I. Introduction
When the parties choose arbitration as their dispute resolution mechanism, they make an informed decision which reflects their confidence that arbitration proceedings will lead to a fast, cost effective and most importantly a binding award which is final. According to international conventions, such as the UNCITRAL Model Law[1] and the New York Convention[2], the fundamental premise of commercial arbitration is finality and enforceability.
It appears that the Arbitration and Conciliation Act of 1996 (“Act”) had realized this when the Act was formulated. It provides under Section 5 that the non-obstante provision takes precedence over all other Acts and expressly prohibits any judicial interference in matters except where explicitly provided for in the Act.[3] Furthermore, Section 35 of the Act declares that any arbitration award is final and binding on the parties.[4] Therefore, Sections 5 and 35 clearly imply that judicial intervention can only occur if explicitly provided in the Act.
Nevertheless, the courts have certainly not remained away from this. The Act provides for judicial intervention in the arbitration process in a systematic and limited manner. Before arbitration commences, courts can make a referral to arbitration by virtue of section 8 of the Act, while arbitrators can be appointed under section 11 of the Act. Once the arbitration process is underway, the courts can provide interim relief through section 9 of the Act and help in the collection of evidence as provided in section 27 of the Act. Following the conclusion of arbitration, important sections such as section 34 (setting aside of an award), section 36 (enforcement of an award), and section 37 (appeals) will be applicable. On paper this makes for a very structured and rational approach, but in practice it has turned out to be a battering ram aimed directly at the finality guaranteed by Section 35.[5]
This article analyses how the Indian courts have always stretched beyond the limits of their authority, and in doing so, they have made the concept of finality a mere fiction. Section II advances the arguments for the finality of the arbitration, advancing the doctrinal arguments of party autonomy, its root in commercial certainty, and institutional integrity, and tracing the use of Sections 34, 36 and 37 to chip away at each of these. In Section III, it discusses the proposed Section 34A and the Appellate Arbitral Tribunal, contending that such a change is not “corrective” but rather another source of delay, cloaked in the guise of judicial restraint. Section IV outlines a range of practical reforms, for stays, extensions and a presumption in favour of enforcement, that could help revive the promise of Section 35. Lastly, Section V concludes.
II. The Case for Arbitral Finality and against Excessive Judicial Intervention
The Commercial Promise of Arbitration
Finality of arbitral awards does not exist on paper, rather, it constitutes the fundamental rationale behind parties’ resorting to arbitration. By including an arbitration agreement in their contract, the commercial parties knowingly and wisely make an informed decision. They wish to settle their dispute through arbitration, which should take place in accordance with their agreement, under the jurisdiction of arbitrators appointed by themselves, and not subject to review in any court to which they chose to exclude at the first place.
The Three Pillars of Finality
The rationale for upholding finality can be stated in terms of three equally important foundations. The first foundation consists of party autonomy. The arbitration process is a consensual undertaking, where parties voluntarily decide to arbitrate their dispute, pick an arbitrator, and submit to the result. When the court intervenes in this process by carrying out the merits-based review of the award, it defeats all elements of the parties’ autonomy. What the parties decide in their agreement is rendered meaningless and a judge who was knowingly excluded gets the final word.
The second is commercial certainty. Commercial activity thrives on calculated risk-taking. When parties come to terms, they calculate their losses and costs, including those involved in resolving any disputes that may arise. The arbitration process, as conceived by its proponents, represents such a calculation, being a clear process leading to a predictable outcome. In India, however, arbitration has evolved into a predetermined beginning stage but an unpredictable contest thereafter. An arbitral award subject to challenge, stay, and appeal under Sections 34, 36(2), and 37 respectively is not a solution, it only opens the next battlefield.
The third element is integrity. An arbitration can be effective only where the parties are sure that the award of the arbitrators will count for something. Whenever a court considers a Section 34 application an opportunity to reassess all aspects of the case including evidence and the reasons behind the decision of the arbitrator and decides that there were errors in the “right judgment,” the message sent out is clear that an award of arbitration carries no conviction. If one is defeated in the arbitration, he has a way out in appealing on some technical grounds. An arbitration award is nothing but permission to continue litigation. Such an approach does great damage not only to the award itself but to arbitration as a system. The New York Convention and the Model Law make the exact opposite assumption, and the countries following it, most importantly Singapore and England, have a friendly arbitral environment.
The Real Cost: Who Actually Suffers
Institutional injustice has to be felt from the perspective of the one who suffers it. The party which manages to win in an arbitration award only to go through several years of court proceedings, winning nothing. This is because it entered into arbitration to avoid litigation. It finds itself embroiled in court proceedings all the same and spends money on lawyers, while the other party uses every procedure possible to stall enforcement. The very point of Section 35[6] becomes nullified. For foreign litigants and companies operating internationally, such circumstances give a good enough reason to prefer Singapore or London as arbitral seats over Mumbai or New Delhi. The fact remains that no matter how good the drafting of India’s arbitration statute can be, if courts insist on re-examining awards, India is going to lose out in the international arena. In Pravin Electricals[7], the court reiterated that at the pre-referral stage, the court’s function is confined to prima facie examination, and it cannot conduct a mini trial nor undertake a full factual evaluation of the dispute. The court’s role is limited to determining whether a valid arbitration agreement exists and it does not extend to pronouncing upon the merits. Post-award, this limited examination must also hold true.
How Section 34 Was Broadly Interpreted
It was possible due to the misuse of the “public policy” ground under Section 34(2)(b)(ii)[8], which itself was not properly defined at all. In Saw Pipes[9], the Supreme Court of India took full advantage of this imprecise wording to introduce “patent illegality” into the public policy ground, and thus the scope of examination extended to all awards deemed contrary to the Indian legal system. Thus, what was meant to be a narrow judicial review ground expanded into an extensive appeal ground. With the 2015 Amendment and the further interpretation in Ssangyong[10], this mischief has been addressed, but the tarnished reputation of India in international arbitration will forever remain there because of the undefined public policy ground.
Sections 36 and 37: Compounding the Problem
However, Section 34[11] is not an independent rule in itself. Under Section 36(1), once the period for making the application to set aside has elapsed, or if such an application was made but turned out to be unsuccessful, then the arbitration award may be enforced as a decree of the court.[12] However, Section 36(2) makes this seem not so final since the court is given discretion to stay the enforcement of the award until Section 34 proceedings conclude.[13] This is because the Section 34 application is never really made, it’s just used as leverage to get the stay.
This is compounded by Section 37. According to Section 37 (1)(c)[14], the appeal in Section 34 cases would be limited to the orders of the court and not an appeal about the arbitral award itself. However, appeals are conducted by the appellate courts, which usually delve into the question of contractual interpretation and facts again, thus making the whole process extend by another two to three years. This was evident in the case of Somdatt Builders[15] where the Supreme Court was forced to step in because a Division Bench of the High Court reinterpreted the contract despite what Section 37 allows for. Significantly, the court took over thirteen years to finally decide on this case, making it a cautionary tale not merely of judicial overreach but of the compounding delays that such overreach inflicts upon commercial parties.
The Concession and Why It Does Not Defeat the Argument
Intellectual integrity compels us to recognize the viewpoint of the courts. There are instances when arbitral tribunals make perverse award, fraud does happen, and there are also instances like that of Shin-Etsu Chemical[16], where it is essential for the court to intervene because of serious doubts regarding the validity of the arbitration agreement itself. An unremedied system is untenable and might become a shield for illegal business transactions by taking cover under the guise of arbitration. However, this does not detract from the case for finality. The trouble did not lie in the existence of Section 34, but in the culture surrounding its invocation, a judicial culture where courts consider the narrow provision as an avenue for appeal. The solution to a bad judicial culture is not an unlimited judicial access but judicial discipline.
III. Section 34A: Repacking Juridical Intervention
The proposal to amend the Arbitration and Conciliation Act, 1996 by including Section 34A[17] which provides for the establishment of Appellate Arbitral Tribunal (“AAT”) to deal with appeals against arbitral awards is essentially an optical illusion at best and self-contradictory at worst. While ostensibly providing protection from judicial interference, the amendment ensures increased involvement of judiciary in the matter. Section 35 of the Act mandates that an arbitral award shall be final and binding. However, as is widely known, the concept of finality cannot exist on paper alone, rather, it must be seen in the context of the exhaustion of challenge mechanism. Every layer of appellate procedure increases the distance between the parties and the concept of finality. In the existing scenario, each challenge mechanism starts from Section 34 and goes through Section 37 and ultimately reaches the Supreme Court under Article 136. The proposed amendment does not eliminate these steps but rather adds one more layer to them. The party who loses his case gets another set of hearings without stepping into any judicial forum, whereas the party who wins his case has to go through yet another round of arbitration.
The commercial consequences of all this are even more chilling. In any commercial arrangement, an unimplemented award is a deadweight liability. The successful party has no way to record the award, finance based on it, settle accounts, or exit an unwanted situation. Section 34A ensures that effect. In addition, if the AAT does have the power to impose interim stays on the award’s enforcement, it recreates the problem that the legislation sought to fix prior to 2015, where merely commencing an application for setting aside meant automatically locking up an award forever in the process. And if not, the parties can easily just sidestep the body and go straight to court. Not only that, but there is also the issue of sheer economics, an appeal process involving an entirely new panel means another set of fees, legal costs, and memoranda. Therefore, challenging an award may come at a financial cost equal to the value of the whole dispute, effectively rewarding the well-funded party while punishing the winner.
At root, the issue with India’s arbitration regime is not who reviews an award, it is what is reviewed under Section 34. Public policy grounds, patent illegality, and merit grounds have been stretched beyond recognition to accommodate judicial interference with an arbitration award. The fix here is not to add reviewers but to narrow grounds for review. By filling the new office with experienced judges and counsels, the AAT only moves the same judicial mindset to a new space. The NY Convention regime governing enforcement of awards made in India internationally would have no idea what to make of this new system of arbitration award in India. An award already examined and revised by two different arbitral institutions could confuse foreign courts trying to execute it under the NY Convention regime, thereby rendering India less suitable to host international commercial arbitrations.
IV. Practical Reforms for Effective Arbitration
There is no doubt about the diagnosis of India’s arbitration system, for which blame cannot be attributed to lack of substantive arbitration laws per se but rather to the system’s surrounding architecture, where enforcing and executing an award becomes economically sensible for any defeated litigant. These recommendations are commercial and are aimed at correcting certain identified shortcomings.
First, Sections 34 and 36(2) should become a calculated risk, not a procedural shelter to delay the enforcement of award. Under the present framework, a party unsuccessful in arbitration can invoke Section 34 and seek stay under Section 36(2) to delay the enforcement of award without bearing any meaningful financial consequence for the delay. It is worth noting that courts have, in practice, already developed a convention of ordering the deposit of award amounts during section 34 proceedings. However, this practice does little to deter frivolous challenges at the threshold. More importantly, once the awarded amount is deposited and the award-holder is permitted to withdraw it, typically upon furnishing security, the award-holder’s entitlement to post-award interest ordinarily ceases from the date of deposit. Consequently, even where the Section 34 challenge ultimately fails, the award-debtor enjoys the benefit of having postponed final compliance without compensating the award-holder for the period during which enforcement remained stalled. The existing framework therefore shifts the economic burden of delay onto the successful party, thereby weakening the compensatory purpose of post-award interest and diminishing the deterrent against meritless challenges. This is not merely a procedural technicality; it creates a structural incentive for speculative and frivolous challenges. The courts should, in appropriate cases, make full use of their powers under Section 31A to award costs against parties pursuing frivolous or meritless challenges under Section 34. Such costs should extend to all reasonable expenses incurred by the successful award-holder, including legal and counsel’s fees, the costs of furnishing security to withdraw the deposited award amount, any other expense reasonably incurred on defending the challenge and the interest on such costs. To ensure that this power is exercised meaningfully, the Act should further be amended to require courts, where they decline to award such costs, to record specific reasons for doing so.
Second, Section 29A[18] requires a fundamental reconsideration and the reform must run in the opposite direction. The present framework, which requires parties to seek judicial intervention for extending the mandate of the arbitral tribunal, is difficult to reconcile with the foundational principle of minimal judicial intervention. Accordingly, the judicial role in granting extensions should be removed entirely. Instead, the parties should be at freedom to extend the mandate of the arbitrator by mutual agreement, just as they are free to agree on the procedural timeline in the first place. This would also eliminate the perception that the Act prescribes a universal standard for all arbitrations (6+12 months), irrespective of their complexity. A high-value, multi-party infrastructure dispute will often need longer period. The determination of an appropriate timetable is therefore best left to the parties and the arbitral tribunal, who are best placed to assess the procedural needs of the dispute before them, rather than to the courts through routine extension proceedings.
Moreover, the proposed reform is consistent with the broader scheme of the Act. Where the delay is attributable to the tribunal, the Act already supplies the answer in Sections 14 and 15, empowering the parties to terminate the mandate of an arbitrator who has failed to act without undue delay and to appoint a substitute arbitrator. In light of this, the requirement under Section 29A for parties to routinely approach the court for an extension of the tribunal’s mandate serves little independent purpose.
What is still missing is a real check on parties who stall on purpose. Although, Section 29A(8) empowers the court to impose actual or exemplary costs upon the party, but this power is seldom exercised in practice. In most extension proceedings, the parties are ad idem on the need for an extension, and the court rarely undertakes a substantive inquiry into whether the delay is attributable to the conduct of either party. Consequently, the provision has had little deterrent effect against tactical or dilatory behaviour.
A more effectively approach would be to expressly empower the tribunal itself should be to impose costs, and exemplary costs where the conduct warrants it, on any party it finds guilty of dilatory tactics. The tribunal, being intimately acquainted with the conduct of the arbitration and the parties’ procedural behaviour, is far better placed than the court to identify and sanction such conduct. This would ensure that the discipline of time remains within the arbitral process itself, rather than being indirectly entrusted to the courts through routine extension proceedings.
V. Conclusion
The root cause of the arbitration crisis in India lies in the lack of institutional will. Far from lacking any definition of finality, the Arbitration and Conciliation Act, 1996 is explicit in its treatment of the subject through Sections 5 and 35. Rather, the problem lies in failure to give effect to the legislative intent. The misconstruction of Section 34 as a mechanism for merits-based review and a tactic to delay the enforcement of the award by seeking stay under Section 36(2), and Section 37 which allows the courts to rehear appeals in effect ensure that the arbitral process becomes only the initial skirmish in what will eventually become years-long litigation in court. The introduction of the new provision of Section 34A will do nothing to remedy this situation; it will merely shift it around. Reform requires limiting grounds for review, making party bear the financial consequence of meritless or frivolous challenge, and fostering an attitude amongst the judges that the norm should be enforcement and not review. Unless India develops this culture, it will continue to send its disputes abroad.
[1] UNCITRAL Model Law on International Commercial Arbitration (UN Doc A/40/17, 1985, as amended 2006).
[2] Convention on the Recognition and Enforcement of Foreign Arbitral Awards (New York, 10 June 1958) 330 UNTS 3 (New York Convention), art V.
[3] Arbitration and Conciliation Act 1996 (India), s 5.
[4] ibid s 35.
[5] ibid ss 8, 9, 11, 14, 27, 34, 36, 37.
[6] Arbitration and Conciliation Act 1996 (India), s 35.
[7] Pravin Electricals Pvt Ltd v. Galaxy Infra and Engineering Pvt Ltd., 2021 SCC Online SC 190.
[8] Arbitration and Conciliation Act 1996 (India), s 34(2)(b)(ii).
[9] Oil and Natural Gas Corporation Ltd v Saw Pipes Ltd (2003) 5 SCC 705.
[10] Ssangyong Engineering and Construction Co Ltd v National Highways Authority of India (2019) 15 SCC 131.
[11] Arbitration and Conciliation Act 1996 (India), s 34.
[12] Arbitration and Conciliation Act 1996 (India), s 36(1).
[13] Arbitration and Conciliation Act 1996 (India), s 36(2).
[14] Arbitration and Conciliation Act 1996 (India), s 37(1)(c).
[15] Somdatt Builders NCC–NEC (JV) v National Highways Authority of India, CIVIL APPEAL NO. 2058 OF 2012.
[16] Shin-Etsu Chemical Co Ltd v Aksh Optifibre Ltd (2005) 7 SCC 234.
[17] Draft Arbitration and Conciliation (Amendment) Bill 2024 (India), cl 34A.
[18] Arbitration and Conciliation Act 1996 (India), s 29A.
Author(s)
Vrinda Bagaria
Student at JGLS