Arbitration in India: What It Is and Why It Matters

India has over 54 million pending cases across its courts — more than 70% of which have been pending for over a year.¹ A commercial dispute filed in a district court today may not reach final resolution for a decade. A High Court appeal can add years more. For a foreign company entering a joint venture, an infrastructure contract, or a shareholder agreement with an Indian party, this is not an abstract statistic — it is a contractual risk that affects pricing, risk allocation, and dispute strategy at the time of drafting itself.

This is where arbitration comes into the picture. But reducing it to an “alternative” or a “faster route” is conceptually inaccurate. In India, arbitration is not simply a faster version of litigation. It is a separate legal regime, governed by a specific statute, shaped by over two decades of Supreme Court judgments, and increasingly aligned with international practice.² Understanding how it works — before a dispute arises — is the starting point for anyone doing business with an Indian counterparty. Most mistakes in arbitration are not made during disputes. They are made at the stage of contract drafting.

What Arbitration Actually Is

Arbitration is a private adjudicatory process. Parties agree — usually in their contract — that if a dispute arises, it will be referred to one or more arbitrators whose decision is final and binding. The arbitrator is not a judge. The proceedings are not in a courtroom. The award is not a public document. The arbitral tribunal performs a quasi-judicial function, and its award carries the same enforceability as a court decree.³

Three features distinguish it from litigation.

Privacy — unlike court proceedings, arbitration is confidential. No public record, no reported judgment, no reputational exposure. Confidentiality in India, however, is not absolute and may be tested in enforcement or challenge proceedings.

Party autonomy — parties choose their arbitrator, the seat, the procedural rules, and the language. The arbitral tribunal is not bound by the Code of Civil Procedure or the Indian Evidence Act.⁴ This autonomy extends to appointing sector-specific experts — an engineer or a project finance specialist in an infrastructure dispute, for instance — which is rarely possible in crowded court rosters.

Finality — an arbitral award is binding and enforceable as a court decree under Section 36 of the Arbitration and Conciliation Act, 1996.⁵ Challenge is available only on narrow statutory grounds under Section 34 — courts cannot re-examine the merits or sit in appeal over the award.⁶ This limited judicial review is the single most important commercial advantage of arbitration.

The Statute That Governs Everything

The Arbitration and Conciliation Act, 1996 is the foundational legislation.⁷ It is based on the UNCITRAL Model Law on International Commercial Arbitration — the same framework that underpins arbitration law in Singapore, Hong Kong, and most major commercial jurisdictions.⁸ This alignment ensures that foreign parties can operate within a familiar legal architecture.

The Act has two distinct parts that serve entirely different purposes.

Part I governs arbitrations seated in India — both domestic disputes and international commercial arbitrations where the parties have chosen India as the seat. It covers the arbitration agreement (Section 7), appointment of arbitrators (Section 11), conduct of proceedings, the award (Section 31), and the limited grounds on which an award can be challenged (Section 34). It reflects a foundational principle of minimal judicial intervention, stated plainly in Section 5.⁹

Part II governs enforcement of foreign awards — awards made outside India in countries that are signatories to the New York Convention (Sections 44–52) or the Geneva Convention (Sections 53–60). A foreign award can be enforced in India under Section 48, subject only to limited grounds of refusal that mirror the Convention.¹⁰

The critical point — and one that took years of Supreme Court litigation to settle — is that Part I and Part II do not overlap. In Bharat Aluminium Co. v. Kaiser Aluminium Technical Services Inc. (2012) 9 SCC 552 — universally known as BALCO — a Constitution Bench held definitively: “Part I of the Arbitration Act, 1996 would have no application to International Commercial Arbitration held outside India.”¹¹ If the seat is abroad, Indian courts cannot entertain a Section 34 challenge or grant Section 9 interim relief for that arbitration. Foreign-seated awards are enforced in India only through Part II, at the recognition and enforcement stage.

BALCO was also critical because it corrected the earlier line of cases. Before 2012, Bhatia International v. Bulk Trading S.A. (2002) 4 SCC 105 had allowed Part I to apply to foreign arbitrations — meaning Indian courts could interfere with arbitrations seated in London or Singapore.¹² BALCO restored certainty and aligned Indian arbitration law with the globally accepted seat-centric doctrine. International parties can now structure their contracts knowing that choosing a foreign seat removes Indian curial supervision over the arbitral process entirely.

How It Compares: Arbitration vs Litigation vs Mediation

Against litigation, the scope of judicial review is the real distinction. A court can review a lower court’s decision on merits, law, and facts across multiple tiers of appeal. An arbitral award can only be challenged under Section 34 — on grounds of incapacity, invalid agreement, lack of notice, excess of jurisdiction, improper tribunal composition, patent illegality, or public policy.¹³ Courts cannot re-appreciate evidence. They cannot substitute their view of the merits. This insulation from appellate review is what gives arbitration its commercial finality. It is not a technicality — it is the entire commercial attraction.

There is also a practical distinction that matters for cross-border projects: litigation timetables are at the mercy of court dockets and judicial vacancies, whereas an arbitral tribunal can schedule hearings more intensively and use documents-only procedures where appropriate. For a project with financing tied to milestones, that procedural control is itself a risk-management tool.

One important development on finality is worth noting. In Gayatri Balasamy v. M/S. ISG Novasoft Technologies Limited (2025 INSC 605), a five-judge Constitution Bench held by a 4:1 majority that courts possess a limited power to modify arbitral awards under Section 34 — in specific circumstances: severing invalid portions, correcting clerical or computational errors, and adjusting post-award interest.¹⁴ The dissent by Justice Viswanathan, which held that no such power exists under the statute, reflects a view widely held by international practitioners. The judgment is binding law but its long-term impact on the finality principle is still being assessed.

Against mediation, the distinction is one of power. Mediation is facilitative — the mediator helps parties reach a settlement but has no power to impose one. If mediation fails, parties walk away. An arbitral award is binding whether or not both sides are happy with it. Under Section 30, parties can attempt mediation during arbitral proceedings and have any settlement recorded as an award, at which point it acquires the enforceability of a decree.¹⁵ The two processes are not mutually exclusive. Sophisticated parties often combine them — commencing arbitration to protect limitation and using mediation windows to explore settlement under the shadow of a binding process.

Why International Parties Choose Indian-Seated Arbitration

India is a signatory to the New York Convention, ratified in 1960. Awards made in India are enforceable in 170+ countries that are party to the Convention.¹⁶ This gives Indian-seated arbitration the same global enforcement reach as awards from Singapore or London. Enforceability, not speed, is often the decisive factor in cross-border contracts.

The 2015 and 2019 amendments addressed the two biggest criticisms of Indian arbitration — delay and judicial overreach. The 2015 amendment introduced Section 29A: a 12-month time limit for the tribunal to make an award (from completion of pleadings, as clarified by the 2019 amendment), extendable to 18 months by party consent.¹⁷ It also narrowed the public policy ground under Section 34 — reversing the expansive readings in ONGC v. Saw Pipes (2003) 5 SCC 705 and ONGC v. Western Geco International Ltd. (2014) 9 SCC 263 that had allowed courts to interfere too readily.¹⁸ This marked a clear shift from an interventionist to a restrained judicial approach.

The 2019 amendment pushed further: greater emphasis on institutional arbitration, the proposed Arbitration Council of India, and a pleadings-within-six-months rule. It also clarified that the strict Section 29A timeline applies to domestic arbitrations; international commercial arbitrations seated in India are to be conducted “as expeditiously as possible.”¹⁹

Institutional arbitration is growing. The Mumbai Centre for International Arbitration (MCIA) operates under internationally benchmarked rules.²⁰ The Delhi International Arbitration Centre (DIAC) is High Court-annexed with dedicated infrastructure.²¹ Neither competes with SIAC or LCIA yet — but for contracts with Indian governing law or Indian parties, Indian-seated arbitration is increasingly the sensible default rather than a compromise.

India is not Singapore, at least not yet. Enforcement can still be slow. Section 34 challenges are routinely used as delay tactics. Costs escalate in ad hoc arbitrations without institutional discipline. But the trajectory since 2015 has been consistently toward a more restrained, pro-arbitration judicial approach — Ssangyong Engineering & Construction Co. Ltd. v. NHAI (2019) 15 SCC 131 being the clearest post-amendment correction on the scope of public policy and the limited role of courts.²²

The Arbitration Agreement: Where Everything Begins

Everything in arbitration flows from one clause. Section 7 of the Act defines an arbitration agreement as a written agreement to refer present or future disputes to arbitration.²³ It can be a clause in a contract or a standalone submission agreement. Writing includes signed documents, exchanges of email, or any communication that records the agreement.

A vague clause — “disputes shall be referred to arbitration” — is technically valid. But it is a trap. Courts have had to fill gaps on the number of arbitrators, the seat, the procedural rules, and the governing law. Every gap is a potential jurisdictional fight before the merits are even reached. Poor drafting shifts the dispute from substance to procedure.

A well-drafted clause specifies the seat, the number of arbitrators, the institutional rules if any, the governing law of the agreement, and the language. These are not formalities. They determine which court supervises the arbitration, which procedural law applies, where the award can be challenged, and whether the arbitration qualifies as an international commercial arbitration under Section 2(1)(f) — which affects Section 29A timelines and the standard of review.²⁴

A clause that says “disputes to be settled by arbitration in India” is better than nothing. A clause that says “disputes shall be finally resolved by arbitration under the MCIA Rules, with the seat at Mumbai, before a sole arbitrator, governed by Indian law” gives everyone — the parties, the arbitrator, and any court asked to intervene — a clear framework from the start. Clarity at the drafting stage eliminates uncertainty at the enforcement stage.

Where Things Stand

Indian arbitration law has changed substantially since 1996 — most of it in the right direction. The regime is now largely aligned with international practice: seat-centric, court-restrained, and globally enforceable. The shift is structural, not cosmetic. Gayatri Balasamy has introduced a new variable on the finality question, and its downstream effects will take time to settle.

The arbitration agreement is where the framework begins — and for any party entering a contract with an Indian counterparty, getting that clause right is the most consequential legal decision before a dispute ever arises.

The author is a law student at Law Centre-1, Faculty of Law, University of Delhi. Views are personal.