in the contractual performance before any disagreement arises-such as fixing or revising the contract price, determining the time for performance or an aspect of conformity of the goods in a sales contract-in which case ED crystallises the parties' rights and obligations and injects detailed meaning into the contract, making it work and preventing disagreements/disputes from arising (contractual 'gap-filling' function). ED can also be an agreed mode of dispute resolution (dispute resolution function), an ADR mechanism. Any contractual matter, factual or legal, can in principle be subject to ED. The parties often agree that ED will be final 7 and binding, 8 in which case it brings finality helping avoid disputes, costs, time and effort to which disputes would otherwise give rise. ED is a speedy, relatively inexpensive, informal, confidential and non-adversarial resolution by an expert. ED is used in a wide range of industries and contracts, which are often based on industry standard-form contracts. For example, ED is used in: sales contracts in the 'commodities' trade often providing for a final and binding inspection certificate in respect of quality, quantity, weight or other aspects of the 6 See (nn 20, 37, 98-99, 142-148) and the accompanying main text. 7 In the sense that parties seek to preclude further resort to another despite resolution mechanism. 8 In the sense that parties have an obligation to comply with the decision and in the case of a party's failure to do so, the other party may enforce it by resorting to court or arbitration (if there is an arbitration clause). goods; 9 the extractive sector, 10 especially the oil and gas industry, where it is used in contracts at all stages of operations, upstream to downstream; 11 the construction sector, where some jurisdictions statutorily regulate 'adjudication', 12 arguably a special form of ED; 13 intellectual property, 14 so much so that the World Intellectual Property Organization (WIPO) has developed its own rules on ED. 15 Other industries using ED include shipbuilding, broadcasting, telecommunications, insurance, banking, finance, insolvency, government procurement, healthcare, defence, 16 arts and antiquity, 17 mergers and acquisitions 18 and accounting. 19 ED is based on the parties' contract and typically governed by contract law in the vast majority of jurisdictions. In some jurisdictions with legislation on construction adjudication, an adjudicator's decision is enforceable in the same manner as a court judgment. 20 In some others, an adjudicator's or an ED decision may be enforceable as a summary judgment (without a full 9  International and Comparative Law Quarterly trial). 21 But generally an ED decision is enforceable only by bringing a breach of contract claim in litigation or arbitration, if there is an arbitration clause in the contract. 22 If such an action is successful, the available remedies will vary depending on the applicable law and include specific performance, damages and injunctions. 23 The inability of ED decisions to be enforceable in their own right is a considerable weakness of ED and a real obstacle to ED growing into an effective and reliable mechanism that is internationally recognised and enforceable. 24 Turning now to DBs, the 'DB' concept as a method of avoiding disputes (by issuing non-binding recommendations) in large-scale/public infrastructure projects emerged in the United States in the 1970s. 25 Since then, DBs have been used widely not only in the US 26 but around the world. The first project outside the US where a DB was employed as a means of avoiding disputes (also by issuing non-binding recommendations) was a hydroelectric project in Honduras in the early 1980s. 27 Subsequently, DBs have been used internationally not just for the purpose of non-binding recommendations but also to resolve disputes. 28 In such cases, DB decisions adjudication decision as if it is a judgment or order of the High Court'); section 27(1), (2), Building and Construction Industry Security of Payment Act 2000 (SOP) (Singapore). 21 As is the case in the UK (see Freedman and Farrell (n 3) 108, 305; Rule 24.1, Civil Procedure Rules). 22  26 'The Eisenhower DRB issued non-binding formal recommendations on a total of three disputes as well as informal assistance (dispute avoidance) on another nine disputes. All of the recommendations and informal assistance led to resolution and there was not post-contract litigation. The Eisenhower experience was considered a success and the DRB process was generally adopted by most contracting authorities across the US, and indeed has become mandatory in at least two jurisdictions in that country' (ibid 64). 27 ibid 65. 28 The International Chamber of Commerce Dispute Board Rules 2015 (<https://iccwbo.org/ dispute-resolution-services/dispute-boards/rules/>) provide for three types of DB: 1. Dispute Review Boards (DRBs): 'DRBs may assist the Parties in avoiding Disagreements, in resolving them through informal assistance, and by issuing Conclusions with respect to Disputes upon formal referral. In formal referrals, DRBs render Recommendations with respect to Disputes. Upon receipt of a Recommendation, the Parties may comply with it voluntarily but are not required to do so.' (art 4(1) and (2)); 2. Dispute Adjudication Boards (DABs): 'DABs may assist the Parties in avoiding Disagreements, in resolving them through informal assistance, and by issuing Conclusions with respect to Disputes upon formal referral. In formal referrals, DABs render Decisions with respect to Disputes … A Decision is binding on the Parties upon its receipt' (art 5(1) and (2)); 3. Combined Dispute Boards (CDBs): 'CDBs may assist the Parties in avoiding Disagreements, in resolving them through informal assistance, and by issuing Conclusions with respect to Disputes upon formal referral. In formal referrals, CDBs render Recommendations with respect to Disputes pursuant to Article 4, but may render Decisions pursuant to Article 5 … If any Party requests a Decision with respect to a given Dispute and no can be 'provisionally binding', until and unless they are challenged by another means of dispute resolution (arbitration or litigation), or 'final and binding'. 29 Like ED, DBs are expert driven, with industry experts appointed as DB members, speedy and less formal and adversarial than litigation and arbitration. Although DBs can be ad hoc, they are often 'standing' boards, used as a permanent dispute avoidance and/or resolution mechanism throughout the life of a project. 30 Disagreements/disputes are addressed by DBs in 'real time', enhancing their ability to prevent disagreements/disputes from arising or escalating and to preserve business relationships. 31 By some estimates, the success rate of DBs in preventing disagreements is more than 95 per cent. 32 The possibility of provisionally binding DB decisions, based on the idea 'pay now-argue later', enables uninterrupted cash flow despite disagreements. All these benefits of DBs and the largely positive international experience of using them have led international development banks and financial institutions to recommend and invest in DBs, especially in developing countries. 33 The emerging global prominence of DBs is also reflected in the adoption of rules on DBs by the International Chamber of Commerce (ICC) 34 and some other bodies promoting ADR. 35 Equally importantly, some leading industry standard-form contracts, such as those of the International Federation of Consulting Engineers (FIDIC), include DBs as an ADR method. 36 Like an ED decision, a DB decision is not generally enforceable in its own right. Unless domestic law provides for its direct enforceability, a party's failure to comply with a DB decision is normally enforceable only as a breach of contract claim by the other party. That said, some jurisdictions make DB decisions enforceable through legislation. 37 An alternative approach is other Party objects thereto, the CDB shall render a Decision … If any Party requests a Decision and another Party objects thereto, the CDB shall make a final decision as to whether it will render a Recommendation or a Decision.' (art 6(1)-(3)). 29   International and Comparative Law Quarterly offered by the decision of the Singapore Court of Appeal. In the case involving a FIDIC contract 38 providing for a provisionally binding DB decision (until/ unless it is revised in an amicable settlement or arbitration), 39 the court held that the clause providing for a dispute resolution by a Dispute Adjudication Board (DAB) 'imposes a distinct contractual obligation on a paying party to comply promptly with a DAB decision regardless of whether the decision is final and binding or merely binding but non-final, and this obligation is capable of being directly enforced by arbitration …'. 40 But these approaches of individual jurisdictions 41 are far from reflecting the future directions that others may take. 42 The prevailing position worldwide is that DB decisions are not enforceable in their own right. This holds back DBs and prevents them from becoming an effective, reliable and globally recognised and used ADR mechanism. 43

III. CURRENT FRAMEWORK
International arbitration has become a highly prominent and widely used ADR method, which is reflected in a well-developed international legal framework comprising, amongst others, the NYC and the UNCITRAL Model Law on using-dispute-boards-in-public-construction-projects/> where the use of DBs is provided for in the legislation on public procurement ( ] decision shall be binding on both Parties, who shall promptly give effect to it unless and until it shall be revised in an amicable settlement or an arbitral award as described below … [7] … If the DAB has given its decision as to a matter in dispute to both Parties, and no notice of dissatisfaction has been given by either Party within 28 days after it received the DAB's decision, then the decision shall become final and binding upon both Parties …'. ('a detailed review of the judgments, including the strong dissenting judgment, indicates that the matter cannot be said to be resolved, and it is likely that other jurisdictions may well come to the opposite view.'). 43 In a similar spirit but in a domestic law context, see Figueroa (n 37) 169 ('the stumbling block for the full acceptance of DBs in Mexico seems to reside in the juridical weakness of DB decisions …. DB decisions lack the legal force of judicial decisions or even arbitral awards … DB decisions seem to be impossible to enforce based on current legislation in Mexico, unless they are approved by and incorporated into a judicial decision.').
International Commercial Arbitration (ML). 44 Under the NYC, the Contracting States have the obligation to recognise and enforce arbitration agreements 45 and recognise as binding and enforce final arbitral awards. 46 The NYC Contracting States have also agreed on a uniform set of grounds for the refusal of the recognition and enforcement of foreign arbitral awards. 47 The ML is not a piece of international legislation but an instrument, which States can use as a model to develop or reform their arbitration laws and which promotes harmonisation amongst arbitration laws worldwide.
The increased use and benefit of mediation as an ADR mode is also recognised in the recent adoption of the SCM, which has created a uniform international regime in respect of settlement agreements resulting from mediation. The SCM Contracting States undertake an obligation to enforce such settlement agreements 48 by, amongst others: allowing a party to invoke the settlement agreement to prove that the matter in dispute with the other party(ies) has already been resolved by the settlement agreement; 49 or providing that if an application or claim relating to a settlement agreement has been made to a court, tribunal (or other forum) which may impact on the relief sought under this convention, the competent authority of a Contracting State 'may, if it considers it proper, adjourn the decision and may also, on the request of a party, order the other party to give suitable security'. 50 The SCM also contains a uniform set of grounds for refusal of the relief sought under the SCM. 51 The absence of an international regime on ED/DBs means that the effects of an agreement on ED/DBs and of ED/DB decisions are subject to the applicable law. The approaches of legal systems vary greatly. Whilst it is not possible to explore the approach of each legal system, in broad terms the approaches can be subsumed into four general categories. At one end of the spectrum, most favourable to ED/DBs, lies the first category of systems that have long recognised and promote ED (and to an extent DBs). 52 These systems recognise that ED/DBs can perform two functions: (1) a gap-filling mechanism making the contract work, preventing any differences/disputes from arising; and (2) an ADR mode to be invoked if/when disputes arise. These systems recognise that 'factual' and 'legal' issues can be resolved by 44

704
International and Comparative Law Quarterly ED/DB. Subject to the grounds of challenge, which are similar but not identical, 53 these legal systems recognise and enforce agreements on ED/DB and ED/DB decisions. Then follow legal systems in the second category that recognise ED/DBs to a more limited extent. They may recognise ED/DB only as contractual gap-filling mechanisms, not as ADR modes. 54 They may also limit ED/DBs to 'factual' issues. 55 The grounds for challenging decisions vary depending on the individual system. 56 These two categories of jurisdictions normally regulate most aspects of ED/ DBs by contract law. In the common law, this body of contract law is largely based on case law, whereas in civil law jurisdictions the rules and principles are normally to be found in a (civil) code and/or court decisions. Some legal systems may subject aspects of ED/DBs to their procedural, rather than, substantive law. 57 In both the common law and civil law jurisdictions, certain forms of ED and DBs can be subject to special legislation. 58 In the third category are systems where it is simply unclear whether the agreements on ED/DBs and ED/DB decisions will be recognised and enforced. This may be so due to a legal system's level of development and/or because the jurisdiction has little experience with ED/DBs. 59 The fourth category lies at the opposite end of the spectrum to that occupied by the first, taking a hostile approach to ED/DBs. Whilst such systems may lack legislative provisions or cases on ED/DBs, there is a risk that in practice an agreement on ED/DB may be rendered invalid or void as contrary to public policy and/or because it ousts the jurisdiction of courts. 60 53 Singapore recognises 'manifest error' as a challenge to ED in law (The Oriental Insurance Co Ltd v Reliance National Asia Re Pte Ltd [2008] SGHC 236, paras 47, 85-86; Tan Yeow Khoon v Tan Yeow Tat [2003] 3 SLR 486), which is not the case in English law. It may also be that Australian law does not recognise the 'material departure from instructions' as a challenge to ED, as is the case in English law (see Saidov, Pt I, (n 24) 322-3) . 54 This appears to be the case in Germany (see, eg, Sessler and Leimert (n 18) 156) and Spain (see E Soler, 'Spain' in Freedman and Farrell (n 3) 479-80). It has been reported that in Dutch law, an agreement to resolve disputes by ED is treated as a settlement agreement; an agreement to use ED as a gap-filling mechanism is not so treated (PE Ernste, 'Binding Advice: The Civil-Law Perspective' in De Ly and Gélinas (n 3) 13). 55 This appears to be the case in Switzerland (art 189(3) Swiss Civil Procedure Code) and Austria (G Zeiler, 'Austria' in Freedman and Farrell (n 3) 468). 56  The legal effects of an agreement to subject an issue or dispute to ED/DB and of an ED/DB decision are, first, to be determined by the law governing the agreement on ED/DB and ED/DB respectively. Given the different approaches of legal systems, the legal effects and implications can vary greatly. Suppose that a long-term gas sales agreement provides that if the contracting parties, based in different countries, disagree on and dispute the applicable index in a price revision clause, the party-appointed expert's decision on the index shall be final and binding. 61 The seller wishes to enforce the decision against the buyer who refuses to comply with it. If the law applicable to ED and the agreement on ED is that of a jurisdiction in the first category, the agreement on ED/DB and the ED/DB decision will be recognised, although the buyer may be able to challenge the decision on various grounds. 62 If the applicable law is that of a jurisdiction in the second category, neither the agreement on ED/DB nor an ED/DB decision will be recognised if the ED/DB is characterised as a dispute resolution clause. For some such jurisdictions, it is also critical whether fixing the index is a factual or legal issue, with the legally recognisable ED/DB confined to factual issues. If the governing law is that of a jurisdiction in the third category, the outcome will be unpredictable. If the law of the fourth category jurisdiction is applicable, an agreement on ED/DB and an ED/DB decision may well be invalid.
The legal implications further depend on the law of the jurisdiction where enforcement is sought. 63 In this scenario-where the law governing an agreement on ED/DB and an ED/DB decision is different from the law of a jurisdiction where enforcement is sought-the fourfold categorisation again gives a broad indication of the extent to which courts in those jurisdictions may be prepared to enforce ED/DB decisions. For example, in the fourth category jurisdiction a court may refuse to enforce the decision, even if the law applicable to ED/DB and the agreement on ED/DB recognises and upholds that agreement and the ED/DB decision. Even if the enforcement is sought through arbitration with a seat in a jurisdiction in the fourth category, there may be a risk that: the arbitration tribunal might not give effect to the ED/DB decision; 64 the law of the seat of arbitration (lex arbitri) might not allow the arbitral tribunal to award certain remedies; 65 or the courts at the suggested that there is a risk that a clause on ED may contravene the rule of sharia against uncertainty (C Shepherd, 'Dubai' in Freedman and Farrell (n 3) 457). That said, it appears that in principle there is no objection in sharia to disputes, other than in areas pertaining to family, crime and public policy, being resolved by established ADR methods (see C Chern, Chern on Dispute Boards: Practice and Procedure (2nd edn, Wiley-Blackwell 2011) 66). 61 See, eg, Contact Energy (n 11). 62 See (n 53) and the accompanying main text. 63 '[I]f a DB decision is to be immediately enforced by a court then it is a consideration of the substantive and procedural laws of the applicable country or countries that will determine whether there is any chance of success.' (N Gould, 'Enforcing a Dispute Board's Decision: Issues and Considerations' [2012] ICLR 442, 445). 64 Freedman and Farrell (n 3) 314. 65 '[T]he law of the seat may not permit the tribunal to award a non-monetary remedy (for example, specific performance or declaratory relief) and a party may be left with no option but to purse a claim for damages.' (ibid).

706
International and Comparative Law Quarterly seat might refuse to enforce the arbitral award that recognises and gives effect to the ED/DB decision. 66 Similarly, if the enforcement of a court judgment, obtained in a jurisdiction upholding ED/DB decisions, is sought in a jurisdiction in the fourth category, there is still a risk that enforcement of the foreign court's judgment will be refused. 67 This complexity of legal implications is also increased by the uncertainty regarding how the rules of private international law ought to operate in respect of various aspects of ED/DBs. 68 76 Whether ED/DB is used as a gap-filling or an ADR mode may also be relevant to what law should apply to a given aspect of ED/DB. 77 If it is a gap-filling mechanism, it can be 66 ibid. 67 See, eg, ibid 313, fn 33. 68 'It is not simply the law of the contract, but potentially a number of legal systems that could apply to the substantive agreement, the classification of the applicable law, jurisdiction, the proper law of the contract, the procedural law, evidence, lex fori, and enforceability' (Gould (n 63) 445). 69 The issue of jurisdiction can also be viewed as part of the grounds of challenging an ED/DB decision. 70 See C Leaua, 'Independence and Impartiality of Experts in Expert Determination Proceedings' in De Ly and PA Gélinas (n 3) 56-9; GS Hök, 'Dispute Adjudication Boards-The International or Third Dimension' (2012) 4 ICLR 420, 431, 433, contending also that whilst a DB clause should constitute a separate agreement to that in the main contract, it should nevertheless be governed by the law applicable to the main contract due to the close connection between the two. Art 1230 French Civil Code ('Termination does not affect contract terms relating to disputeresolution, nor those intended to take effect even in the case of termination, such as confidentiality or non-competition clauses.') is sometimes interpreted as suggesting the autonomous nature of a DB clause (see, eg, CR Seppälä, 'Recent Case Law on Dispute Boards' in De Ly and Gélinas (n 3) 122, fn 37). 71 See Seppälä (n 70) 122, fn 37. 72 'Some arbitral tribunals have considered that dispute adjudication board procedure is governed by the lex arbitri, in view of the fact that such a procedure is part of a multi-tier dispute resolution mechanism and it is a mandatory pre-arbitration step' (Leaua (n 70)  argued that the law governing the contract should govern all aspects of ED/DB just as it governs other contractual issues. 78 If the ED/DB is intended as an ADR mechanism, the arguments in favour of the procedural law applicable to the dispute resolution 79 may be relevant. 80 Whilst non-exhaustive, these points show that the legal framework and thinking in the world today is such that there is much uncertainty and complexity involved in determining the legal implications of an ED/DB clause and ED/DB decisions. Tackling this complexity requires time and resources to ascertain the relevant conflict-of-laws rules, which will in turn determine the law(s) applicable to ED/DBs and agreements on ED/DB 81 and the laws of the jurisdictions where enforcement might be sought. The overall lack of certainty and predictability of legal outcomes flowing from ED/DBs is detrimental to business planning and increases risks, creating a sense of insecurity. All this leads to high 'transaction costs', casts doubt on the effectiveness and reliability of ED/DB, which can otherwise bring many benefits, and ultimately discourages the use of ED/DBs.

IV. THE NEED FOR AN INTERNATIONAL CONVENTION?
As already noted, there is no international framework on ED/DBs. There are codes and rules adopted by bodies and organisations that, if contractually incorporated, can govern a procedure to resolve disputes/differences by ED/ DBs. 82 There are also international industry standard-form contracts providing for the resolution of differences/disputes by ED/DBs. 83 Such standard-form contracts may, amongst others, provide for a procedural regime for ED/DBs, a status of an ED/DB decision 84 or for grounds, such as 'fraud, collusion or manifest error', on which ED/DB decisions can be challenged.
Such codes and standard-form contracts cannot perform the functions that the NYC and SCM perform in arbitration and mediation simply because they cannot impose an obligation on States to recognise and enforce agreements on ED/DBs and ED/DB decisions. Only an international legislative instrument can turn an ED/DB decision into an 'enforceable title'. Similarly, 78 ibid 57. Some argue, however, that in this case it is appropriate to imply procedural rules from the contract by interpreting the parties' intentions and relying on principles, such as good faith (Seppälä (n 70) 126). 79 The procedural rules applicable to court proceedings or lex arbitri if there is an arbitration agreement. There may still be arguments in favour of the applicability of the law governing the contract (see Leaua (n 70) 57). 80 Leaua (n 70) 57. 81 Which in international contracts is often foreign to at least one contracting party. 82 See (nn 28, 34-35) and the accompanying main text. 83 See (nn 30 and 42) and the accompanying main text. 84 Such as whether it is binding, non-binding or 'provisionally binding' (until/unless it is challenged in accordance with the contractual procedure), 'final' (seeking to preclude recourse to another mode of resolving disputes) or whether it can be appealed through arbitration or litigation (in which case ED/DB may be a precondition to arbitration/litigation).

708
International and Comparative Law Quarterly these codes and standard-form contracts are not concerned with a comprehensive set of grounds on which agreements on ED/DB and ED/DB decisions can be challenged internationally. This task is best left to an international treaty which will be directly concerned with the grounds of challenge. As a State-made instrument, a convention would be the best forum for reaching international consensus on the scope within which States are prepared to recognise and enforce agreements on ED/DBs and ED/DB decisions. Imagine a world where such an international convention exists. A party to a contract, incorporating an ED/DB mechanism, would be able to enforce this agreement in a Contracting State (to the convention) to ensure that a difference/dispute on a matter, falling within the ED/DB clause, be resolved by ED/DB. This party would be able to enforce the ED/DB decision in any Contracting State against the party refusing to abide by the decision. 85 The enforcement of the ED/DB decision would be much more effective than it currently is even in the first category jurisdictions because it would be enforceable in its own right, as an 'enforceable title', without the need for further dispute resolution proceedings to establish a contractual breach and that the preconditions for the applicable remedies have been met. 86 Such a legal regime would make ED/DBs a reliable and effective mechanism globally. The more widely adopted this convention would be, the greater the number of jurisdictions where agreements on ED/DBs and ED/DB decisions would be recognised and enforced. The ability of parties operating internationally to ascertain the legal implications of using ED/DBs will be greatly enhanced, reducing transaction costs. This legal regime would infuse parties with confidence in ED/DBs, encourage their use and implement the benefits associated with them. The parties would be able to evaluate risks of using ED/DBs, such as grounds on which a State can refuse to recognise an agreement on ED/DB or an ED/DB decision, and decide whether an ED/DB should be incorporated. These propositions accord with recent surveys, indicating that the enforceability of a decision is critical to the choice of an ADR mechanism. 87 85 See similarly in the SCM context, A Handbook on The Singapore Convention on Mediation (Singapore International Dispute Resolution Academy 2021) 5, 11-13 <https://www. singaporeconvention.org/sites/singaporeconvention.org/files/SMUSOL Singapore Convention MediationHandbook.pdf>. 86 See, similarly, Sessler and Leimert (n 18) 154 ('the expert determination has to be enforced in an additional proceeding before a state court or arbitral tribunal. Even if the inclination of state courts to declare an expert determination void due to obvious inequity or incorrectness is low, it nevertheless constitutes a considerable disadvantage of such proceedings.'). 87 See 'International Dispute Resolution Survey: 2020 Final Report' (Singapore International Dispute Resolution Academy 2020) 6-7 <https://sidra.smu.edu.sg/sites/sidra.smu.edu.sg/files/ survey/index.html>, according to which 71 per cent of the respondents ranked enforceability as the most important consideration.
What are other policy arguments in favour of this convention? First, freedom of contract and party autonomy in commercial matters are increasingly recognised internationally. ED/DBs are a creature of contract and consensual in nature. The freedom of contract perspective dictates that if parties have contractually agreed on an ED/DB mechanism, such agreements and resulting ED/DB decisions ought to be legally recognised and enforced. Secondly, ED/DBs help avoid disputes and preserve business relationships, which is especially important in and conducive to the successful completion of large-scale, long-term projects. Thirdly, as an ADR mechanism ED/DBs are speedier, cheaper, less adversarial and formal than litigation or arbitration, resulting in lower transaction costs and greater economic efficiency. Therefore, extensive use of ED/DBs worldwide could contribute to global economic growth. 88 Fourthly and generally, by creating a common framework international conventions are often thought to reduce transaction costs. 89 The costs of negotiating contracts could reduce. The uncertainty of legal implications of an ED/DB clause and an ED/DB decision would also be mitigated because those implications, being subject to the unified regime, may no longer depend on the vagaries of conflict-of-laws, the applicable law and the laws of jurisdictions where enforcement might be sought. Finally, the international community has been increasingly recognising the value of ADR. With the attention given to arbitration and mediation, the time is ripe to afford similar recognition to ED/DBs, making international commercial law and the ADR architecture more comprehensive and recognising the reality of ED/DBs being extensively used in many regions, projects and contracts. As ADR modes, ED/DBs produce 'savings in the administration of justice by States'. 90 By helping avoid and resolve disputes in a less adversarial and costly way than arbitration and litigation, they contribute to a positive and 'harmonious' international economic environment. 91 There are, of course, arguments against such a convention. 92 First, affording an 'enforceable title' to an ED/DB decision can undermine the 'rule of law'. Experts/DB members can be industry, not legal, experts entrusted to resolve legal matters. ED/DBs are faster and more informal than other dispute 88 The ability of international commercial law to produce economic benefits globally is increasingly recognised (see

710
International and Comparative Law Quarterly resolution methods, limiting the experts'/DBs' ability to consider issues fully and rigorously and do justice to the issues at hand. Further, not all regimes require experts/DBs to adhere to natural justice, fairness, independence and impartiality. These factors may prevent ED/DB decisions from administering justice by duly applying substantive and/or procedural law, as is done in litigation or arbitration. Some suggest that ED can amount to a decisionmaking 'by tossing a coin' 93 and, if so, it is arguable that the convention should not recognise and enforce such decisions. Secondly, ED/DBs are not yet as well known internationally as arbitration or possibly even mediation. It therefore may be premature to launch an international law-making project on mechanisms which are not yet fully understood and tested. A related (third) point is that it is unclear if the international community supports such an initiative. It may be pointed out that successful international commercial law conventions are those that were supported by the industries. 94 Only a high level of industry support would justify such an international law-making project. 95 Finally, many would regard the relative informality of ED/DBs as their advantage over arbitration, which has become increasingly formalised and adversarial. This informality enables experts/DBs to be flexible, low cost, quick and relatively non-adversarial. If ED/DB decisions become enforceable under the convention, that may damage this strength and radically transform their informal character. The fact that ED/DB decisions are enforceable internationally and there are uniform grounds for challenging them (especially if those are extensive) may induce: (a) parties to design sophisticated and formalised contractual regimes; and (b) experts/DBs to invest more time and effort in ED/DB procedures and insist on elaborate contractual arrangements with the contracting parties, primarily as protection against professional liability. ED/DBs may then become more formalised, expensive, timeconsuming, adversarial and lose their quintessential character 96 by being akin to arbitration, in which case the value of ED/DB becomes questionable.
On balance it is submitted that, whilst not compelling, the case for an international convention on ED/DBs is strong. Promoting freedom of contract, this convention would respond to the reality of ED/DBs being used extensively in practice. It would also be consistent with and continue an apparent policy of international commercial law to promote ADR. This regime will generate tangible benefits for parties, industries and global 93  economy, contributing to a more positive international commercial environment. At the very least, the time has come for the international community to start thinking about and debating the need for such a convention. Even if the criticisms of this initiative do not win the day and the work on the convention begins, they should influence the convention's design. Thus, concerns about the impact on the rule of law and the informality of ED/DBs should be reflected in the design of grounds of challenge. The arguments, questioning the extent to which EDs/DBs are known and used and calling for clarity on preferences of individual industries, not only draw attention to the need for further research and engagement with industry/legal practitioners but may also influence the convention's scope and structure. For instance, it may be that only certain industries support this convention, in which case there may be a case for confining its scope to certain sectors or designing the convention applicable to all sectors but, in addition, creating industry-specific protocols to address the needs of specific industries. 97 If the international community does not adopt this convention, the arguments in its favour indirectly make the case for greater recognition and facilitation of ED/ DBs amongst legal systems. The development of a Model Law on ED/DBs can help legal systems develop or reform their laws on ED/DBs, promoting harmonisation. The legal framework proposed by the Model Law will be especially useful to jurisdictions in the third category. It may also invite the fourth category jurisdictions to rethink their position. Even the second category jurisdictions may change their position. For example, if the Model Law provides for the possibility of an expert deciding on questions of law, not just those of fact, or ED/DB performing the function of contractual gap-filling and ADR, this can cause these jurisdictions to rethink their stance. The Model Law would thus be a welcome development that can trigger law reform worldwide, especially in the absence of an international convention. Even if the convention initiative goes ahead, it should not rule out the promulgation of a Model Law. Some countries might prefer not to ratify the convention but to develop or reform their law.

A. General
Suppose that the arguments supporting this convention prevail. It is obviously up to the drafters, representing different States and legal systems, to negotiate and agree on all aspects of the convention. This article does not propose that the convention should govern all aspects of ED/DBs. As has been noted, it is the risk of ED/DBs not being recognised by some legal systems and ED/DB decisions not being enforceable in their own right that holds back their reliability and effectiveness as ADR mechanisms. By analogy with the NYC and SCM, this convention should be concerned with the recognition of agreements on ED/DBs and the enforcement of ED/DB decisions. Whilst this article cannot address all aspects of the proposed convention, it will make the first step towards imagining this international regime by evaluating a number of key issues relating to its scope of application.

B. EDs and DBs?
Should this convention focus only on ED or also include DBs? The purpose and legal nature of these mechanisms are similar. They are both consensual mechanisms created by a contract, used to prevent and resolve disputes and composed of experts, technical, industry or legal. The differences between them are more relative than definite. DBs are often established as part of large-scale, long-term projects for their entire duration. The issues or differences/disputes, which can concern all aspects of the contract, are often resolved 'live' whilst the project is ongoing, reinforcing DBs' ability to prevent differences/disputes and facilitating the completion of projects. DBs are often composed as boards or panels, although they may consist of a sole individual. EDs, in contrast, feature in all types of commercial contracts, whether large-or small-scale, short-or long-term. EDs tend to be used for a specific matter, such as fixing or resolving a disagreement on a contract price. 98 Whilst EDs can be composed as panels, they frequently take the form of an individual expert. ED and DBs thus perform similar functions and the differences are of emphasis, not principle. For this reason, it would be best that both these mechanisms be included in the convention. The convention would then provide the international regime on contractual methods of resolving differences/disputes based on a third-party expert/panel that is not arbitration.

C. Which Decisions?
EDs/DBs can result in different types of decision. Experts/DBs can be contractually required to make either a binding or non-binding decision. The former, it is submitted, must fall within the convention's scope. It is binding decisions, with which one contracting party refuses to comply, that the other party can justifiably expect to be able to enforce. A non-binding decision cannot be expected to be enforceable and should fall outside the convention. ED/DB decisions are often contractually required to be 'final', seeking to preclude further resort to another dispute resolution mechanism. The ability of decisions to bring finality is a major advantage and the argument for binding and final decisions to be enforceable is even stronger than in the case of merely binding decisions. 'Final and binding' decisions should fall within the convention's scope.
The contract can provide for a 'provisionally binding' decision, as is frequently the case with DBs. Such decisions are binding until and unless a party follows a contractually specified method of challenging them, including initiating another dispute resolution mode (litigation or arbitration). It can be argued that this possible lack of finality militates against incorporating such decisions in the convention because doing so would generate unnecessary complexity by creating an additional legal regime that may not put an end to the parties' disagreement/dispute. However, it is suggested that on balance they should fall within the convention for two reasons. One, noted in the previous paragraph, is that being binding (even if provisionally) such decisions can be reasonably expected to be complied with. The other reason is that, based on the idea 'pay now-argue later', provisionally binding decisions ensure continuous cash flow to the parties without interrupting the project despite the difference/dispute being unresolved. The value of this commercial purpose is already well recognised by some domestic regimes on construction adjudication 99 that enforce such decisions notwithstanding their provisional character, 100 the ICC Rules on DBs 101 and some international industry standard-form contracts, 102 and, it is suggested, should be protected and facilitated internationally.
The issue subject to ED/DB can be factual, such as establishing the level of impurities in the wheat to determine compliance with the specifications in a sales contract, or legal, such as whether the preconditions of a force majeure or hardship clause are met. 103 As noted, some jurisdictions recognise ED/ DBs to the extent that they concern factual issues. More generally, it is frequently suggested that: ED/DBs are best suited to resolving factual issues, especially those of technical (non-legal) nature because they are best resolved through industry/technical expertise; and legal issues should be left to other dispute resolution modes. 104 Several arguments can be advanced in response. If freedom of contract is to be given precedence, the parties' choice to have a legal issue resolved by a third party (an expert or DB) ought to be respected. As masters of their agreement, the 99 See, eg, section 108(3) Housing Grants, Construction and Regeneration Act 1996 (HGCRA) (UK); section 6(10) CCA (Ireland); section 60 CCA (New Zealand); section 34 SOP (Singapore); also Freedman (n 52) 35. 100 See (nn 20, 21, 38-40) and the accompanying main text. 101 See art 5, ICC DB Rules. See further (nn 28, 32 and 34). 102 See (nn 38-40) and the accompanying main text. 103 In this case, the issue is one of contract interpretation, which is a matter of law. In addition, an interpretation of a clause needs to be carried out having regard to other relevant sources, such as, in the common law, a body of case law providing detailed guidance on how similar clauses have been interpreted previously. 104 The concern about the convention undermining the 'rule of law' has already been highlighted.

714
International and Comparative Law Quarterly parties should bear the risk of resolving their differences/disputes by ED/DB. The distinction between matters of 'fact' and 'law' is also notoriously difficult to apply. 105 The resolution of an issue can involve both matters and disentangling them may be impracticable. For example, a service provider is normally required to provide a service with 'reasonable skill and care'. Whether this standard is met requires: (a) identifying facts, such as whether there exist industry practices and/or standards in respect of the service in question; and (b) engaging in legal analysis, such as whether an industry or some other non-governmental standard should (based on the guidance provided by the law) 106 influence the meaning of 'reasonable skill and care'. Therefore, a decision, involving both factual and legal inquiries, would have to fall in its entirety either within or outside the convention. Removing such decisions from the convention would be regrettable because many decisions, especially those involving contract interpretation, comprise questions of fact and law. Many instances involving ED/DBs will not be covered by the convention then, failing to promote ED/DBs sufficiently. Finally, whether an ED/DB decision concerns a factual or legal matter is typically irrelevant for contracting parties as both impact equally on their legal and commercial position. Where the buyer must pay the price against, amongst others, a quality certificate that complies on its face with the sales contract providing for the 'finality' of the certificate, the fact that the factual information in the certificate is mistakenly presented as complying with the contract is of no consolation to the buyer.

D. ED/DB as an ADR Mode and/or as a Gap-Filling Mechanism?
Should the convention cover ED/DBs as an ADR mode and a gap-filling mechanism? Since, as an ADR mode, ED/DBs result in a decision on an existing disagreement/dispute that is functionally equivalent to an arbitration award or a mediation-based settlement agreement, ED/DBs should be covered by the convention. The convention should recognise and enforce agreements to resolve disputes by ED/DB and ED/DB decisions.
The answer is more difficult when ED/DBs are used as a contractual gapfilling mechanism. It can be argued that such an ED/DB decision is not a stand-alone independent 'decision' capable of enforcement but merely a step within the contractual framework to determine what the contract means and/ or how exactly it must be performed. As such, ED/DB decisions cannot be equivalent to an 'enforceable title' and are akin to a contractual provision or an implied term which can only be enforced through a breach of contract claim. Similarly, an agreement on ED/DB as a gap-filling mechanism cannot be recognised and enforced because it merely provides for a way of determining the meaning of the contract and/or how it must be performed.
Including ED/DBs as a contractual gap-filling mechanism in the convention would result in 'special treatment' of contractual terms providing for ED/DBs in comparison with other contractual terms which are not enforceable in themselves under any international convention. In contrast, contractual terms providing for ED/DBs would be so recognised and enforced. The scope of the proposed convention would go further than that of the NYC and SCM. In essence, this convention would be a regime that provides specific performance of contractual obligations to resolve an issue in accordance with an ED/DB clause and to perform the contract in accordance with an ED/DB decision. Not only is it difficult to justify giving such a privilege to a contractual term providing for ED/DBs, but it is also unlikely that States will agree on such an arrangement, as evidenced by the Convention on Contracts for the International Sale of Goods 1980 (CISG) whose drafters could only agree on limited availability of the remedy of specific performance. 107 Why would the international community, that was unable to agree on the ability of the CISG to enforce obligations governed by it, now wish to enforce obligations arising from ED/DB clauses when the convention's focus is not on governing those obligations 108 but on recognising and enforcing agreements on ED/DB and ED/DB decisions?
This convention will also encroach upon substantive contract law and freedom of contract to a greater extent than if it is only concerned with ED/ DB as an ADR mode because it will enable parties to challenge the existence and scope of agreements on ED/DB and ED/DB decisions. By creating such grounds of challenge, the convention would become an international contract law regime limiting contractual freedom insofar as it concerns a choice to use ED/DB as a gap-filling mechanism. It is difficult to see how and why States would sanction such a fragmented but powerful intervention into the contract law domain.
Nevertheless, two arguments can support incorporating ED/DBs as a gapfilling mechanism. One points to the difficulty of conceptualising the precise function of ED/DBs in an individual case. Suppose that a sale contract provides that the parties 'shall make reasonable efforts' or 'are encouraged to agree' on an element in the contract price and if they do not reach an agreement by a certain date, that element shall be fixed by a party-nominated expert in the 'final and binding' manner. It can be argued that no real difference/dispute has arisen between the parties and the ED mechanism lies in between the gap-filling and difference/dispute resolution. Another example is one where an ED/DB decision is provisionally binding and/or a necessary 107 Art 28 CISG. 108 That is, not to govern the formation and content of such agreements.

716
International and Comparative Law Quarterly precondition to another dispute resolution mode. In this case, it can be pointed out that ED/DB is not the final and definitive ADR mode and is no more than a transition to the final dispute resolution. As such, it lies somewhere between the gap-filling and ADR functions. Whilst there may be some force in this argument and some conceptual overlap between the two functions, it should be possible to definitively characterise the intended function of ED/DB. In the first example, the absence of an agreement on the price element by the specified date constitutes a 'difference'. Thus, if an ED/DB clause includes the resolution of a 'difference', 109 the ED/DB fulfils a difference/dispute resolution, not a gapfilling, function. This is so even though the parties' obligations are not as strict as they could have been 110 because that was the parties' chosen standard of duty. The conclusion in the second example depends on whether ED/DB is preconditioned on the parties having to agree on an issue. If so, the conclusion should be the same as in the first example. If not and the relevant issue was simply entrusted to ED/DB, the latter is intended as a gap-filling mechanism even if provisionally.
The second argument, supporting the inclusion of ED/DBs in both their functions, is one of policy, promoting ED/DB in all their manifestations. This policy would emphasise the benefits of preventing disputes. Whilst the NYC and SCM signal the international community's concern with some of these benefits, it remains to be seen whether the benefits of ED/DBs, especially those flowing from their gap-filling function, are appreciated so highly as to lead to their being incorporated in the convention in all their functions. On balance, the arguments against incorporating ED/DBs in their gap-filling function in the convention strongly outweigh those in favour.

E. ED/DB as a Precondition to Arbitration or Litigation
ED/DB may be a contractually required step before the dispute/difference can be finally resolved by arbitration or litigation. These agreements can be of two kinds: (a) an agreement requiring the parties to resort to ED/DBs in the first place before the matter can be submitted to arbitration or litigation; and (b) an agreement giving the parties the right to resort to ED/DB and, if this right is exercised, not allowing the matter to be submitted to arbitration or litigation until and unless the ED/DB process is completed. Should such agreements be recognised and enforced by the convention? The answer depends on whether ED/DB is intended as an (interim) gap-filling or ADR mechanism. If the former, for the reasons given above such agreements should not fall within the convention. If the latter, the answer depends on whether an ED/DB decision is 'provisionally binding' or merely advisory. 109 See (n 115). 110 Compared to, say, the parties' obligation to make 'best' efforts to agree.
As has been argued, there are good reasons why binding decisions, even if provisionally, should be and non-binding decisions should not be covered by the convention. 111 F. Distinguishing ED/DBs from Arbitration ED/DBs as an ADR mode are not very different from arbitration. Their nature is consensual as they come into being by virtue of the parties' agreement. The parties entrust the resolution of their disputes/differences to a third party. Whilst ED/DBs can be provisionally binding, they can equally be final and binding, like an arbitration award. That is why legal systems find it difficult to differentiate clearly between these ADR modes. To be an effective instrument promoting legal certainty internationally, the proposed convention must articulate the criteria differentiating ED/DBs from arbitration. 112 It is helpful to propose the differentiating criteria, all of which should be conceptualised as based on the parties' intentions objectively interpreted: the question is whether the parties intended the agreed ADR mode to be in the nature of arbitration or ED/DB. 113 Various factors relevant to ascertaining this intention will now be analysed. 114 First, the contractual language can make the parties' intention clear, such as where the contract provides that the decision-maker must act 'as an expert, not an arbitrator'. Secondly, there being a 'formulated dispute or difference' between the parties is often seen as an intention for the ADR mode to be arbitration. 115 It is submitted, however, that such an approach would be justifiable if ED/DBs were used exclusively as a gap-filing mechanism. Because ED/DBs are also used as ADR modes, this criterion cannot distinguish them from arbitration and in fact must be a precondition to ED/ DBs. Thirdly, if the contracting parties' intention is to grant the decisionmaker(s) immunity against liability to the parties arising from the discharge of the decision-maker(s)' duties, that may point in favour of arbitration. This is so because some jurisdictions regard arbitrators as acting in a 'quasijudicial' capacity and grant arbitrators certain immunities, 116 whereas 111 See Section V(C) above. 112 A different legal regime will obviously apply, entailing different legal consequences, depending on how the agreed ADR mode is characterised. 113 ). For different approaches in cases to the meaning of a 'dispute', see Chern (n 60) 45-6; for the discussion of a position that a dispute only crystallises if a party's claim is rejected, see Gould (n 63) 462-3. 116 These are jurisdictions, mainly in the common law (eg section 29 Arbitration Act 1996 (UK)), that adhere to the so-called 'status school' of thought, as opposed to those (eg, most continental European jurisdiction) that regard the parties' relationship with arbitrators as being 'contractual',

718
International and Comparative Law Quarterly experts/DBs do not enjoy such status and can incur professional liability to the contracting parties. 117 Fourthly, it may be thought that the intention for the decision-maker to be a lawyer points in favour of the intended mode to be arbitration, whereas the intention for a non-lawyer to resolve the dispute points to ED/DBs because the latter may be perceived to be well suited for resolving technical or factual issues. 118 It is submitted, however, that this cannot be a differentiating factor. Non-lawyers can be appointed as arbitrators, 119 just as lawyers can be appointed for the ED/DB process. But if States wish to preserve their power to regulate which disputes can be resolved by ED/DBs, this should be incorporated in the grounds of challenge and in the provisions on parallel applications or claims and the possibility of stay of proceedings. 120 Fifthly, arbitrators are frequently viewed as acting in a 'judicial' or 'quasijudicial' role. 121 A closely related set of features of arbitration, viewed also as manifestations of a 'judicial' function, is that an arbitrator must: adhere to the principles of natural justice, including affording each party equal treatment and a fair opportunity to present its case; 122 observe due process; 123 abide by the applicable rules of procedure and evidence; 124 make a decision based on the evidence, matters and arguments presented by the parties. 125 The way the practice and domestic laws on ED/DBs have developed can be contrasted with these features of arbitration. ED/DBs are not generally bound by the rules of natural justice or due process. 126 The procedural and evidentiary rules are not applicable to them. Unless the parties have incorporated procedural rules for experts/DBs to follow, experts/DBs will decide on the applicable procedure. Nor must an ED/DB decision be reached based on the evidence, matters and arguments presented by the parties. On the contrary, it is a feature of ED/DBs that experts/DBs can: take into account matters not raised by the parties; carry out their own inquiries and investigations; and, ultimately, reach a decision based on their own opinion, knowledge, judgment and expertise. 127 This spirit of informality and flexibility is explained by the need for ED/DBs to be expert-driven, speedy, informal, relatively non-adversarial and low-cost. Thus, if the parties intended not to follow any procedure and to entrust the chosen expert to decide on procedure, and/or intended the expert to reach a decision based on his/her expertise or judgment, that would point to the ADR mode not being arbitration.
Sixthly, arbitrators are required to be impartial and independent, 128 whereas domestic laws diverge when it comes to ED/DBs. Generally, the common law jurisdictions do not impose these requirements, 129 whereas some civil law jurisdictions require experts/DBs to be impartial and independent. 130 The convention would require a policy decision in relation to whether experts/ DBs ought to be impartial and independent. The SCM subjects mediators to these requirements because 'a failure by the mediator to disclose to the parties circumstances that raise justifiable doubts as to the mediator's impartiality or independence' 131 is, in principle, a ground for refusal to grant relief under the SCM. 132 Put differently, although the SCM is not and the proposed convention would not be directly concerned with obligations of mediators and experts/DBs respectively, to a significant extent they do so through grounds on which relief can be refused, which in turn means that mediators/experts/DBs ought to comply with the criteria in the grounds of challenge. If a choice is made not to incorporate independence and impartiality in the grounds of challenge under the convention, the contractual requirement that a decision must be independent and impartial may point to the intended ADR mode being arbitration.
Seventhly, the need for justice in arbitration normally leads to a requirement that arbitral awards must state their underlying reasons, 133 although parties may be permitted to agree on a non-reasoned award. 134 ED/DB decisions may or may not be required to state reasons. 135 Thus, although in arbitration the need for a reasoned award is the preferred position, 136 there is no principled difference between the two ADR modes in this respect. Finally, the intended 'finality' of the decision of the chosen ADR method is not a differentiating factor because, whilst decisions can be non-final (such as provisionally binding), ED/DB decisions can also be final.
To conclude, whether the chosen ADR mode is arbitration or ED/DB should be based on the objective interpretation of the parties' intention. In determining it, the key factors in favour of its being ED or DB should include: (1) the contractual language to this effect; (2) the absence of the intention to grant immunity to the decision-maker; and (3) evidence that the decision-maker: . was not intended to act in a 'judicial/quasi-judicial' capacity; . was not intended to adjudicate a difference/dispute based on the parties' submissions; . was not intended to observe the rules of natural justice, such as giving each party an opportunity to present its case; . could or was intended to apply his/her own expertise, skill, opinion and judgment to resolve the dispute/or difference. 137 With ED/DB being an ADR method, the existence of a formulated difference/ dispute must be a precondition to ED/DB coming into being and the applicability of the proposed convention. Whether experts/DBs must be subject to the impartiality and independence requirements is an open question to be resolved in the context of the grounds of challenge.

G. ED/DB Distinguished from Mediation
ED/DBs need to be differentiated from mediation, especially bearing in mind the SCM's regime on settlement agreements resulting from mediation. 138 The latter is defined in the SCM as 'a process, irrespective of the expression used or the basis upon which the process is carried out, whereby parties attempt to reach an amicable settlement of their dispute with the assistance of a third person or persons ('the mediator') lacking the authority to impose a solution upon the parties to the dispute'. 139 There are several differences. ED/DBs are not processes by which parties seek to 'reach an amicable settlement' 140 and the third party is not appointed to assist them to reach any such settlement. ED/ DB decisions, if they are final and binding or provisionally binding, are made solely by the expert/DB and resolve the dispute/difference based on the expert's/ DB's judgment. Finally, an ED/DB decision is not an agreement. 141

H. The Convention and Domestic Legislative Regimes
Some common law jurisdictions have enacted legislation providing for resolving disputes between parties to 'a construction contract' by 'adjudication', 142 arguably a form of ED. 143 With varying detail, these regimes regulate different aspects of adjudication. Some jurisdictions have legislation providing for the resolution of disputes by DBs in public construction contracts. 144 Should the proposed convention apply to agreements on ED/DB and ED/DB decisions falling within such legislation?
The answer depends on whether such application is problematic from the perspective of these jurisdictions. One problematic area concerns experts'/adjudicators'/DBs' obligations under such regimes. As has been noted, there is a close link between such obligations and the grounds of challenging ED/DB decisions under the convention. To be meaningful these obligations must correspond to or at least be taken into account in the grounds of challenge. Domestic legislation imposes a different range of obligations on experts/adjudicators/DBs 145 and some of domestic legislation on certain forms of ED, such as adjudication, and DBs, it has been suggested that it is best for the convention to provide an option for Contracting States to make a reservation in relation to ED/DBs falling within such domestic legislation. In a similar vein, because some industries may have a particular interest in this convention, whilst others may not, this article has highlighted the possibility of considering limiting the convention's scope to certain sectors or, in addition to the convention being applicable to all sectors, developing industry-specific protocols to reflect the particular needs of certain industries. Finally, the article has advanced several criteria of what an 'international' ED/DB should mean under the convention.