India stands at a pivotal moment in its campaign to establish itself as a global hub for international commercial arbitration. The project is not merely rhetorical. It is supported by the Arbitration and Conciliation (Amendment) Acts of 2015, 2019 and 2021, by the inauguration of the Arbitration Bar of India in May 2024, and by the establishment of an office of the Permanent Court of Arbitration in New Delhi.1 The message to the international community is that India is open for business and that its dispute-resolution framework is being aligned with global standards.
These efforts remain persistently undermined by a long history of expansive judicial intervention, which has eroded the very advantages that make arbitration an attractive alternative to litigation: efficiency, cost-effectiveness and, most critically, finality. The commercial bargain of arbitration rests on the promise of a swift and conclusive resolution. That promise has been repeatedly broken by a judiciary struggling to define the limits of its supervisory role. The crux of the difficulty lies in the vacillating interpretation of Section 34 of the Arbitration and Conciliation Act, 1996 (“the Act”), which governs the setting aside of arbitral awards. For decades, Indian courts have failed to maintain a consistent and limited approach to review, creating a climate of profound jurisprudential uncertainty.
The judicial pendulum has swung from the broad, interventionist patent illegality doctrine in Oil & Natural Gas Corp. Ltd. v. Saw Pipes Ltd., which permitted a merits-based review, to a pro-arbitration correction in Bharat Aluminium Co. v. Kaiser Aluminium Technical Services Inc. and Ssangyong Engineering & Construction Co. v. National Highways Authority of India.2 More recently, in Gayatri Balasamy v. ISG Novasoft Technologies Ltd., a Constitution Bench of the Supreme Court carved out a limited power for courts to modify awards in exceptional circumstances, introducing yet another layer of judicial creativity.3
In response to this judicially created problem of inconsistent and often excessive review, the Draft Arbitration and Conciliation (Amendment) Bill, 2024 (“the Draft Bill”) proposes a radical and untested solution: the establishment of state-sanctioned appellate arbitral tribunals through a new Section 34A. The Draft Bill was released for public consultation by the Department of Legal Affairs on 18 October 2024, comments closed on 3 November 2024, and it has not since been introduced in Parliament.4 The reform is presented as a means of reducing the burden on national courts and of providing a specialised, expert-driven review mechanism within the arbitral framework itself. It is, however, a perilous inflection point. While ostensibly designed to cure judicial overreach, the appellate arbitral tribunal risks institutionalising the very pathologies it seeks to remedy: delay, cost and uncertainty. It threatens to become a poisoned apple, a seemingly attractive solution that fatally compromises arbitration’s commercial advantages, rather than a beneficial second bite at the cherry.
This paper examines whether a statutory, de facto mandatory appellate layer can be doctrinally reconciled with the fundamental tenets of arbitration. It argues that the proposal, as presently conceived, is a structurally flawed and doctrinally incoherent reform, a legislative overcorrection born of a deep-seated trust deficit in the judiciary’s capacity for self-restraint. By tracing the historical arc of judicial intervention, deconstructing the proposed legislative text, comparing the proposal with international models, and assessing its economic impact, the paper demonstrates that the appellate arbitral tribunal framework is fundamentally at odds with the commercial logic of arbitration. The path chosen, whether to proceed with this litigation-mimicking model or to pursue holistic ecosystem reform, will determine whether India ascends as a global leader in arbitration or remains a jurisdiction in which the long shadow of the courthouse deters the international users it seeks to attract.
The proposal to establish appellate arbitral tribunals cannot be understood in a vacuum. It is a direct legislative reaction to decades of judicial oscillation. The judiciary’s inability to adhere consistently to the principle of minimal intervention, a cornerstone of modern arbitration law, created a jurisprudential crisis of confidence. That crisis prompted the legislature to contemplate a structural overhaul designed to wrest control of post-award review from the courts. The appellate arbitral tribunal is not a proactive, technocratic fix for court backlogs. It is a reactive legislative vote of no confidence in the judiciary’s capacity for self-regulation. The narrative unfolds across three phases: an era of expansive intervention, a period of pro-arbitration correction, and a recent return to judicial creativity that precipitated the present initiative.
The history of arbitration in India before 1996 was largely defined by the Arbitration Act, 1940. That colonial-era statute fostered a culture of excessive judicial intervention.5 Courts could involve themselves at nearly every stage of the arbitral process, from the appointment of arbitrators to the enforcement of the award, which had to be made a rule of court to be effective. The framework was procedurally cumbersome. Arbitral proceedings frequently became a mere prelude to protracted court litigation, defeating the purpose of seeking an alternative to the formal judicial system.
It was against this backdrop that the Arbitration and Conciliation Act, 1996 was enacted. Based on the UNCITRAL Model Law on International Commercial Arbitration, the 1996 Act represented a paradigm shift.6 Its Statement of Objects and Reasons stated that the goal was to minimise the supervisory role of courts in the arbitral process.7 That intent was crystallised in Section 5, which contains a non obstante clause providing that, in matters governed by Part I, no judicial authority shall intervene except where so provided in that Part.8 Judicial intervention was to be the exception, not the rule. The Act was designed to grant finality to awards and to respect the autonomy of parties who had chosen to resolve their disputes outside the court system. That foundational principle is the standard against which all subsequent judicial interpretations must be measured.
The legislative promise of minimal intervention was significantly undermined by the Supreme Court’s 2003 decision in Oil & Natural Gas Corp. Ltd. v. Saw Pipes Ltd.9 The dispute arose from a supply contract in which ONGC withheld liquidated damages for delay. The tribunal held that ONGC had failed to prove actual loss and was therefore not entitled to the damages.10 On a challenge under Section 34, the Court was asked to define the scope of the public policy of India ground for setting aside an award.
Departing from the narrower interpretation in Renusagar Power Co. Ltd. v. General Electric Co., the Court held that a wider meaning was required so as to prevent the frustration of legislation and of justice.11 An award could accordingly be set aside if it was patently illegal, and patent illegality was defined in exceptionally broad terms: an award would be patently illegal if it was contrary to the substantive provisions of Indian law, to the provisions of the 1996 Act, or to the terms of the contract.12 The innovation reintroduced merits-based review through the back door. By allowing courts to set aside an award for an error of law or a perceived misinterpretation of the contract, Saw Pipes empowered the judiciary to act as a de facto court of appeal, second-guessing the arbitrator’s substantive findings.13
Commentators criticised the decision in unusually direct terms. Fali S. Nariman wrote that it had virtually set at naught the entire Arbitration and Conciliation Act of 1996, letting the unruly horse of public policy run free and opening the floodgates for challenges.14 The pendulum had swung decisively towards intervention, and the core commercial advantage of arbitration, a swift and final resolution, was placed in jeopardy.
The interventionist approach of Saw Pipes created significant uncertainty and drew criticism from both domestic and international arbitration communities. A gradual course correction followed. The first significant signal came in 2012 with the Constitution Bench decision in Bharat Aluminium Co. v. Kaiser Aluminium Technical Services Inc. (BALCO).15 The primary holding was jurisdictional: Part I of the Act, including Section 34, does not apply to foreign-seated arbitrations. The underlying philosophy was equally important. By emphasising the territoriality principle and the significance of the arbitral seat, the Court strongly endorsed party autonomy and signalled a move away from the interventionist tendencies of the past.16 BALCO began the realignment of Indian arbitration jurisprudence with international practice.
The judicial shift was followed by a decisive legislative intervention. Acting on the 246th Report of the Law Commission of India, which had heavily criticised the expansive interpretation in Saw Pipes, Parliament enacted the Arbitration and Conciliation (Amendment) Act, 2015.17 The Amendment was a direct legislative rebuke to the Saw Pipes doctrine. It narrowed the operation of the public policy ground, clarifying through express Explanations that the test does not entail a review on the merits of the dispute.18 It also introduced Section 34(2A), which codified patent illegality as a distinct ground but confined its application to arbitrations other than international commercial arbitrations. Crucially, an award could not be set aside merely on the ground of an erroneous application of the law or by reappreciation of evidence.19 This was a clear legislative command to cease reviewing the merits of domestic awards.
The intent and effect of the 2015 Amendment received their full judicial imprimatur in Ssangyong Engineering & Construction Co. v. NHAI.20 The Court held that the expansive interpretation of the fundamental policy of Indian law adopted in Saw Pipes and in ONGC Ltd. v. Western Geco International Ltd. was no longer good law.21 The public policy ground was restricted to the narrow confines articulated in Renusagar. The new patent illegality ground under Section 34(2A) was interpreted strictly: it applies only to illegality going to the root of the matter, and not to a mere erroneous application of the law.22 Ssangyong was widely regarded as the high-water mark of India’s pro-arbitration jurisprudence, the moment when the legislative intent for minimal judicial intervention was finally and authoritatively upheld. The pendulum had swung back towards finality and party autonomy.
Just as a stable jurisprudence seemed to have been achieved, the Supreme Court, in its decision of 30 April 2025 in Gayatri Balasamy v. ISG Novasoft Technologies Ltd., introduced a fresh element of uncertainty.23 The underlying dispute arose from an employment claim in which the tribunal had awarded the claimant damages, and the Madras High Court, rather than confining itself to setting the award aside, twice varied the sum awarded. That approach sat uneasily with Project Director, NHAI v. M. Hakeem, in which the Court had held that a court exercising jurisdiction under Section 34 possesses only the power to set aside an award, not to modify it.24
A five-judge Constitution Bench resolved the conflict by a majority of four to one. The Court held that, although the Act does not expressly confer a power of modification, a narrow and limited power to do so is available under Sections 34 and 37.25 The majority confined that power to severing the invalid portion of an award from the valid, to correcting clerical, computational and typographical errors apparent on the face of the record, and, in defined circumstances, to varying post-award interest; it added that the power under Article 142 of the Constitution must be exercised with great care.26 The reasoning was pragmatic: forcing parties into a fresh round of arbitration to cure a correctable flaw would cause undue procedural hardship.
However carefully circumscribed, Gayatri Balasamy represented a new exercise in judicial creativity. It signalled that, despite the clear legislative boundaries set by the 2015 Amendment, the judiciary remained willing to fashion remedies not expressly provided in the statute. However well intentioned, the decision sent a powerful message to the legislature: the judiciary could not be relied upon to self-regulate its interventionist tendencies. The cycle of judicial expansion, legislative correction and fresh judicial innovation appeared to be repeating itself. Having lost faith in its ability to constrain the judiciary through textual amendments to Section 34, the legislature has now considered a more drastic solution: removing the courts from the first level of review altogether. The appellate arbitral tribunal proposal is the culmination of this fractured dialogue, a legislative attempt to end the uncertainty created by the judicial pendulum once and for all.
The introduction of appellate arbitral tribunals through a new Section 34A would be the most significant structural change to Indian arbitration law since 1996. Presented as a pragmatic solution to judicial backlogs, a forensic reading of the proposed text reveals a mechanism that sits awkwardly with the 1996 Act and is riddled with structural ambiguities. The central flaw is an illusion of choice: ostensibly permissive language conceals what would in practice be a near-universal appellate framework. The tribunal, as engineered, is not a voluntary enhancement of party autonomy. It is a top-down, state-designed tier that alters the nature of arbitration in India and creates direct tension with finality, party autonomy and minimal intervention.
The architecture of the proposed system is an exercise in what may be called coercive permissiveness. The legislature has avoided explicit mandatory language, likely to pre-empt criticism that party autonomy is being violated. Instead, it has engineered a framework in which textual framing, purposive interpretation, judicial doctrine and regulatory pressure combine to compel a result that is, for all practical purposes, mandatory for parties choosing institutional arbitration in India.
The operative clause of the proposed Section 34A provides that an arbitral institution may provide in its rules for an appellate arbitral tribunal to entertain applications made under Section 34.27 On a prima facie reading, “may” suggests discretion and optionality, and the Draft Bill does contemplate party agreement: recourse to the appellate tribunal is available where the parties have agreed to it, and where they have so agreed no application lies to a court.28 The critical question is where that agreement is in practice formed. The discretion to create the mechanism is vested in the arbitral institution, and the parties’ agreement will ordinarily be given long before any dispute has arisen, by the simple act of adopting an institution’s rules. Once an institution incorporates an appellate tribunal into those rules, they bind every party that agrees to arbitrate under its aegis. The choice is therefore made at the institutional level and cascades to the parties as a term of the institutional package rather than as a separately negotiated concession.29 The drafting creates a pathway by which a near-universal regime can be implemented through institutional rules without any requirement of express post-dispute consent. It also leaves ad hoc arbitration outside the scheme altogether, since only a recognised arbitral institution can constitute such a tribunal.
The true import of Section 34A must be read through the purposive interpretation doctrine, which requires a provision to be construed so as to give effect to legislative intent and to remedy the mischief the law seeks to address.30 The mischief, as established above, is excessive, inconsistent and time-consuming judicial intervention at the post-award stage under Section 34. A purely optional appellate tier would be an ineffective remedy. It would create parallel streams of jurisprudence, one from the appellate tribunals and another from the courts. It would spawn a new category of procedural litigation over whether parties had validly consented to the tribunal. Most importantly, it would fail to reduce the burden on the courts, because parties could simply decline the mechanism and proceed directly to a Section 34 challenge. To give effect to the legislative intent of decongesting the judiciary and creating a uniform review process, the provision invites a reading as the exclusive procedural channel for institutional arbitrations.
Indian jurisprudence has a long history of reading the permissive term “may” as the mandatory term “shall” where a power conferred upon a public body is coupled with a public duty, and, more generally, of treating the mandatory or directory character of a provision as a question of purpose rather than of vocabulary.31 The power granted to arbitral institutions under Section 34A is not for their private benefit. It is linked to the public objective of promoting efficiency in dispute resolution and decongesting an overburdened judicial system. Institutions would, in this context, act as delegates of a state-sanctioned reform.
Interpreting “may” as merely directory would also produce unworkable consequences, particularly in relation to the stay of enforcement under Section 36. If the appellate tribunal were optional, a losing party could file an appeal with the tribunal while the winning party simultaneously approached the court for enforcement, forcing the losing party to apply to the court for a stay. Parties would litigate in two forums concurrently, the tribunal on the challenge and the court on the stay, creating procedural confusion and defeating the purpose of the reform.32 On the Draft Bill’s own logic, a mandatory reading is therefore necessary to secure a coherent system in which the appellate tribunal has exclusive jurisdiction over the initial challenge.
Beyond legal doctrine, the regulatory architecture already in place creates a powerful practical mandate. The Arbitration Council of India is charged by statute with grading arbitral institutions by reference to criteria including infrastructure, the quality and calibre of arbitrators, and performance and compliance with time limits.33 The provision of an appellate arbitral tribunal, a mechanism designed by the legislature as the primary solution to a core problem in Indian arbitration, would very likely become a parameter in that grading matrix. No serious institution could readily absorb the reputational and commercial cost of a low grade from the national regulator. Regulatory pressure would therefore push every major institution towards establishing such a tribunal, and parties choosing those institutions would be contractually bound by their rules. The choice is effectively made for the parties by the interplay of regulation and market forces, rendering the permissive language of the provision largely moot.
A framework that is mandatory in practice creates serious tension with the foundational principles of the 1996 Act. First, it confronts Section 35, which provides that an arbitral award shall be final and binding on the parties.34 An appellate tier transforms what was intended to be a final award into a provisional one, subject to a full merits appeal within the arbitral system itself. The result is a concept of provisional finality, which alters the nature of the award and delays its conclusiveness.
Second, it undermines party autonomy, the grundnorm of arbitration.35 True party autonomy would involve parties freely negotiating and agreeing to an appellate mechanism, preferably after a dispute has arisen and they can assess its costs and benefits. An appellate tribunal embedded in institutional rules as a standing feature removes that choice. It is not a product of free contractual negotiation but an imposed element of the institutional framework.
Third, it creates a deep irony with respect to Section 5 and the principle of minimal judicial intervention.36 The proposal seeks to reduce the involvement of courts only by increasing the overall level of intervention in the arbitral process itself. It substitutes one layer of review for another, adding a tier to the dispute-resolution pyramid rather than streamlining it. The cure for excessive review, it seems, is more review.
Beyond these doctrinal clashes, the Draft Bill is plagued by critical structural omissions. The most significant lacuna is its silence on the appellate tribunal’s power to grant a stay of enforcement of the original award under Section 36. The Draft Bill does not amend Section 36, so that power remains with the courts.37 That omission creates the potential for duplicative parallel proceedings, a serious flaw in a reform aimed at efficiency. The Draft Bill also provides no guidance on the constitution, neutrality or qualifications of appellate panellists. The same institution that administered the first-instance arbitration would be responsible for managing the appeal against the award rendered under its auspices, which raises real concerns about institutional bias.38
Most alarmingly, the Draft Bill would extend the patent illegality ground so that it is no longer confined to awards other than those in international commercial arbitration, and would therefore reach awards made in international commercial arbitrations seated in India.39 That would reverse the progress made by the 2015 Amendment and cemented in Ssangyong. It would reintroduce merits-based review for international awards and make India a significantly less attractive and more unpredictable seat for foreign parties. This single provision, if enacted, could on its own derail India’s ambition of becoming a global arbitration hub. The appellate arbitral tribunal, intended to streamline the post-award phase, thus risks substituting judicial delay with a more complex form of procedural confusion, while simultaneously making India an unattractive jurisdiction for international users.
India’s proposal for a state-sanctioned appellate tier is not conceived in a global vacuum. Several leading international arbitral institutions have, for years, offered their own versions of appellate mechanisms. A comparative analysis reveals, however, that while India may be borrowing the structure of an appellate system, it is diverging sharply from the principles that make those international models viable, limited and consistent with the ethos of arbitration. India is attempting to solve a domestic problem of systemic judicial inefficiency with a borrowed international structure designed for the entirely different purpose of party-driven error correction. That mismatch in philosophy and function renders the Indian model an outlier.
The appellate mechanisms offered by institutions such as the American Arbitration Association, JAMS and the International Institute for Conflict Prevention and Resolution evolved from a fundamentally different philosophy.40 They are a market-driven response to a niche demand from sophisticated parties, typically in high-stakes, complex disputes. Those parties, while valuing the efficiency of arbitration, want a contractual safety net, an internal and expert-driven mechanism to correct clear, aberrant and outcome-determinative errors without resorting to the unpredictable and limited review of national courts.41 The mechanisms are, in essence, a product designed to enhance party choice, allowing users to customise their process by trading a degree of finality for a greater chance of substantive correctness.
In stark contrast, the Indian proposal is a top-down, state-designed intervention. It is not a response to market demand from users but a legislative solution to a systemic problem of judicial backlog and jurisprudential inconsistency.42 Its primary objective is not to enhance party autonomy but to re-engineer the State’s dispute-resolution architecture by diverting cases away from the courts. This philosophical chasm between a private, contractual product and a public, systemic solution explains the profound structural differences between the Indian proposal and its international counterparts.
The divergence becomes stark when operational features are compared. The Indian proposal is an anomaly, lacking the essential safeguards that make international appellate mechanisms workable and fair. Table 1 summarises the contrast.
| Feature | Proposed Indian appellate arbitral tribunal | AAA, JAMS and CPR | Singapore (International Arbitration Act) | England (1996 Act) |
|---|---|---|---|---|
| Nature | Internal arbitral appeal | Internal arbitral appeal | Judicial setting aside | Judicial appeal on a question of law |
| Party consent | Agreement given in advance through institutional rules | Strictly opt-in by express agreement | Non-derogable statutory review | Opt-out; parties may exclude |
| Scope of review | Broad Section 34 grounds, including patent illegality | Narrow grounds: material error of law; clearly erroneous findings of fact | Narrow UNCITRAL Model Law grounds | Question of law only, subject to leave |
| Reviewing body | Appellate tribunal within the same institution | Appellate panel drawn from a specialist roster | National courts | National courts |
| Safeguards | No specified timelines; remand not prohibited | Short timelines of twenty-one to ninety days; remand typically prohibited | Three-month time limit | Twenty-eight-day limit; high threshold for leave |
| Philosophy | State-centric: backlog and uniformity | Party-centric: contractual error correction | Pro-arbitration finality | Finality balanced with correctness in English law |
Table 1: Comparative features of appellate and review models
As Table 1 illustrates, the Indian proposal is an outlier on nearly every critical metric. International models are built on explicit, opt-in consent.43 They confine appeals to narrow, specific grounds so as to prevent a full-scale relitigation of the merits: the American Arbitration Association permits an appeal only for an error of law that is material and prejudicial or for determinations of fact that are clearly erroneous, and the CPR procedure is to similar effect.44 They are designed for speed, with strict timelines and, typically, an express prohibition on remanding cases to the original tribunal.45 The Draft Bill lacks these safeguards and thereby opens the door to protracted proceedings.
The relatively low uptake of optional appellate mechanisms in jurisdictions where they are available is itself a powerful market signal.46 The great majority of sophisticated commercial users, when given the choice, prioritise the speed and certainty of a final award over the possibility of correcting an error through an appeal. They understand and accept the commercial bargain of arbitration. The Indian proposal discounts that preference and instead channels parties into a litigation-style review process.
The success of leading international seats is not built on multi-layered review but on a steadfast commitment to finality, enforced by both legislation and the judiciary.
Singapore has risen as a premier global hub not by offering appeals, but by cultivating institutional excellence and judicial restraint. Its International Arbitration Act gives the force of law to the UNCITRAL Model Law and supplements Article 34 with two further grounds concerning fraud or corruption and breach of natural justice.47 The grounds are narrow and directed to jurisdiction, procedural fairness and public policy, and an application must be made within three months of receipt of the award.48 That unwavering commitment to finality supplies the predictability that international users demand.
London offers a different but equally instructive model. Section 69 of the English Arbitration Act 1996 permits a highly restrictive judicial appeal on a question of law.49 An appeal can be brought only by agreement of the parties or with the leave of the court, and leave is granted only if the determination of the question will substantially affect the parties’ rights, the question was one the tribunal was asked to determine, and either the tribunal’s decision is obviously wrong or the question is one of general public importance and the decision is at least open to serious doubt. Any application must be brought within twenty-eight days of the award.50 The right of appeal is not mandatory and is frequently excluded by the adoption of institutional rules; in practice, applications are few and successful ones fewer still.51 When Parliament revisited the 1996 Act in the Arbitration Act 2025, it left Section 69 untouched, the Law Commission having concluded that the provision represents a defensible compromise between the finality of awards and the correction of blatant errors of law.52 The English model thus strikes a careful balance, respecting the finality of the award while providing a judicial backstop for egregious and outcome-determinative errors of law in exceptional cases.
Both models, though different in technique, share a common philosophical commitment: they treat the arbitral award as final and limit recourse to exceptional circumstances. The proposed Indian appellate tribunal does the opposite. It treats the first-instance award as merely provisional and institutionalises the appeal as a routine next step. In doing so, India is not emulating the practices of successful jurisdictions. It is creating a system that is philosophically and functionally their antithesis.
A pragmatic assessment of the proposal reveals its most damaging flaw: its impact on the efficiency calculus for commercial users. Arbitration’s primary value proposition in the commercial world is not its ability to achieve perfect substantive justice, but its capacity to deliver a final and enforceable decision with speed and certainty. By institutionalising a formal appellate layer, the proposed reform misunderstands that economic logic. It risks transforming Indian arbitration into a multi-tiered litigation process, laden with additional costs and delays. The proposal will not make Indian arbitration more attractive. It will actively deter the international users India seeks to court.
The most immediate consequence of the framework is the institutionalisation of delay. The proposal does not replace court review; it precedes it. The result is a new and elongated three-stage process: first, the initial arbitration to obtain an award; second, an appeal to the appellate tribunal on the broad grounds of Section 34; and third, judicial recourse or enforcement, including an appeal from the appellate tribunal’s decision under Section 37, or an application to enforce the award.
This structure undermines the core promise of arbitration as a swift alternative to the Indian court system. What was conceived as a one-stop resolution mechanism is set to become a multi-stage marathon. As Justice Surya Kant, now the Chief Justice of India, observed in September 2025, arbitration in India has itself become a victim of protracted timelines and excessive adjournments.53 An appellate tier would codify that pathology, making delay a feature of the system rather than an aberration within it. For commercial parties, time is money. A process that promises three rounds of adjudication before finality is commercially unviable.
A new procedural tier will inevitably escalate both the direct and the indirect costs of arbitration in India. The direct costs are substantial. Parties will have to pay a fresh panel of appellate arbitrators, who are likely to be senior and expensive practitioners; a second set of institutional administrative charges; and significant additional legal fees for drafting appellate briefs and arguing a second full hearing on the merits. Practitioner estimates already place the legal budget for a mid-sized Indian arbitration in the range of forty to sixty lakh rupees, a figure that an added appellate stage would push sharply higher.54
The indirect costs may be more damaging still. Prolonged delay means that capital, assets and bank guarantees remain tied up for much longer. Business operations are disrupted, investment decisions are postponed, and the opportunity cost of locked-up resources mounts with each passing month. That extended period of uncertainty is a significant economic drag, eroding the financial benefits that a swift arbitral process is supposed to provide.
The proposal misreads the economic principles that underpin the choice of arbitration. Sophisticated commercial parties do not choose arbitration by accident. They make a calculated trade-off. They voluntarily waive the extensive, multi-layered review offered by the judicial system in exchange for what may be termed the finality premium, the commercial value derived from a swift, certain and conclusive decision.55 Finality allows businesses to close their books on a dispute, release contingent liabilities, enforce their rights and redeploy capital into productive activity.
For these users, access to finality is the most crucial component of commercial justice. The proposal is rooted in a different paradigm: the public-law concept of access to justice, which prioritises substantive correctness and provides multiple layers of appeal as a safeguard against error. By imposing that litigation-based model onto a commercial framework, the reform commits a category error. It presumes that commercial parties desire what the State thinks they should desire, namely more opportunities to be proven right, rather than what they have contractually bargained and paid for, namely a quick end to the dispute. This paternalistic approach strips Indian arbitration of its finality premium and devalues the very product that India needs to sell to the global market.
The introduction of an appellate tier is likely to create a bifurcated and fragmented market. Some domestic parties, particularly risk-averse public sector undertakings, may welcome an additional layer of internal review. Their institutional culture often mirrors that of government litigation, prioritising exhaustive review over speed. International commercial parties, the primary target of India’s global hub ambitions, will be strongly deterred. They choose seats such as London, Singapore and Hong Kong precisely because those legal frameworks are perceived as offering supportive and impartial local courts and a reliable record of enforcement.56 The prospect of a three-tiered, multi-year process in India will be a powerful disincentive. The proposed extension of patent illegality to international commercial arbitration would compound the problem, making India an unacceptably unpredictable and interventionist jurisdiction.
The result would be a two-track perception of Indian arbitration: a slow, litigation-like system for domestic disputes and an unattractive, high-risk option for international business. That outcome would be fatal to the goal of attracting foreign direct investment and becoming a preferred seat. In the global competition for arbitration work, certainty and efficiency are the currencies of success. By devaluing both, the proposal threatens to leave India on the sidelines.
The proposed introduction of appellate arbitral tribunals is a solution born of legitimate frustration with judicial inconsistency, but it is a structurally flawed and doctrinally problematic reform. It risks becoming a cure worse than the disease, replacing the uncertainty of judicial review with the certainty of institutionalised delay and cost. If reform is to proceed, it must be radically re-envisioned to align with global practice and with the foundational principles of party autonomy and finality. A more sustainable path lies not in adding layers of review but in strengthening the integrity and quality of the first-instance arbitral process.
Should the legislature remain committed to an appellate arbitral tier, the Draft Bill must be substantially amended. The following changes are not minor tweaks. They are essential guardrails.
Preserve party autonomy by making the mechanism explicitly opt-in. This is the single most critical amendment. The statute should require clear, written consent from all parties to submit to the appellate tribunal’s jurisdiction, and that consent should so far as possible be obtained after the first-instance award has been rendered, so that parties can decide on the basis of the award and the dispute actually before them.
Narrow the grounds for appeal. The scope of review must be delinked from the broad, subjective and litigation-prone grounds of Section 34. The statute should define a new, narrow and exhaustive set of grounds, mirroring those used by international institutions: a material and prejudicial error of law that is central to the award, or a determination of fact clearly unsupported by the record.57 That would prevent the appellate tribunal from becoming a forum for a complete rehearing.
Mandate strict, non-extendable timelines. The international comparators are instructive: an appeal to be filed within thirty days of the award, and a decision to follow within a defined and short period after the appellate panel is constituted.58 A limit of ninety to one hundred and twenty days would be consistent with that practice.
Ensure neutrality and independence. To address the conflict of interest inherent in an institution reviewing awards rendered under its own auspices, the statute should provide for a national, independent roster of qualified appellate arbitrators created and maintained by the Arbitration Council of India rather than by individual institutions.59 Inclusion on the roster should turn on proven appellate experience and a deep understanding of arbitration law.
Confer comprehensive and exclusive powers. To avoid parallel proceedings, the appellate tribunal must be granted the exclusive power to decide applications for a stay of enforcement of the original award under Section 36, which would require an amendment to that section.60
A far more effective path to enhancing India’s status as an arbitration hub lies not in adding layers of review but in strengthening the quality, integrity and finality of the first-instance process. A high-quality initial award is the best antidote to the desire for an appeal.
Strengthen arbitrator quality and accountability. The most effective way to reduce the number of flawed awards is to improve the quality of the arbitrators who render them. The Arbitration Council of India should be equipped to enforce rigorous standards of accreditation, continuing professional development and a strict code of ethics, an approach already recommended by the High Level Committee chaired by Justice B.N. Srikrishna.61 A robust system of training and certification will build user confidence and improve decision-making at the source.
Improve and expedite the enforcement regime. The real test of an arbitration-friendly jurisdiction is how quickly a final award can be converted into a tangible asset. Legislative and judicial effort should be focused on streamlining enforcement under Section 36, including dedicated commercial benches with trained judges and strict timelines for disposal. A swift and certain enforcement process enhances the value of the finality premium.
Build professional and institutional capacity. India must invest in the human capital required to support a world-class ecosystem: a new generation of sophisticated practitioners, a specialist arbitration bar, and well-resourced institutions with high-quality administrative staff.62 Fostering a professional culture that understands and respects the commercial imperative of finality is as important as any legislative reform.
India faces a clear choice. It can proceed with a paternalistic, litigation-mimicking model that sacrifices commercial efficiency for an abstract, state-imposed notion of correctness. Or it can adopt a party-centric, market-driven approach that respects the commercial bargain for finality and invests in the foundational elements of a credible arbitration ecosystem. The path chosen will determine whether India’s ambition to become a global arbitration hub is realised or remains an elusive one.
India’s journey towards becoming a global arbitration powerhouse is at a critical juncture. Legislative amendments and the establishment of specialised institutions reflect a genuine commitment to that goal. Yet this progress is threatened by a proposed reform that, while born of a legitimate frustration with judicial overreach, misunderstands the economic and doctrinal foundations of arbitration. The plan to introduce appellate arbitral tribunals is a perilous overcorrection. It mistakes the symptom, judicial delay and inconsistency, for the underlying disease of judicial interventionism, and in doing so it prescribes a cure that will almost certainly worsen the patient’s condition.
This paper has argued that the proposal is a poisoned apple: a structurally flawed and doctrinally incoherent framework that would institutionalise delay, escalate costs and erode the finality that is the lifeblood of commercial arbitration. The historical analysis reveals a deep-seated trust deficit between the legislature and the judiciary, with the appellate arbitral tribunal representing an attempt to bypass a judiciary no longer trusted to exercise self-restraint. The doctrinal deconstruction of the proposed Section 34A exposes a use of coercive permissiveness, creating a framework that is mandatory in practice and that clashes with finality and party autonomy.
The comparative analysis demonstrates a fundamental mismatch between India’s state-centric proposal and the market-driven, party-centric appellate models that exist internationally. India is attempting to use a tool designed for niche, contractual error correction to solve a systemic problem of judicial inefficiency, a category error that ignores clear market signals from sophisticated users who overwhelmingly prioritise finality over correctness. The economic assessment concludes that, by transforming arbitration into a three-tiered process, the proposal destroys the finality premium, confusing the public-law concept of access to justice with the commercial imperative of access to finality.
India now faces a consequential choice. It can proceed down the path of a paternalistic, litigation-mimicking model that sacrifices the commercial efficiency valued by international business for an abstract, state-imposed notion of substantive correctness. That path leads to a fragmented market, deters foreign investment, and relegates India to the status of a secondary, less-trusted arbitration jurisdiction. Alternatively, India can embrace a party-centric, market-driven approach that respects the commercial bargain for finality. That requires abandoning the appellate arbitral tribunal proposal in favour of holistic reforms that strengthen arbitrator quality, streamline enforcement, and build a professional culture that champions the core values of arbitration. That is the path taken by the world’s leading seats. The choice India makes will determine whether it finally ascends as a global leader in dispute resolution, or remains a jurisdiction where the long and costly shadow of the courthouse continues to deter the very international users it so urgently seeks to attract.
*****
1. See Establishment of Arbitration Bar Aligns with the Vision to Position India as a Premier Destination for Arbitration, Says Dr S. Jaishankar at ABI Launch, SCC Online Blog (May 13, 2024), https://www.scconline.com/blog/post/2024/05/13/arbitrate-in-india-dr-s-jaishankar-at-the-inauguration-of-the-arbitration-bar-of-india/; Establishment of PCA Another Step Towards India Becoming International Arbitration Hub: Justice Surya Kant, India Legal, https://indialegallive.com/constitutional-law-news/courts-news/establishment-of-pca-another-step-towards-india-becoming-international-arbitration-hub-justice-surya-kant/.
2. Oil & Nat. Gas Corp. Ltd. v. Saw Pipes Ltd., (2003) 5 SCC 705; Bharat Aluminium Co. v. Kaiser Aluminium Tech. Servs. Inc., (2012) 9 SCC 552; Ssangyong Eng’g & Constr. Co. v. Nat’l Highways Auth. of India, (2019) 15 SCC 131.
3. Gayatri Balasamy v. ISG Novasoft Techs. Ltd., 2025 INSC 605 (Apr. 30, 2025).
4. Draft Arbitration and Conciliation (Amendment) Bill, 2024, § 34A (India), released for public consultation by the Department of Legal Affairs, Ministry of Law and Justice on Oct. 18, 2024, comments closing Nov. 3, 2024. The Draft Bill had not been introduced in either House of Parliament at the time of writing.
5. The Arbitration Act, No. 10 of 1940, India Code (1940).
6. U.N. Comm’n on Int’l Trade Law, UNCITRAL Model Law on International Commercial Arbitration, U.N. Doc. A/40/17, annex I (1985) (as amended 2006).
7. The Arbitration and Conciliation Act, No. 26 of 1996, Statement of Objects and Reasons, India Code (1996).
8. The Arbitration and Conciliation Act, No. 26 of 1996, § 5, India Code (1996).
9. Oil & Nat. Gas Corp. Ltd. v. Saw Pipes Ltd., (2003) 5 SCC 705.
10. Saw Pipes, (2003) 5 SCC 705.
11. Renusagar Power Co. Ltd. v. Gen. Elec. Co., 1994 Supp (1) SCC 644; Saw Pipes, (2003) 5 SCC 705 (holding that a wider meaning was required so as to prevent frustration of the legislation and of justice).
12. Saw Pipes, (2003) 5 SCC 705 (holding that an award contrary to the substantive provisions of law, to the provisions of the Act, or to the terms of the contract is patently illegal).
13. See Harsh Gagrani & Ritika Jhurani, Unbridled Horse on a Run: A Critique of the Judgment in ONGC v. Saw Pipes (SSRN Working Paper, 2010), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1554313; cf. Sidharth Sharma, Public Policy Under the Indian Arbitration Act: In Defence of the Indian Supreme Court’s Judgment in ONGC v. Saw Pipes, 26(1) J. Int’l Arb. (2009) (defending the decision).
14. Fali S. Nariman, Ten Steps to Salvage Arbitration in India: The First LCIA-India Arbitration Lecture, 27 Arb. Int’l 115 (2011).
15. Bharat Aluminium Co. v. Kaiser Aluminium Tech. Servs. Inc., (2012) 9 SCC 552.
16. BALCO, (2012) 9 SCC 552 (emphasising the seat of the arbitration and the territoriality principle).
17. Law Commission of India, Report No. 246: Amendments to the Arbitration and Conciliation Act, 1996 (2014).
18. The Arbitration and Conciliation (Amendment) Act, No. 3 of 2016, India Code (2016).
19. Id. § 18 (substituting the Explanation to § 34(2)(b) and inserting § 34(2A)).
20. Ssangyong Eng’g & Constr. Co. v. Nat’l Highways Auth. of India, (2019) 15 SCC 131.
21. Ssangyong, (2019) 15 SCC 131; see Oil & Nat. Gas Corp. Ltd. v. W. Geco Int’l Ltd., (2014) 9 SCC 263.
22. Ssangyong, (2019) 15 SCC 131 (confining § 34(2A) to illegality going to the root of the matter and excluding a mere erroneous application of the law).
23. Gayatri Balasamy v. ISG Novasoft Techs. Ltd., 2025 INSC 605.
24. Project Dir., Nat’l Highways Auth. of India v. M. Hakeem, (2021) 9 SCC 1.
25. Gayatri Balasamy, 2025 INSC 605 (majority op.).
26. Gayatri Balasamy, 2025 INSC 605 (confining modification to severable portions of an award, to clerical, computational and typographical errors apparent on the face of the record, and to post-award interest in defined circumstances, and counselling care in the exercise of the power under Article 142).
27. See Appellate Arbitral Tribunals Under Arbitration (Amendment) Bill, 2024: Issues and Challenges, Mondaq, https://www.mondaq.com/india/arbitration-dispute-resolution/1670434/appellate-arbitral-tribunals-under-arbitration-amendment-bill-2024-issues-and-challenges (recording that arbitral institutions “may” provide for an appellate arbitral tribunal to entertain applications under § 34(1)).
28. See Navigating Commercial Disputes: The Draft Arbitration and Conciliation (Amendment) Bill, 2024: An Analysis, Bar & Bench, https://www.barandbench.com/columns/the-draft-arbitration-and-conciliation-amendment-bill-2024-an-analysis (noting that the appellate arbitral tribunal is to hear applications where the parties have agreed to take recourse to it, and that ad hoc arbitrations would remain with the courts).
29. See Ansh Arora & Tanishi Ahuja, India’s Draft Arbitration Bill 2024: Appellate Arbitration and Jurisdictional Reforms, Indian Journal of Arbitration Law Blog (Jan. 13, 2025), https://www.ijal.in/post/india-s-draft-arbitration-bill-2024-appellate-arbitration-and-jurisdictional-reforms.
30. Heydon’s Case (1584) 76 Eng. Rep. 637; Bengal Immunity Co. Ltd. v. State of Bihar, AIR 1955 SC 661.
31. State of U.P. v. Jogendra Singh, AIR 1963 SC 1618 (holding that “may” is capable of meaning “must” or “shall” in an appropriate context); see also Govindlal Chhaganlal Patel v. Agric. Produce Mkt. Comm., (1975) 2 SCC 482 (on the mandatory and directory distinction).
32. See Ishant S. Joshi & Vatsala Tyagi, Appellate Arbitral Tribunals: A Critical Analysis of Section 34A of the Draft Arbitration and Conciliation Amendment Bill, 2024, The Arbitration Workshop (Jan. 5, 2025), https://www.thearbitrationworkshop.com/post/appellate-arbitral-tribunals-a-critical-analysis-of-section-34a-of-the-draft-arbitration-and-concil.
33. The Arbitration and Conciliation Act, No. 26 of 1996, § 43-I, India Code (1996) (inserted as part of Part IA by the Arbitration and Conciliation (Amendment) Act, No. 33 of 2019, and brought into force on Oct. 12, 2023).
34. The Arbitration and Conciliation Act, No. 26 of 1996, § 35, India Code (1996).
35. BALCO, (2012) 9 SCC 552.
36. The Arbitration and Conciliation Act, No. 26 of 1996, § 5, India Code (1996).
37. Joshi & Tyagi, supra note 32 (noting the absence of any provision empowering an appellate arbitral tribunal to stay enforcement under § 36).
38. Joshi & Tyagi, supra note 32 (noting the absence of guidance on the constitution and qualifications of appellate panellists).
39. See Proposed Amendments to the (Indian) Arbitration and Conciliation Act: If It Ain’t Broke, Don’t Fix It? Part II, Bar & Bench (Nov. 2024), https://www.barandbench.com/law-firms/view-point/proposed-amendments-indian-arbitration-and-conciliation-act-part-ii (noting that the proposed provision does not confine patent illegality to awards other than those in international commercial arbitration).
40. See American Arbitration Association, Optional Appellate Arbitration Rules (eff. Nov. 1, 2013); JAMS, Optional Arbitration Appeal Procedure; CPR Dispute Resolution Services, Appellate Arbitration Procedure.
41. R. Platt, The Appeal of Appeal Mechanisms in International Arbitration: Fairness over Finality?, 30 J. Int’l Arb. 531 (2013).
42. See Nihareeka Ghadage & Soham Bhagwat, Draft Arbitration Bill 2024: A Closer Look at Appellate Tribunals and Remand of Partial Awards, IBC Laws, https://ibclaw.in/draft-arbitration-bill-2024-a-closer-look-at-appellate-tribunals-and-remand-of-partial-awards-by-nihareeka-ghadage-soham-bhagwat/.
43. JAMS, supra note 40, R. A(a); see also AAA Adopts Optional Appellate Arbitration Process, Proskauer Rose LLP (Dec. 2013), https://www.proskauer.com/alert/aaa-adopts-optional-appellate-arbitration-process (noting that a party may not unilaterally appeal an award).
44. American Arbitration Association, supra note 40, R. A-10; CPR Dispute Resolution Services, supra note 40, R. 8.2.
45. American Arbitration Association, supra note 40, R. A-3 (appeal to be initiated within thirty days of receipt of the underlying award); JAMS, supra note 40, R. D(e) (decision within twenty-one days absent good cause; no remand to the original arbitrator); see also The Appealing Possibility of Appeals in Arbitration, American Bar Association (Winter 2024), https://www.americanbar.org/groups/litigation/resources/newsletters/appellate-practice/winter2024-appealing-possibility-appeals-arbitration/.
46. Irene Ten Cate, International Arbitration and the Ends of Appellate Review, 44 N.Y.U. J. Int’l L. & Pol. 1109 (2012).
47. International Arbitration Act 1994, § 24 (Sing.); UNCITRAL Model Law, supra note 6, art. 34 (given the force of law by the First Schedule to the International Arbitration Act).
48. UNCITRAL Model Law, supra note 6, art. 34(3).
49. Arbitration Act 1996, c. 23, § 69 (UK).
50. Id. §§ 69(3), 70(3).
51. See Section 69 Almost 20 Years On, Kluwer Arbitration Blog (June 24, 2015), http://arbitrationblog.kluwerarbitration.com/2015/06/24/section-69-almost-20-years-on/.
52. Arbitration Act 2025, c. 4 (UK); see Arbitration Act 2025, Part 3: What Did Not Change?, Mayer Brown (Mar. 2025), https://www.mayerbrown.com/en/insights/publications/2025/03/arbitration-act-2025-part-3-what-did-not-change (recording that no amendment was made to § 69 of the 1996 Act).
53. Arbitration, Once Swift Alternative to Litigation, Now Plagued by Delays: Justice Surya Kant, The Print (Sept. 2025), https://theprint.in/india/arbitration-once-swift-alternative-to-litigation-now-plagued-by-delays-justice-surya-kant/2747968/ (reporting the valedictory address at the third Delhi Arbitration Weekend).
54. See Cost of Arbitration in India: What to Expect?, Amlegals (Sept. 2025), https://amlegals.com/cost-of-arbitration-in-india-what-to-expect/ (reporting legal budgets of forty to sixty lakh rupees for mid-sized disputes).
55. Ten Cate, supra note 46.
56. Queen Mary University of London & White & Case, 2021 International Arbitration Survey: Adapting Arbitration to a Changing World (2021), https://www.qmul.ac.uk/arbitration/research/2021-international-arbitration-survey/ (recording London, Singapore, Hong Kong, Paris and Geneva as the most preferred seats, and identifying support from the local judiciary, neutrality and a reliable enforcement record as the drivers of that preference).
57. American Arbitration Association, supra note 40, R. A-10; CPR Dispute Resolution Services, supra note 40, R. 8.2.
58. American Arbitration Association, supra note 40, R. A-3; JAMS, supra note 40, R. D(e).
59. The Arbitration and Conciliation Act, No. 26 of 1996, § 43-I, India Code (1996).
60. The Arbitration and Conciliation Act, No. 26 of 1996, § 36, India Code (1996).
61. High Level Committee to Review the Institutionalisation of Arbitration Mechanism in India, Report of the High Level Committee (2017) (chaired by Justice B.N. Srikrishna) (recommending the grading of arbitral institutions and the accreditation of arbitrators).
62. High Level Committee to Review the Institutionalisation of Arbitration Mechanism in India, supra note 61.