Reimagining Domestic Arbitration in India: A Hybrid Appellate Model

Domestic arbitration in India appears to be at a crossroads. The Arbitration and Conciliation Act, 1996 (Arbitration Act) was enacted to provide a modern, efficient and commercially responsive framework for dispute resolution, but recent developments suggest that public confidence in the regime is facing serious challenges.

Governmental bodies are reducing their reliance on arbitration, and arbitration jurisprudence remains uncertain. These developments do not establish that arbitration has failed, but they warrant closer examination of whether the framework delivers the certainty, efficiency and integrity that parties expect.

The response should not be retreat. Arbitration retains significant advantages over litigation, including specialised adjudication, procedural flexibility and cost and time efficiency. The better question is whether the framework can accommodate greater judicial scrutiny where parties want it.

This article proposes a hybrid appellate model founded on party autonomy. Parties would choose at the outset between the existing limited-intervention framework and an alternative route providing a single appeal on procedure and merits, consolidating the existing post-award challenge and appeal mechanisms rather than adding another layer of proceedings.

i. Concerns about Integrity

Government bodies are increasingly stepping away from arbitration. In June 2024, the Ministry of Finance issued an Office Memorandum describing arbitration as time-consuming and a drain on the exchequer’s money, and restricting arbitration to disputes under INR 10 crore. The Public Works Department of the Government of NCT of Delhi followed with a circular dated 21 April 2025 mandating deletion of arbitration clauses from the General Clauses of Contract, effectively making litigation the preferred mode of dispute resolution.

The most significant development is the circular issued by the Ministry of Road Transport and Highways (“MoRTH”) on 12 January 2026. Applying to BOT (Toll), HAM and EPC contracts for national highway projects and NHAI standard procurement contracts, it mandates administrative settlement as the primary mechanism, and disfavours arbitration, where the aggregate claim exceeds INR 10 crore. It applies to contracts executed after 12 January 2026, and existing contracts are to be amended by supplementary agreements. As one of the largest participants in India’s infrastructure ecosystem, MoRTH is the strongest indicator of the trend.

No single circular proves that arbitration is failing, and the MoRTH direction is merely administrative. But when multiple government stakeholders across sectors independently move away from arbitration, signalling concerns about the arbitral process and arbitrators, the pattern is difficult to dismiss.

ii. Inconsistency in Arbitration Jurisprudence

At the GHAC Arbitration Week 2026, held from 4 to 6 September 2026 at the GIFT City Club in Gandhinagar, Justice R.F. Nariman identified inconsistency in Indian arbitration jurisprudence as a significant concern in his keynote address. He observed that several Supreme Court decisions arose because uncertain questions of law required reconsideration by larger Benches, a pattern particularly problematic for a system meant to be efficient and predictable.

If legal principles repeatedly change, parties face uncertainty about both the conduct and outcome of proceedings.

The most striking illustration is whether an unstamped instrument containing an arbitration agreement can be enforced. SMS Tea Estates v. Chandmari Tea Co., held that it cannot be acted upon, followed by Garware Wall Ropes v. Coastal Marine Constructions & Engg. Ltd. A three-judge bench in N.N. Global Mercantile v. Indo Unique Flame, overruled SMS Tea Estates, holding the arbitration agreement independent through separability. A five-judge bench in N.N. Global Mercantile, reversed course, holding that an unstamped instrument does not exist in law. Finally, a seven-judge bench in Re: Interplay Between Arbitration Agreements and Indian Stamp Act, held that an arbitration agreement is not rendered void or unenforceable because it is unstamped. This back-and-forth over more than a decade, through ever larger benches, shows how difficult certainty is to achieve.

Justice Nariman also criticised DMRC Ltd. v. Delhi Airport Metro Express, where “miscarriage of justice” was invoked at the curative stage to revisit factual findings in an arbitral award after statutory challenges were exhausted. Citing Rupa Ashok Hurra v. Ashok Hurra, he stressed that curative jurisdiction is meant for exceptional situations such as violation of natural justice or bias, not the merits. Using it to re-examine factual findings takes away the finality the statutory framework promises.

In Gayatri Balasamy v. ISG Novasoft Technologies, the question was whether “set aside” under Section 34 includes a power to modify an award; Justice Nariman questioned this expansion, given the arbitrator’s corrective powers under Section 33.

The problem extends beyond these examples. On pre-deposits, S.K. Jain v. State of Haryana, a three-judge bench, upheld a 7% pre-deposit clause, but ICOMM Tele Ltd. v. Punjab State Water Supply and Sewerage Board, a two-judge bench, struck down a 10% clause. This was reaffirmed in Lombardi Engineering Ltd. v. Uttarakhand Jal Vidyut Nigam and accepted in Santosh Associates v. Haryana State Industrial and Infrastructure Development Corporation, which noted that S.K. Jain had higher bench strength and referred the matter to a larger bench. The issue remains unresolved.

Seat and venue are similarly uncertain. In Union of India v. Hardy Exploration, a three-judge bench held that designating a venue does not automatically designate a seat. But in BGS SGS Soma JV v. NHPC Ltd, another three-judge bench held that a venue is the seat unless the contrary is indicated, creating ongoing confusion on a fundamental question.

Finally, the referral court’s powers under Sections 8 and 11 remain unsettled. Vidya Drolia v. Durga Trading held them complementary, applying a prima facie test to weed out “ex-facie non-arbitrable and frivolous disputes.” But SBI General Insurance v. Krish Spinning held this approach “legislatively overruled” by the 2015 Amendment to Section 11(6-A), limiting the court to the prima facie existence of the arbitration agreement. The two decisions pull in opposite directions.

These instances underscore the need for greater judicial discipline. While reconsideration by larger Benches is inherent in a hierarchical system, inconsistency between courts of coordinate jurisdiction creates avoidable uncertainty. Where a larger Bench has settled a position, later decisions should ordinarily follow it absent compelling grounds for reconsideration through the appropriate judicial process.

Identifying these problems is necessary, but abandoning arbitration would be a mistake. It reduces the burden on an overburdened judiciary, provides adjudicators with specialised expertise, offers procedural flexibility, and, where it works as intended, resolves commercial disputes more efficiently than litigation.

The MoRTH circular highlights the seriousness of the confidence deficit; it is not a model to follow. Moving high-value disputes to administrative settlement or civil courts merely shifts the problems to a slower and less specialised forum. The more productive response is to ask what reforms might restore confidence.

Major arbitral institutions have developed optional appellate mechanisms going well beyond conventional judicial review. The International Institute for Conflict Prevention and Resolution (CPR) introduced its Arbitration Appeal Procedure in 1999, and the International Centre for Dispute Resolution (ICDR) introduced its Optional Appellate Arbitration Rules in 2013, allowing appeal for material and prejudicial errors of law or clearly erroneous determinations of fact.

Where the parties agree, either in their arbitration clause or post-dispute, an appellate panel may affirm, reverse or substitute the award within a compressed timeframe. This is full merits review of law and fact. These rules developed in response to the lack of appealability of awards under statutes such as the U.S. Federal Arbitration Act (FAA), as affirmed in Hall Street Associates, L.L.C. v. Mattel, Inc., which held that parties could not contract for expanded judicial review beyond the FAA’s narrow grounds.

These models show that substantive appellate review can coexist with arbitration’s private nature, balancing finality with the parties’ preference for an additional layer of review.

Justice Nariman’s keynote included a concrete proposal: a full appeal on facts and law, at one stage, before a Division Bench of the High Court. His reasoning was that modern arbitration increasingly involves complex, high-value disputes between sophisticated parties, where accuracy is as important as speed and efficiency.

Building on this proposal, this article suggests a slightly different hybrid model grounded in party autonomy, allowing parties to choose between two routes when entering into an arbitration agreement.

Route 1- The existing framework. Parties proceed under the Arbitration Act, with the existing limited grounds of challenge under Section 34 and appeal under Section 37, preserving the current structure for parties content with minimal judicial intervention.

Route 2- The hybrid appellate framework. Parties agree at the outset that a full appellate remedy will be available after the award, covering questions of law and procedure as well as the merits and factual findings. To avoid an additional layer of prolonged litigation, the proposal provides for a single consolidated appeal: the existing Section 34 challenge and Section 37 appeal, effectively two stages of post-award proceedings, are merged into one appellate proceeding in which the aggrieved party raises procedural, statutory and substantive challenges together. This is not an additional appeal but a restructuring of the existing framework, available only to parties who voluntarily choose it.

Route 2 can also be adopted through a multi-tier arbitration clause under which a dissatisfied party may have the award reconsidered by a second tribunal, whose outcome is final and binding. This is not novel: in Centrotrade Minerals & Metal Inc. v. Hindustan Copper Ltd., the Supreme Court upheld a clause providing for arbitration before an Indian Council of Arbitration tribunal and, if either party remained dissatisfied, a second and binding arbitration before the ICC in London, as consistent with party autonomy. A multi-tier clause is thus a legally sound route to the enhanced review Route 2 contemplates, if drafted with precision.

This model balances the competing demands of arbitration. Meaningful appellate scrutiny can strengthen confidence in the accuracy and integrity of arbitral decisions, consolidation prevents broader review from becoming prolonged litigation, and optionality preserves party autonomy. The proposal does not dilute arbitration’s advantages but recalibrates the balance between finality, accuracy and efficiency.

This article makes two recommendations.

1. Legislative amendment to the Arbitration and Conciliation Act, 1996. Parliament should insert a new provision, for instance Section 34A, into Part I of the Act, enabling parties to elect the hybrid appellate framework at the time of contracting in place of Sections 34 and 37. It should specify the appellate forum, the grounds of appeal (law, procedure and factual merits), the limitation period, and that the appeal is a single, consolidated proceeding.

2. Withdrawal of executive instructions discouraging arbitration. The Ministry of Finance’s June 2024 Office Memorandum, the Public Works Department circular of 21 April 2025 and the MoRTH circular of 12 January 2026 should be reconsidered and withdrawn. As argued above, the answer to concerns about arbitral integrity is reform, including through the hybrid model, not retreat to litigation or administrative settlement.

The challenges facing domestic arbitration in India are real. Governmental retreat from arbitration reflects a deepening loss of institutional confidence, and inconsistent jurisprudence undermines the predictability essential to any credible dispute resolution system. But the response should be reform, not retreat: a framework offering meaningful appellate review where parties want it, without dismantling the efficiency and flexibility that make arbitration valuable. A hybrid appellate model, grounded in party autonomy and designed to consolidate rather than multiply post-award proceedings, offers one path towards that objective, alongside greater judicial discipline and consistency in the development of arbitration law.

About the authors: Shally Bhasin and Varun Pathak are Partners and Divyansh Morolia is an Associate at Shardul Amarchand Mangaldas & Co.

Disclaimer: The opinions expressed in this article are those of the author(s). The opinions presented do not necessarily reflect the views of Bar & Bench.

If you would like your Deals, Columns, Press Releases to be published on Bar & Bench, please fill in the form available here.