Richard Spencer BEcon, LLB, MEcStud
Part I: Introduction
Commercial arbitration rests on a fundamental bargain. Parties trade the procedural protections of public courts–including rights of appeal–for speed, confidentiality, party autonomy, and finality. The bargain assumes that parties make this choice knowingly and that the benefits of finality outweigh the risks of unreviewable error. This article examines whether that assumption holds under Australia’s uniform commercial arbitration legislation, with particular reference to the opt-in appeal mechanism in section 34A of the Commercial Arbitration Act 2013 (Qld) and its interstate equivalents.
The tension between finality and justice is not new. The UNCITRAL Model Law on International Commercial Arbitration, adopted in 1985 and revised in 2006, deliberately eschewed appeal rights in favour of narrow grounds for setting aside awards. The drafters’ choice reflected a policy preference: in international commerce, the benefits of a final, enforceable award outweigh the costs of occasional error. Australian legislatures adopted this framework for domestic arbitration through a series of uniform statutes commencing with New South Wales in 2010. But Australia departed from the pure Model Law position by including section 34A–an opt-in appeal mechanism that permits appeals on questions of law where the parties agree and the court grants leave.
The opt-in structure creates what this article identifies as a structural vulnerability. Before a dispute arises, parties may not appreciate the significance of appeal rights or may lack the bargaining power to insist upon them. After an award is made, the prevailing party has no incentive to agree to appellate review of a favourable result. The window for securing appeal rights is, in practical terms, closed once the arbitration concludes. The consequence is that errors of law–even manifest errors–may be unreviewable where the parties did not have the foresight, sophistication, or leverage to negotiate appeal rights ex ante.
This vulnerability is not unique to Australia. The major international arbitration hubs have adopted divergent approaches to appellate review, ranging from England’s opt-out model (where appeals exist unless excluded) to Singapore’s prohibition on appeals for international arbitrations. Each approach reflects different weightings of finality against accountability. But Australia’s opt-in model occupies an unusual position: it formally preserves appeal rights while structurally impeding their exercise.
This article proceeds in six parts. Part II examines the Queensland legislative framework, including the narrow grounds for setting aside under section 34 and the demanding requirements for appeal under section 34A. Part III compares Australia’s approach with those of the major international arbitration hubs: England, Singapore, Hong Kong, and the ICC/Paris framework. Part IV identifies the structural vulnerabilities in the Australian regime: the timing asymmetry that renders post-award agreement practically unobtainable, the informational problems that disadvantage less sophisticated parties, the accountability gap for errors that fall outside the narrow setting-aside grounds, and the conditions that limited oversight creates for arbitral misconduct. Part V addresses the institutional safeguards that the system relies upon to ensure quality despite limited appeal–reasoned awards, institutional scrutiny, professional standards, and conflict controls–and asks whether these mechanisms adequately address the structural risks. Part VI situates the analysis within the scale of international arbitration, examining the sectors dependent on arbitral justice and the trust assumptions that underpin the system. Part VII concludes.
The article’s scope is deliberately diagnostic rather than prescriptive. The question of whether Australia should adopt broader appeal rights, or move toward the pure Model Law position, involves policy considerations beyond the present analysis. The aim here is more modest: to identify the structural risks inherent in the current regime, to examine whether the safeguards are adequate, and to surface questions that warrant scholarly attention. The stakes are considerable. Arbitration has become the dominant mode of resolving international commercial disputes, with major institutions handling cases valued in the tens of billions of dollars annually. The assumption that one unreviewable decision will be correct–or at least not so wrong as to constitute injustice–deserves rigorous examination.
Part II: The Queensland Framework
A. Legislative Foundation and the Model Law
The Commercial Arbitration Act 2013 (Qld) (‘the Act’) commenced operation on 17 May 2013, replacing the Commercial Arbitration Act 1990 (Qld).[1] The Act substantially adopts the UNCITRAL Model Law on International Commercial Arbitration (‘Model Law’), and its explanatory notes emphasise that it ‘closely follows commercial arbitration legislation enacted in other Australian jurisdictions’ to ensure national harmonisation.[2]
The Act’s paramount object, stated in section 1AC, is ‘to facilitate the fair and final resolution of commercial disputes by impartial arbitral tribunals without unnecessary delay or expense’.[3] This formulation reveals the legislative priority: finality is not merely incidental to arbitration but is elevated to a paramount concern. Section 5 reinforces this policy by providing that ‘in matters governed by this Act, no court is to intervene except as provided by this Act’.[4]
B. Section 34: Setting Aside on Narrow Grounds
Section 34 establishes setting aside as the primary recourse against an arbitral award, providing that ‘[r]ecourse to the Court against an arbitral award may be made only by an application for setting aside … or by an appeal under section 34A’.[5] The grounds for setting aside are exhaustively specified and narrowly confined to procedural and jurisdictional defects: incapacity or invalidity of the arbitration agreement; denial of proper notice or inability to present one’s case; the award exceeding the scope of submission; procedural irregularity; non-arbitrability; or conflict with public policy.[6]
Critically, these grounds do not permit review of the merits. An arbitrator who applies the wrong legal test, misconstrues a contract, or draws unsupportable inferences from the evidence has not thereby exceeded jurisdiction or violated public policy. The threshold for establishing the setting-aside grounds has been interpreted restrictively. In TCL Air Conditioner (Zhongshan) Co Ltd v Castel Electronics Pty Ltd, the Full Federal Court held that an award will not be set aside ‘unless there is demonstrated real unfairness or real practical injustice in how the … dispute resolution was conducted or resolved’.[7] This demanding standard–requiring not merely technical irregularity but substantial prejudice–has been consistently applied in Queensland[8] and other Australian jurisdictions.[9] Indeed, the Queensland Court of Appeal has upheld an award where the arbitrator adopted a valuation methodology not advanced by either party, on the basis that the appellant had been afforded an opportunity to respond and its failure to recall expert witnesses was a ‘forensic choice’.[10]
C. Section 34A: The Opt-In Appeal Mechanism
Section 34A provides an avenue for appeal on questions of law, but subject to significant constraints. The section provides that an appeal lies to the Court on a question of law arising out of an award if the parties agree, before the end of the appeal period, that an appeal may be made under the section, and the Court grants leave.[11] Two cumulative conditions must be satisfied: party agreement and judicial leave. The Act explicitly notes that ‘the Model Law does not provide for appeals under section 34A’–signalling that this provision is an Australian departure from the international template.[12]
The requirement for party agreement creates a temporal asymmetry. Agreement must be reached ‘before the end of the appeal period’–that is, within three months of receipt of the award.[13] In practice, parties who did not include an appeal mechanism in their original arbitration agreement face a narrow window in which to secure the opposing party’s consent after the award is made. As commentators observed shortly after the Act’s commencement, ‘it is doubtful that a party who has been successful at arbitration would agree to a review of that award by the Court’.[14] The winning party has no incentive to expose a favourable award to appellate scrutiny. The losing party cannot unilaterally invoke section 34A.
Even where party agreement exists, the Court must grant leave. Section 34A(3) specifies that leave may only be granted if the Court is satisfied that: the determination of the question will substantially affect the rights of one or more parties; the question is one which the tribunal was asked to determine; on the basis of the findings of fact in the award, either the decision is ‘obviously wrong’, or the question is one of ‘general public importance’ and the decision is ‘at least open to serious doubt’; and despite the agreement to arbitrate, it is ‘just and proper’ for the Court to determine the question.[15] As the Supreme Court observed in Mackay Sugar Ltd v Sugar Australia Pty Ltd, ‘[t]he gateway of leave is intended to accord a greater degree of finality to an award made by arbitrators under the Act than exists upon an appeal from a judgment of a court’.[16]
D. Uniformity Across Australian Jurisdictions
The section 34A mechanism is uniform across all Australian states and territories.[17] The South Australian Supreme Court of Appeal confirmed in Inghams Enterprises Pty Ltd v Southern Cross Farms Australia Pty Ltd that ‘the provision in s 34A is uniform across each Australian jurisdiction’.[18] This uniformity is deliberate: the Standing Committee of Attorneys-General developed a Model Commercial Arbitration Bill to achieve national consistency.[19] The consequence is that the structural characteristics of section 34A–the opt-in requirement, the double gateway, the demanding leave threshold–are features of Australian commercial arbitration law generally. The vulnerabilities identified in subsequent Parts are accordingly national, not merely Queensland, concerns.
Part III: Comparative Analysis
The Australian opt-in appeal mechanism is one of several approaches adopted by major arbitration jurisdictions. This Part examines four comparators: England, Singapore, Hong Kong, and the ICC/Paris framework. The comparison reveals that common law jurisdictions have adopted markedly different positions on the finality-accountability spectrum, and that Australia’s approach–while formally preserving appeal rights–structurally impedes their exercise in ways that distinguish it from both the English opt-out model and the pure Model Law approach.
A. England: The Opt-Out Model
The Arbitration Act 1996 (UK) section 69 permits appeals to the court on questions of law arising out of an award, unless the parties have agreed to exclude such appeals.[20] This is an opt-out model: appeal rights exist by default, and parties who wish to secure absolute finality must take positive steps to exclude them. Many institutional rules contain such exclusions, but parties using ad hoc arbitration or rules that do not exclude section 69 retain access to the courts.
The leave threshold under section 69 is demanding, mirroring the Australian section 34A threshold. The practical operation of section 69 suggests that even where appeal rights formally exist, they are rarely exercised successfully. The Commercial Court Reports for 2022–2023 indicate that no section 69 appeals succeeded in either 2021–2022 or 2022–2023, despite approximately 40 applications per year.[21] The Law Commission reviewed the Arbitration Act in 2023 and recommended retention of section 69, and the Arbitration Act 2025, which received Royal Assent on 24 February 2025, maintains section 69 without substantive amendment.[22]
B. Singapore: The Most Restrictive Position
Singapore has adopted the most restrictive approach to appellate review among the major common law arbitration hubs. The International Arbitration Act 1994 (Singapore) provides no right of appeal on questions of law for international arbitrations.[23] For parties who have chosen Singapore as their seat for international arbitration, there is no path to curial review of legal error, however manifest. The position differs for domestic arbitrations, where the Arbitration Act 2001 (Singapore) permits appeals on questions of law, though parties may exclude this right.[24]
This bifurcation reflects a deliberate policy choice. Singapore’s emergence as a leading arbitration hub has been built partly on its reputation for non-intervention in arbitral awards. The Ministry of Law consulted in 2019 on whether to introduce an opt-in appeal mechanism for international arbitrations, and the Singapore Academy of Law’s Law Reform Committee recommended such a mechanism.[25] The recommendation was not implemented. Singapore has thus consciously rejected even the limited appeal rights that Australia provides.
C. Hong Kong: Opt-In Under a Merged Regime
Hong Kong occupies an intermediate position. The Arbitration Ordinance (Cap 609) introduced a single merged regime for domestic and international arbitrations in 2011, replacing the previous dual-track system.[26] The Ordinance permits appeals on questions of law where the parties expressly agree–an opt-in model similar to Australia’s section 34A.[27] The retention of opt-in appeal rights in Hong Kong was influenced by the construction industry, which had been accustomed to appeal rights under the previous domestic arbitration regime.[28]
D. Paris and the ICC: Pure Model Law Adherence
France, as the seat of the International Chamber of Commerce, adheres closely to the Model Law position. French law provides no appeal on the merits or on points of law from arbitral awards.[29] The ICC compensates for the absence of appellate review through institutional mechanisms: ICC awards are subject to scrutiny by the ICC Court before issuance–a form of quality control that reviews awards for formal defects and may draw the tribunal’s attention to substantive issues, though the Court cannot amend the award’s substance.[30]
E. Summary and Implications
The comparative analysis reveals a spectrum of approaches. England adopts an opt-out model where appeals exist unless excluded. Singapore prohibits appeals for international arbitrations entirely, while permitting them for domestic disputes. Hong Kong and Australia adopt opt-in models where appeals are available only by party agreement. France and the ICC adhere to the pure Model Law position with no appellate review.
Australia’s opt-in model is formally more permissive than Singapore’s prohibition but structurally more restrictive than England’s opt-out approach. The critical difference is the allocation of the burden of action. Under the English model, a party who wants finality must act (by excluding section 69). Under the Australian model, a party who wants appeal rights must act (by securing agreement to section 34A). Given the timing asymmetry–the difficulty of securing post-award agreement from a prevailing opponent–the Australian model tends toward finality in practice even where it formally preserves the possibility of appeal.
Part IV: Structural Vulnerabilities
The preceding Parts have established the doctrinal framework and comparative context. This Part identifies five structural vulnerabilities in the Australian opt-in appeal regime: timing asymmetry, informational problems, the accountability gap, the ‘dead-end’ scenario, and conditions conducive to misconduct.
A. Timing Asymmetry
The most significant structural problem with section 34A is temporal. Agreement to appeal rights must be reached ‘before the end of the appeal period’–that is, within three months of receipt of the award.[31] In practice, this creates two distinct windows for securing appeal rights, each with characteristic difficulties.
The first window is pre-dispute. Parties negotiating an arbitration agreement may include a provision permitting appeals under section 34A. But at this stage, neither party knows whether it will be the claimant or respondent in any future dispute, let alone whether it will prevail. The incentive to secure appeal rights depends on risk preferences and bargaining position, not on any concrete assessment of the dispute that will eventually arise. Parties focused on the commercial terms of their transaction may treat dispute resolution clauses as boilerplate.
The second window is post-award. Here the asymmetry becomes acute. The prevailing party has obtained a binding award in its favour. It has no incentive–and would be poorly advised–to expose that award to appellate scrutiny. The losing party, who might benefit from appeal rights, cannot invoke them unilaterally. The three-month window is, for practical purposes, closed.
B. Informational Problems
The timing asymmetry intersects with informational asymmetries between different categories of parties. Large corporations with experienced in-house counsel and access to specialist arbitration practitioners are well-positioned to appreciate the significance of appeal rights and to negotiate express provisions in their arbitration agreements. These ‘repeat players’ understand the system and can structure their contracts accordingly.
Small and medium enterprises, by contrast, typically use standard form contracts without modification. They may not appreciate that an arbitration clause excludes, by default, the appeal rights they would enjoy in litigation. Consumer and small business parties are most disadvantaged: they are unlikely to negotiate dispute resolution terms at all. This creates a systematic pattern: sophisticated parties can secure appeal rights if they want them; unsophisticated parties cannot. The opt-in structure thus tends to benefit repeat players at the expense of one-shot participants.
C. The Accountability Gap
The combination of section 34’s narrow grounds and section 34A’s structural inaccessibility creates an accountability gap. Section 34 addresses procedural failures, jurisdictional defects, and public policy violations. It does not permit review of the merits. Section 34A addresses errors of law, but only where the parties agreed to appeal rights.
The gap between these provisions is substantial. It includes: errors of law where no appeal was agreed; errors of fact; unreasonable findings; and arbitrary or capricious decisions that do not rise to the level of ‘real unfairness or real practical injustice’.[32] In each of these categories, the aggrieved party has no remedy. The award stands, enforceable under Division 7 of the Act, with no path to correction.
D. The Dead-End Scenario
The accountability gap produces what might be called the ‘dead-end’ scenario: a case where an arbitrator makes a manifest error, the parties did not agree to section 34A appeal rights, and the error does not engage any of the narrow section 34 grounds.
Consider a case where an arbitrator determines a contractual valuation dispute by adopting a methodology neither party advanced and that was not put to the expert witnesses during their evidence. The methodology departs from what a court would consider permissible. But the arbitrator raises the issue during final submissions, the aggrieved party makes submissions in response, and the party’s ‘forensic choice’ not to recall its experts forecloses any claim that it was denied an opportunity to be heard. The error does not constitute an inability to present one’s case; it does not exceed jurisdiction; it does not violate public policy; and it cannot be appealed because the parties did not agree to section 34A rights. The result is that the arbitrator’s decision is final and binding. There is no remedy.
E. Conditions Conducive to Misconduct
The limited oversight created by sections 34 and 34A raises a further concern: the conditions it creates for arbitral misconduct, whether through partiality, corruption, or capture. Arbitration is a private system: proceedings are confidential, awards are generally not published, there is no body of precedent constraining arbitral decision-making, and there is limited external review.
In such a system, the incentives operating on arbitrators deserve scrutiny. Arbitrators are appointed by parties and their continued career depends on continued appointments. Repeat players are more valuable sources of appointments than one-shot participants. These dynamics are well-recognised in the literature.[33] The ICC Task Force on Corruption in International Arbitration has acknowledged that limited review of awards can be exploited by unscrupulous parties and arbitrators.[34] The point is structural: limited oversight creates conditions in which misconduct, if it occurs, is less likely to be detected and corrected.
Part V: Institutional Safeguards and their Limits
The preceding Part identified structural vulnerabilities in the Australian opt-in appeal regime. A fair assessment requires engagement with the safeguards the system relies upon to ensure quality despite limited appellate review. This Part examines four categories of safeguard: transparency mechanisms, professional standards, conflict and bias controls, and the enforcement gateway.
A. Transparency Mechanisms
The Act requires arbitrators to provide reasoned awards. Section 31(2) provides that the award ‘shall state the reasons upon which it is based’ unless the parties have agreed otherwise.[35] Reasoned awards serve several functions: they discipline arbitral reasoning, enable parties to understand the basis for decisions, and provide a foundation for any challenge. Some institutions supplement statutory requirements with additional quality controls. ICC awards are subject to scrutiny by the ICC Court before issuance.[36] A growing number of institutions publish anonymised awards, creating a body of arbitral precedent.
The limits of these mechanisms should be acknowledged. Reasoned awards are only as disciplining as the arbitrator’s commitment to rigorous reasoning. Institutional scrutiny is ex ante, not ex post, and cannot correct errors that survive the scrutiny process. Published awards remain a minority of total awards.
B. Professional Standards and Reputational Accountability
Arbitrators are subject to professional standards enforced through multiple channels. Many arbitrators are members of professional bodies such as the Chartered Institute of Arbitrators or the Resolution Institute, which maintain codes of conduct and disciplinary procedures. The arbitration community is small and reputation-dependent. An arbitrator who develops a reputation for poor-quality awards is unlikely to receive future appointments.
The limits are equally apparent. Disciplinary procedures are rarely invoked. Reputational accountability operates with a lag and creates its own perverse incentive: the desire for future appointments may lead arbitrators to favour repeat players who are more likely to appoint them again.
C. Conflict and Bias Controls
The Act provides mechanisms to address arbitrator partiality. Section 12 permits a party to challenge an arbitrator where ‘circumstances exist that give rise to justifiable doubts as to the arbitrator’s impartiality or independence’.[37] Arbitrators are subject to disclosure obligations, and the IBA Guidelines on Conflicts of Interest provide a widely-adopted framework.[38] The structure of three-member tribunals provides an additional check.
These controls address disclosed or discoverable conflicts but are less effective against undisclosed relationships or subtle forms of cognitive bias. A challenge under section 12 requires the challenging party to identify and prove the circumstances giving rise to justifiable doubts–a burden that may be difficult to discharge.
D. The Enforcement Gateway
Courts retain supervisory jurisdiction at the enforcement stage. Even after the three-month window for setting aside has closed, a party against whom enforcement is sought may resist enforcement on grounds substantially similar to the setting-aside grounds.[39] The New York Convention preserves these defences for international awards.[40] However, the enforcement gateway is narrow: the grounds for refusing enforcement mirror the setting-aside grounds and do not permit review of errors of law or fact.
E. Are the Safeguards Adequate?
The safeguards described above are real and meaningful. They provide layers of quality control, professional accountability, and structural balance. The question is whether they adequately address the structural vulnerabilities. This article suggests that the safeguards operate effectively in the core case: sophisticated parties, experienced arbitrators, well-functioning institutions. But they are weaker at the margins: for unsophisticated parties who cannot evaluate arbitrator quality; against arbitrators whose reputations are established; in ad hoc arbitrations outside institutional frameworks; and against errors apparent only on legal analysis–precisely the errors appellate review is designed to correct.
Part VI: The Stakes
The structural analysis would be of limited significance if arbitration were a marginal mode of dispute resolution. It is not. International commercial arbitration has become the dominant mechanism for resolving cross-border commercial disputes, and domestic arbitration plays an increasingly significant role in sectors ranging from construction to financial services.
A. The Scale of International Arbitration
The major arbitration institutions handle thousands of cases annually, with aggregate sums in dispute measured in the tens of billions of dollars. The Singapore International Arbitration Centre reported 625 new cases in 2024, involving parties from 72 jurisdictions, with a total sum in dispute of US$11.86 billion.[41] The International Chamber of Commerce typically handles over 800 cases annually. A 2018 study estimated the total value of disputes in international arbitration at approximately US$86 billion annually.[42]
B. Dependent Sectors
Certain sectors are particularly dependent on arbitration as their primary dispute resolution mechanism. Construction and engineering disputes, which comprise over 45% of ICC caseload, are overwhelmingly resolved through arbitration.[43] Energy and resources disputes follow a similar pattern. International trade relies heavily on arbitration for disputes under sale of goods contracts and distribution agreements. Maritime and shipping disputes have their own specialist arbitration ecosystem. For parties in these sectors, arbitration is not an optional alternative to litigation; it is the expected and often the only realistic mode of dispute resolution.
C. Trust Assumptions
The arbitration system operates on trust assumptions that deserve articulation. Parties who agree to arbitrate assume that arbitrators will apply the law correctly, that procedural fairness will be observed, that awards will reflect the merits of the dispute, and that institutional mechanisms will provide quality control. Limited appellate review is premised on the assumption that the one shot will be a good shot. If that assumption fails in a material proportion of cases, the legitimacy of the system is called into question.
D. Hub Ambitions and Regime Design
The structural analysis has implications for Australia’s ambitions as a regional arbitration hub. Queensland and Western Australia together generate the majority of international arbitration activity in Australia, driven by the natural resources, energy, and construction sectors that dominate their economies.[44] Brisbane, alongside Sydney, Melbourne, and Perth, is promoted as a ‘key seat’ offering time zone convenience with East Asia.[45] Yet Australia’s market share in Asia-Pacific arbitration remains modest; surveys suggest that an Australian seat is ‘typically requested’ in only a small fraction of regional disputes.[46]
Singapore has built its position as the leading Asia-Pacific hub partly by marketing absolute finality–the complete absence of appeals for international arbitrations. One practitioner guide notes that Singapore ‘recognizes the finality of arbitral awards and limits the grounds for appeal’, which ‘reduces a successful party’s risk of having the arbitral award set aside/reversed on appeal’.[47] England, by contrast, retains appeal rights by default, offering a safety valve that some parties value. Australia’s opt-in model occupies an unstable middle position: less certain finality than Singapore (where there is no ambiguity about the absence of appeals), less reliable protection than England (where parties automatically have recourse unless they exclude it).
Whether this regime design serves Australia’s hub ambitions–or impedes them–is a question that policymakers might usefully consider alongside the structural concerns identified in this article. A party who wants absolute finality may prefer Singapore precisely because there is no possibility of appeal; a party who wants the reassurance of appeal rights can obtain them automatically in London. Australia’s middle position may satisfy neither preference fully, which could contribute to the limited uptake of Australian seats despite the jurisdiction’s other advantages: a common law system, pro-enforcement courts, and geographic proximity to Asian markets.
Part VII: Conclusion
This article has examined the structural vulnerabilities in Australia’s opt-in appeal mechanism for commercial arbitration awards. Part II established the doctrinal framework: section 34 provides narrow grounds for setting aside; section 34A provides for appeals on questions of law, but only where the parties agree and the court grants leave. The ‘double gateway’ structure is uniform across all Australian jurisdictions.
Part III compared Australia’s approach with the major international arbitration hubs. England’s opt-out model provides appeal rights by default; Singapore prohibits appeals for international arbitrations entirely; Hong Kong and Australia adopt opt-in models; France and the ICC adhere to pure Model Law finality. Australia’s position–formally preserving appeal rights while structurally impeding their exercise–occupies an unstable middle ground.
Part IV identified five structural vulnerabilities: timing asymmetry that renders post-award agreement practically unobtainable; informational problems that disadvantage unsophisticated parties; an accountability gap for errors falling outside the narrow statutory grounds; dead-end scenarios where manifest errors are unreviewable; and conditions conducive to arbitral misconduct in an environment of limited oversight.
Part V examined the institutional safeguards the system relies upon and concluded that while these mechanisms are meaningful, they do not fully address the structural vulnerabilities. Part VI situated the analysis within the scale of international arbitration, the trust assumptions that underpin the system’s legitimacy, and the implications for Australia’s ambitions as a regional arbitration hub.
The analysis has been deliberately diagnostic rather than prescriptive. Whether Australia should move toward the English opt-out model, the Singaporean prohibition, or retain its current structure is a policy question involving considerations beyond this article’s scope. What the analysis demonstrates is that the current structure creates risks: risks that manifest errors will go uncorrected, that unsophisticated parties will be disadvantaged, and that limited oversight will create conditions in which misconduct, if it occurs, will be difficult to detect.
These risks warrant attention from practitioners, parties, and policymakers. Practitioners advising on arbitration agreements should ensure their clients understand the implications of the opt-in structure and consider whether to negotiate express appeal rights. Parties entering arbitration should appreciate that the ‘one shot’ they are taking may be unreviewable. And policymakers should consider whether the current balance between finality and accountability serves the interests of commercial parties and the integrity of the dispute resolution system.
The stakes are considerable. Arbitration has become the dominant mode of resolving international commercial disputes. The assumption that one unreviewable decision will be correct–or at least not so wrong as to constitute injustice–is an assumption on which billions of dollars and countless commercial relationships depend. That assumption deserves rigorous and continuing examination.
Footnotes
[1] Commercial Arbitration Act 2013 (Qld) s 41.
[2] Explanatory Notes, Commercial Arbitration Bill 2012 (Qld) 1.
[3] Commercial Arbitration Act 2013 (Qld) s 1AC(1).
[4] Ibid s 5.
[5] Ibid s 34(1).
[6] Ibid s 34(2)(a)–(b).
[7] TCL Air Conditioner (Zhongshan) Co Ltd v Castel Electronics Pty Ltd [2014] FCAFC 83, [55] (Allsop CJ, Middleton and Foster JJ).
[8] Clarke Energy (Australia) Pty Ltd v Power Generation Corporation [2025] QSC 64, [42] (Kelly J).
[9] Amasya Enterprises Pty Ltd v Asta Developments (Aust) Pty Ltd [2016] VSC 326, [47] (Croft J).
[10] Mango Boulevard Pty Ltd v Mio Art Pty Ltd [2018] QCA 39, [82] (Morrison JA).
[11] Commercial Arbitration Act 2013 (Qld) s 34A(1).
[12] Ibid s 34(1) n.
[13] Ibid s 34A(6).
[14] Emma Schilling and Sam Hibberd, ‘Choose Your Own Adventure: The New Commercial Arbitration Act 2013 (QLD)’ (2013) Lexology (online).
[15] Commercial Arbitration Act 2013 (Qld) s 34A(3)(a)–(d).
[16] Mackay Sugar Ltd v Sugar Australia Pty Ltd [2013] QSC 233, [30] (P McMurdo J).
[17] Commercial Arbitration Act 2010 (NSW) s 34A; Commercial Arbitration Act 2011 (Vic) s 34A; Commercial Arbitration Act 2011 (SA) s 34A; Commercial Arbitration Act 2012 (WA) s 34A; Commercial Arbitration Act 2011 (Tas) s 34A; Commercial Arbitration (National Uniform Legislation) Act 2011 (NT) s 34A; Commercial Arbitration Act 2017 (ACT) s 34A.
[18] Inghams Enterprises Pty Ltd v Southern Cross Farms Australia Pty Ltd [2022] SASCA 7, [45] (Doyle JA, Livesey and Bleby JJA agreeing).
[19] Standing Committee of Attorneys-General, Model Commercial Arbitration Bill (2010).
[20] Arbitration Act 1996 (UK) s 69(1).
[21] Commercial Court, Commercial Court Report 2022–2023 (Judiciary of England and Wales, 2023) Table 7.
[22] Arbitration Act 2025 (UK), received Royal Assent 24 February 2025.
[23] International Arbitration Act 1994 (Singapore) s 24.
[24] Arbitration Act 2001 (Singapore) s 49.
[25] Singapore Academy of Law, Law Reform Committee, Report on Right to Appeal Against International Arbitration Awards on Questions of Law (2019).
[26] Arbitration Ordinance (Cap 609) (Hong Kong).
[27] Ibid s 5, Sch 2.
[28] See Legislative Council Brief, Arbitration Bill (2010) (Hong Kong).
[29] Code de procédure civile (France) art 1518.
[30] ICC Arbitration Rules 2021, art 34.
[31] Commercial Arbitration Act 2013 (Qld) s 34A(6).
[32] TCL Air Conditioner (Zhongshan) Co Ltd v Castel Electronics Pty Ltd [2014] FCAFC 83, [55].
[33] See, eg, Catherine Rogers, Ethics in International Arbitration (Oxford University Press, 2014) ch 7.
[34] ICC Task Force on Corruption, Report: Using Red Flags to Identify Corruption in International Arbitration (2024).
[35] Commercial Arbitration Act 2013 (Qld) s 31(2).
[36] ICC Arbitration Rules 2021, art 34.
[37] Commercial Arbitration Act 2013 (Qld) s 12(2).
[38] International Bar Association, IBA Guidelines on Conflicts of Interest in International Arbitration (2014).
[39] Commercial Arbitration Act 2013 (Qld) ss 35–36.
[40] Convention on the Recognition and Enforcement of Foreign Arbitral Awards, opened for signature 10 June 1958, 330 UNTS 3 (entered into force 7 June 1959) art V.
[41] Singapore International Arbitration Centre, Annual Report 2024 (2025).
[42] PwC and Queen Mary University of London, International Arbitration Survey: The Evolution of International Arbitration (2018).
[43] International Chamber of Commerce, Dispute Resolution Statistics (2023).
[44] Chambers and Partners, International Arbitration 2020: Australia (2020).
[45] DLA Piper, ‘Doing Business in Australia: Why International Commercial Arbitration is an Attractive Option to Resolve Disputes’ (2023).
[46] Australian Disputes Centre, ‘International Arbitration between Hong Kong and Australia’ (2023).
[47] K&L Gates, ‘International Arbitration and the Singapore International Arbitration Centre’ (2022).