The advantages and disadvantages of arbitration decide whether a business should keep its disputes out of court or not. Arbitration in India, governed by the Arbitration and Conciliation Act, 1996, offers party autonomy, confidentiality, a subject-expert decision-maker, a final award with no appeal on merits, and cross-border enforceability under the New York Convention. Its disadvantages are just as real: high and rising costs (arbitrator fees have run into crores in some ad hoc references), delays that defeat the promised speed, repeated court intervention at the appointment, challenge and enforcement stages, and no binding precedent. This article weighs both sides against current Indian law, real cost figures, and the leading Supreme Court rulings.
This article sets out the advantages and disadvantages of arbitration in India, the real cost and delay data, the map of where a dispute still goes back to court, and a framework for when arbitration is the right choice.
Arbitration was meant to be the quicker, cheaper, private cousin of litigation. For a long time that pitch went largely unquestioned. Then, in 2025, the Rajasthan High Court looked at a single arbitration that had run for seven years, in which a tribunal of one former Supreme Court judge and two former High Court judges had been paid close to Rs 13 crore in fees, and called it “luxury litigation.” That phrase has since travelled through boardrooms and law offices, because it captures a real tension at the heart of the subject.
The tension is this. On paper, arbitration in India carries genuine structural advantages, most of them written into the Arbitration and Conciliation Act, 1996, which is modelled on the UNCITRAL Model Law. In practice, the way arbitration actually runs, especially ad hoc arbitration between Indian parties, has drifted a long way from that promise. The law has been amended in 2015, 2019 and 2021 to close the gap, and a fresh amendment bill is pending, yet the trade-offs remain live.
So the honest answer to “should I arbitrate” is: it depends on the dispute, the amount at stake, the seat, and whether you can steer the process into an institution. The sections below give you the material to make that call.
Advantages and disadvantages of arbitration at a glance
The advantages and disadvantages of arbitration tend to be two sides of the same feature. The privacy that protects a company’s reputation is the same privacy that keeps an unfair award out of public scrutiny. The finality that gives certainty is the same finality that leaves a wrong award almost impossible to correct. Reading the two columns together, rather than as a simple scorecard, is the only way to use them.
Here is the short version before the detail.
| Advantage | Matching disadvantage |
|---|---|
| Party autonomy: you choose the arbitrator, seat, procedure and language | That autonomy is only as good as the clause you drafted; a bad clause locks you into a bad process |
| A subject-expert decides your dispute, not a generalist judge | No appeal on the merits, so an expert who gets it wrong cannot be corrected |
| Confidential proceedings (Section 42A) | No public record, no precedent, and reduced scrutiny of the tribunal |
| Final and binding award, limited challenge | Section 34 challenges and Section 37 appeals still pull parties back into court for years |
| Global enforceability under the New York Convention | Domestic and foreign enforcement is itself a fresh court process with its own limitation clock |
| Flexible, lighter procedure; hearings on your schedule | Per-hearing fees plus routine extensions create an incentive to prolong |
| Minimal court intervention by design (Section 5) | The Act itself opens court doors at Sections 9, 11, 34 and 36 |
| Often faster than India’s backlogged courts | Ad hoc arbitration can be as slow, and more expensive, than the litigation it replaced |
A common question is whether these trade-offs apply equally to every dispute. They do not. A cross-border supply contract and a Rs 40 lakh domestic payment dispute sit at opposite ends: the first leans heavily towards arbitration, the second often does not. Keep that spread in mind as each point is unpacked below.
How arbitration works under the 1996 Act
Arbitration under the 1996 Act is a private adjudication that the parties agree to in advance, usually through an arbitration clause in their contract. The dispute is decided by one or more arbitrators the parties help choose, the tribunal issues a binding award, and the court’s role is deliberately kept to a minimum. If you want the full procedure, from notice of arbitration to enforcement, our explainer on how arbitration works under the Arbitration and Conciliation Act, 1996 walks through each stage.
The foundation is the arbitration agreement. Under Section 7 of the Arbitration and Conciliation Act, 1996, the agreement must be in writing, and it is what gives the tribunal its authority. Without a valid arbitration agreement, there is no arbitration, which is why a court under Section 8 will refer parties to arbitration only when such an agreement exists. This consent-based design is the source of both the biggest advantages and some of the sharpest limits discussed later.
The arbitration agreement and reference
Once a dispute arises and one side invokes the clause, a tribunal is constituted and the reference begins. The tribunal is not bound by the Code of Civil Procedure or the strict rules of the Evidence Act, which is what makes the procedure lighter than a court trial. And the guiding principle sits in Section 5: no judicial authority is to intervene except where the Act itself allows it. In theory, that keeps the process self-contained.
In practice, readers often ask whether an arbitration clause is mandatory once it is in the contract. Broadly, yes: if there is a valid clause and one party still sues in court, the other can ask the court under Section 8 to send the matter to arbitration. A quick point of disambiguation, since it trips up newcomers: “arbitration” is dispute resolution, while “arbitrage” is a finance term for exploiting price differences. They are unrelated despite the similar spelling.
From the 1940 Act to the 1996 Act and its amendments
The current statute did not appear out of nowhere. India ran on the Arbitration Act, 1940 for over half a century, a regime widely criticised for excessive court supervision. The 1996 Act replaced it, importing the UNCITRAL Model Law to make Indian arbitration friendlier to international commerce. Three later amendments then tried to fix what practice exposed: the 2015 Amendment added a strict timeline and a fee schedule and curbed expansive judicial review; the 2019 Amendment created the Arbitration Council of India and a statutory confidentiality duty; and the 2021 Amendment allowed an unconditional stay of enforcement where an award is tainted by fraud. Each of these maps directly onto an advantage or a disadvantage below.
Advantages of arbitration
The advantages of arbitration are, at their core, control and enforceability: control over who decides, how, and in private, plus an award that travels across borders more easily than any court decree. These are not marketing claims; most of them are anchored in specific sections of the 1996 Act and in Supreme Court authority. Take them one at a time.
Party autonomy
Party autonomy is the defining advantage. The parties choose the number of arbitrators (Section 10), participate in appointing them (Section 11), and fix the seat or place of arbitration (Section 20), the language, and much of the procedure. A construction dispute can go before an engineer-arbitrator; a shipping dispute before a maritime specialist. This freedom extends surprisingly far. In PASL Wind Solutions Pvt. Ltd. v. GE Power Conversion India Pvt. Ltd., (2021) 7 SCC 1, the Supreme Court held that even two Indian parties may choose a foreign seat of arbitration, and the resulting award is enforceable in India as a foreign award. That is a degree of self-determination no domestic court process offers.
A subject-expert decision-maker
In court, you take the judge assigned to the roster, however unfamiliar the subject. In arbitration, you can pick a decision-maker who already understands the technical terrain, whether that is a power-purchase tariff, a software licensing model, or a re-insurance treaty. For complex commercial disputes, that expertise can mean a sharper, faster grasp of the real issue. In practice, this is one of the strongest reasons infrastructure and technology contracts opt for arbitration: the parties would rather explain their dispute once, to someone who gets it, than educate a generalist bench from scratch.
Confidentiality
Arbitration is private, and since 2019 that privacy is a statutory duty. Section 42A, inserted by the 2019 Amendment, obliges the arbitrator, the institution and the parties to keep the proceedings confidential, except where disclosure is necessary to enforce or challenge the award. For a company defending a product-quality claim or a sensitive joint-venture dispute, keeping the matter out of the public gaze, and out of the press, is a real commercial advantage. But hold this thought, because the same confidentiality is listed later as a disadvantage: what is private is also unscrutinised.
Finality and limited appeal
An arbitral award is final and binding under Section 35. There is no full appeal on the merits the way there is from a trial court to a High Court. The only route is a narrow challenge to set the award aside, and the Supreme Court has steadily disciplined even that. In Ssangyong Engineering and Construction Co. Ltd. v. National Highways Authority of India, (2019) 15 SCC 131, the Court confirmed that after the 2015 Amendment, courts can no longer reopen the merits of an award under the guise of “public policy.” For a winning party, that finality is certainty: the fight ends, mostly, when the award is signed.
Cross-border enforceability
This is the single biggest structural advantage, and the one litigation simply cannot match. India is a party to the New York Convention of 1958, which more than 170 countries have signed (confirm the current count on the official contracting states list). A foreign award from a notified Convention country is enforced in India under Sections 44 to 49, and once the court is satisfied it is enforceable, the award is treated as a decree. A court decree from, say, a Delhi court has no such automatic recognition in Singapore or London. For any contract with an international counterparty, this reach is often the deciding factor.
Neutrality, flexibility, and minimal court intervention
For cross-border deals, arbitration offers neutral ground: neither party has to litigate in the other’s home courts, before the other’s judges, in the other’s language. The procedure is flexible (Section 19 frees the tribunal from the CPC and the Evidence Act), so hearings can be scheduled around the parties, including by video. And Section 5 restricts judicial interference to what the Act permits, a design intent that the 2015 and 2019 amendments reinforced. The Constitution Bench in Bharat Aluminium Co. v. Kaiser Aluminium Technical Services Inc., (2012) 9 SCC 552 cemented this by holding that Indian courts cannot interfere with a foreign-seated arbitration, settling the “seat” theory that underpins minimal intervention. The independence of the arbitrator is protected too. In TRF Ltd. v. Energo Engineering Projects Ltd., (2017) 8 SCC 377 and later in Perkins Eastman Architects DPC v. HSCC (India) Ltd., 2019 SCC OnLine SC 1517, the Supreme Court held that a person interested in the dispute cannot appoint the sole arbitrator, a safeguard against a stacked tribunal. There is even a technical advantage at the doorway: in In Re: Interplay Between Arbitration Agreements under the Arbitration and Conciliation Act 1996 and the Indian Stamp Act 1899, (2024) 6 SCC 1, a seven-judge bench held that an unstamped arbitration agreement is still enforceable, removing a technical roadblock that used to stall references at the appointment stage.
Disadvantages of arbitration
The disadvantages of arbitration are cost, delay, court entanglement, and the flip side of finality: a wrong award you cannot really fix. What makes these worth taking seriously is that the strongest evidence for them is recent and Indian, not borrowed from foreign textbooks. Lead with the money.
Cost: the “luxury litigation” problem
The claim that arbitration is cheaper has taken a beating. In the seven-year Rajasthan matter that drew the “luxury litigation” label, a three-member tribunal of retired senior judges was paid close to Rs 13 crore, with per-sitting fees reported at around Rs 2.5 lakh per arbitrator across more than 160 sittings, as reporting on the order set out. That is not an isolated headline; it points to a structural problem in ad hoc arbitration that the cost section below breaks down in numbers. For a small or mid-sized business, the arbitrator’s fees can swallow a meaningful slice of the claim itself.
Delay in practice
The law promises speed. Section 29A requires the award within twelve months of the completion of pleadings, extendable by six months with party consent. Reality has not kept pace. A study of 202 extension orders in the Delhi High Court between 2015 and 2024 found extensions granted in about 98 percent of cases. When arbitrators are paid per hearing, and extensions are almost automatic, the “tareekh pe tareekh” culture that plagues the courts can reappear inside the arbitration itself. Isn’t a twelve-month rule that is waived nineteen times out of twenty really just a suggestion? That is the criticism, and the data gives it weight.
Narrow challenge, no merits appeal
Finality cuts both ways. If the arbitrator misreads the contract or gets the law wrong, you generally cannot appeal that error. Your only recourse is a challenge under Section 34 to set the award aside, on limited grounds, filed within three months (plus a possible 30 days). The history here is instructive. In Oil and Natural Gas Corporation Ltd. v. Saw Pipes Ltd., (2003) 5 SCC 705, the Court introduced “patent illegality” as a ground, and in Oil and Natural Gas Corporation Ltd. v. Western Geco International Ltd., (2014) 9 SCC 263 it widened public-policy review further, so much so that Section 34 came to be criticised as a “disguised appeal” that dragged awards back into years of litigation. The 2015 Amendment and Associate Builders v. Delhi Development Authority, (2015) 3 SCC 49 reined this in, confirming that a court cannot reappreciate evidence or sit in appeal. You can read the current position in our guide to the grounds for setting aside an arbitral award under Section 34. Even so, a narrow challenge is not the same as a fair-second-look, and that is a genuine cost of choosing arbitration.
No doctrine of precedent
Court judgments build the law; arbitral awards do not. An award binds the parties to that dispute and no one else, and because most awards are confidential, they are not published and cannot guide future tribunals. Two arbitrations on near-identical facts can reach opposite results, with no way to reconcile them. For a business that wants a clear rule it can rely on across many contracts, that absence of precedent is a real drawback.
The limits of consent: joinder and arbitrability
Arbitration rests on agreement, which means it struggles the moment a dispute reaches beyond the people who signed. Joining a non-signatory or consolidating related disputes is hard. The Supreme Court softened this in Cox and Kings Ltd. v. SAP India Pvt. Ltd., (2024) 4 SCC 1, recognising the “group of companies” doctrine so that a non-signatory can sometimes be bound where there was a mutual intention to bind it. But the test is still consent, so a genuinely unrelated third party stays outside. There is also a hard boundary on what can be arbitrated at all. In Vidya Drolia v. Durga Trading Corporation, (2021) 2 SCC 1, the Court set a fourfold test for non-arbitrability, keeping matters like serious fraud, criminal offences, insolvency and matrimonial disputes out of arbitration entirely. If your dispute or your opponent falls into these gaps, arbitration cannot give you a complete remedy.
Limited discovery and unequal bargaining power
Arbitration offers nothing like the full document discovery of a civil suit, which can hurt a party that needs records held by the other side to prove its case. And because arbitration clauses are often buried in standard-form contracts, the weaker party (an employee, a small vendor, a consumer) may be bound to arbitrate without ever having negotiated it. A common concern raised in employment and consumer contexts is exactly this: a mandatory clause can quietly close the courtroom door to the side least able to afford a private tribunal. It is a fairness problem that the pro-arbitration framing tends to skip over.
Cost of arbitration in India
The cost of arbitration in India has several layers, and arbitrator fees are only one of them. Understanding what you actually pay, and why ad hoc arbitration inflates it, is essential before you sign a clause that commits you to this route. The numbers below are drawn from reported orders and practitioner analyses, not estimates.
| Cost head | What it covers | Typical range or example |
|---|---|---|
| Arbitrator fees (Fourth Schedule model) | The tribunal’s fee, often per sitting in ad hoc references | Fourth Schedule caps the fee at Rs 30 lakh per arbitrator; senior ad hoc arbitrators can exceed this |
| Per-sitting fees | Charged each hearing in ad hoc arbitration | Around Rs 2.5 lakh per arbitrator per sitting reported in the “luxury litigation” order |
| Legal fees | Counsel and solicitor costs, usually the largest head | Can be roughly three-quarters of a party’s total spend |
| Institutional / administrative fees | Charged by an arbitral institution if you use one | Scaled to claim value; often lower overall than open-ended ad hoc fees |
| Venue, transcription, experts | Hearing rooms, records, expert witnesses | Variable; higher in long ad hoc references |
| An illustrative total | Pursuing a mid-sized claim end to end | A Rs 1 crore claim can cost Rs 30 to 40 lakh to arbitrate |
What you actually pay
Notice that the arbitrator is not usually the biggest line item; counsel is. Practitioner analyses suggest legal fees can account for around 74 percent of a party’s arbitration cost, with arbitrator fees and administrative charges making up the rest. That matters because it means switching to a cheaper arbitrator does not, by itself, make arbitration affordable. The whole process has to be disciplined.
Why retired-judge ad hoc arbitration got so expensive
The premium names in Indian arbitration are retired Supreme Court and High Court judges, and demand for them far outstrips supply. In an ad hoc reference, with no institution fixing the fee, those arbitrators effectively set their own terms, frequently per hearing. In Central Organisation for Railway Electrification v. ECI SPIC SMO MCML (JV), 2024 INSC 857, a five-judge bench separately curbed the practice of one party curating the arbitrator panel, holding that unilateral appointment mechanisms in public-sector contracts violate the equality guarantee. The per-sitting model then does the rest: it links the arbitrator’s earnings to the number of hearings, and the second-order effect is predictable. When more sittings mean more fees, and Section 29A extensions are granted almost as a matter of course, the process has a built-in incentive to run long rather than to conclude.
Is arbitration viable for small and medium disputes?
For a large, complex, high-value commercial dispute, spending 30 to 40 lakh to resolve it privately and finally can be worth it. For a Rs 20 or 40 lakh dispute, the same spend is close to irrational, and this is where readers most often ask whether arbitration is affordable at all. The candid answer is that pure ad hoc arbitration usually is not, for smaller matters. The fix, covered further below, is to specify an institution and a fee cap in the clause, or to keep smaller disputes in court or in mediation. Choosing arbitration by reflex, without sizing the cost against the claim, is the most common and expensive mistake.
Where arbitration goes back to court, section by section
Arbitration goes back to court more often than its “avoid the courts” reputation suggests, and the 1996 Act itself is what opens those doors. Section 5 promises minimal intervention, but the same Act then provides for court involvement at several defined points, and each one adds time. Mapping them makes the disadvantage concrete rather than vague.
Before the award
Two provisions bring the court in early. Under Section 9, a party can ask a court for interim measures (to freeze assets or preserve the status quo) before or during the arbitration, which is useful but is still a court proceeding. Under Section 11, when the parties cannot agree on an arbitrator, the court appoints one, and this stage has become a notorious source of delay. The rulings in TRF Ltd. and Perkins Eastman, useful as they are for independence, also spawned satellite litigation over who may appoint whom, and the Constitution Bench in Central Organisation for Railway Electrification added a further layer for government contracts. A dispute can sit at the appointment stage for a year before the tribunal is even constituted.
After the award
Three more provisions bring the court back once the award exists. Section 34 allows a challenge to set the award aside, Section 36 governs enforcement (and the conditions for any stay), and Section 37 lists the limited orders that can be appealed. The point is not that any single provision is unreasonable; it is that a losing party with resources can chain them (a Section 34 petition, then a Section 37 appeal) and keep a “final” award contested for years. This is a large part of why sophisticated Indian parties increasingly prefer a Singapore or London seat, where the supervisory courts intervene less and move faster. If the promise of arbitration was staying out of court, the reality is that you visit court at the start, sometimes in the middle, and often at the end.
Enforcing an arbitral award, domestic and foreign
Enforcing an arbitral award is a separate step from winning it, and the mechanics differ sharply between domestic and foreign awards. This is where arbitration’s greatest advantage (cross-border reach) and one of its quiet frustrations (enforcement is itself a court process) both live. Take them in turn.
Enforcing a domestic award
A domestic award is final and binding under Section 35 and is enforced as if it were a decree of the court under Section 36. For years, merely filing a Section 34 challenge operated as an automatic stay on enforcement, which let losing parties freeze awards indefinitely. That changed. In Hindustan Construction Company Ltd. v. Union of India, (2020) 17 SCC 324, the Supreme Court struck down the provision that had reinstated the automatic stay, confirming that a challenge does not by itself halt enforcement. An award-holder still has to file execution and wait out the challenge, but the balance has shifted back towards the winner. The limitation period to enforce a domestic award runs to twelve years.
Enforcing a foreign award under the New York Convention
For a foreign award, the picture is stronger. Sections 44 to 49 give effect to the New York Convention, so an award made in a notified Convention country is recognised and enforced in India, and once the court is satisfied under Section 49, it is deemed a decree. The grounds to resist are narrow and mirror Article V of the Convention; after the 2015 Amendment, an enforcing court cannot review the merits. The limitation period to seek enforcement of a foreign award is three years. Our detailed walkthrough of enforcing a foreign arbitral award in India covers the documentary requirements. For businesses running international contracts, this enforceability is exactly why the drafting of the dispute-resolution and enforcement clauses matters so much, a skill that structured programmes such as Skill Arbitrage’s course on international contract negotiation, drafting and enforcement are built around. In practice, a well-drafted foreign-seated arbitration clause can be the most valuable line in a cross-border contract.
Arbitration vs litigation: which to choose
Arbitration vs litigation is not a contest one side always wins; it is a fit question that depends on the dispute. The honest comparison below sets the main factors against each other without pretending arbitration is uniformly better. Use it as a checklist, not a verdict.
| Factor | Arbitration | Litigation |
|---|---|---|
| Cost | Can be high, especially ad hoc; predictable if institutional | Lower court fees, but long timelines add cost |
| Time | Faster in theory (Section 29A); slower in ad hoc practice | Often years, given court backlogs |
| Privacy | Confidential (Section 42A) | Public record and open court |
| Appeal | No merits appeal; narrow Section 34 challenge | Full appellate ladder |
| Decision-maker | Party-chosen subject expert | Assigned judge, generalist |
| Cross-border enforcement | Strong, via the New York Convention | Weak; decrees rarely enforced abroad automatically |
| Precedent | None; award binds only the parties | Builds binding precedent |
| Discovery | Limited | Full discovery available |
| Third parties | Hard to join non-signatories | Can implead third parties freely |
The factors that actually decide it
Read down that table and a pattern appears. Arbitration wins on privacy, expertise and cross-border enforcement; litigation wins on cost for small claims, on appeal rights, on precedent, and on reaching third parties. So the deciding factors are usually the nature of the counterparty (domestic or foreign), the value at stake, and whether you need a public precedent or a private result. A general counsel weighing a routine domestic recovery will often stay in court; the same counsel handling a cross-border technology licence will almost always arbitrate.
When litigation is the better call
Litigation remains the better route in several situations: when you need an injunction against a stranger to the contract, when the subject matter is non-arbitrable under Vidya Drolia, when a binding precedent would help your wider business, or when the amount is too small to justify arbitrator and counsel fees. Choosing arbitration in these cases, just because the contract had a clause, is a mistake that surfaces only once the costs mount.
Ad hoc vs institutional arbitration, and why it matters for cost
Ad hoc versus institutional arbitration is, in India, the single biggest lever on cost and delay. Most Indian arbitration is still ad hoc, meaning the parties and tribunal run the process themselves with no administering institution, and that is precisely where the fee and timeline problems concentrate. Understanding the difference is how you capture arbitration’s advantages while dodging its worst disadvantages. Our overview of the types of arbitration, including ad hoc and institutional, sets out the categories in full.
Why ad hoc arbitration drives up cost and delay
In an ad hoc reference there is no institution to fix fees, enforce timelines, or administer hearings, so arbitrators often bill per sitting and the schedule drifts. The “luxury litigation” example was ad hoc. This is a big part of why arbitration in India has not delivered the speed and economy it promises, and why so many Indian parties prefer foreign institutional seats. A useful comparison of how the two models stack up on cost appears in LawSikho’s analysis of how institutional and ad hoc arbitration compare on cost.
How institutional arbitration fixes part of it
An arbitral institution supplies rules, a fee schedule, timelines and administration, which brings discipline that ad hoc references lack. The trend is finally moving. The Delhi International Arbitration Centre reported a 2024 caseload of over 10,000 cases across more than 15,000 hearings, up from around 2,857 cases in 2020, per a survey of India’s institutional arbitration landscape. The Mumbai Centre for International Arbitration reported its caseload rising from 23 cases in 2023 to 34 in 2024, with a total disputed value around 257 million dollars, according to its 2025 annual report. Institutional cases tend to conclude closer to the statutory timeline, with far fewer extensions. In practice, the most reliable way to get the advantages of arbitration without the cost blowout is to name a credible institution in the clause.
The reform outlook: the 2024 Amendment Bill and recent rulings
The reform outlook is aimed squarely at the disadvantages, especially cost, delay and court intervention. A government expert committee and a draft amendment bill are both on the table, and the Supreme Court has been course-correcting through 2024 and 2025. None of this is settled law yet, so treat it as direction of travel rather than current rule.
What the 2024 Bill proposes
An expert committee chaired by a former Law Secretary was constituted in June 2023 and submitted its report in February 2024. The Department of Legal Affairs then released a draft Arbitration and Conciliation (Amendment) Bill, 2024 for public consultation in October 2024. As of mid-2026, it has not been enacted, so the operative statute remains the 1996 Act as amended through 2021. The draft proposals are ambitious: statutory recognition of an emergency arbitrator, a much-debated appellate arbitral tribunal, tighter clocks for appointing arbitrators and ruling on jurisdiction, and deletion of the Fourth Schedule fee grid in favour of institution-set fees. Our dedicated explainer on the draft Arbitration and Conciliation (Amendment) Bill, 2024 tracks the detail. Early signals suggest the reform is likely to lean hard towards institutional arbitration, though the appellate-tribunal idea is contested for the very reason that it could add another layer of challenge.
Recent Supreme Court course-corrections
The Court has not waited for Parliament. In Gayatri Balasamy v. ISG Novasoft Technologies Ltd., 2025 INSC 605, a five-judge bench held (by a 4:1 majority) that courts have a limited power to modify an arbitral award under Sections 34 and 37, for instance to sever an invalid part, correct a clerical or computational error, or adjust post-award interest. That is genuinely two-edged: it can save parties a costly fresh arbitration, but critics warn it chips at the finality that made arbitration attractive in the first place. Together with the appointment-fairness ruling in Central Organisation for Railway Electrification, it shows a judiciary trying to make arbitration fairer without making it slower, which is the whole balancing act this article is about.
When arbitration is worth it: a decision framework
Arbitration is worth it when the dispute is cross-border, commercially sensitive, technically complex, or high in value, and when you can route it through a credible institution. It is a poor fit for small domestic claims, matters needing precedent or third-party relief, and pure ad hoc references with no fee discipline. Rather than choose by reflex, run the dispute through a short framework.
When arbitration is the right choice
Lean towards arbitration when the counterparty is foreign and enforceability across borders matters, when confidentiality has real commercial value, when the dispute needs a subject-matter expert, or when the amount at stake comfortably justifies the cost. These are the situations where the advantages compound and the disadvantages stay manageable.
When to think twice
Pause when the claim is small relative to arbitrator and counsel fees, when you may need to bring in third parties, when the subject matter is non-arbitrable, or when a public precedent would serve your wider interests better than a private result. In these cases the disadvantages tend to dominate, and litigation or mediation may serve you better.
Drafting the clause to capture the advantages
Most of arbitration’s disadvantages are set at the drafting stage, not discovered later. A clause that names an institution, fixes the seat, specifies the number of arbitrators, and adopts an institutional fee model captures the speed and cost control that ad hoc references lose. A common and costly pitfall is a bare, one-line clause (“disputes shall be settled by arbitration”) that leaves every one of these choices open to a later fight. The advantages of arbitration are real, but they are earned in the clause, not assumed. Building that drafting skill is exactly what LawSikho’s Certificate Course in Arbitration is designed to teach.
Frequently asked questions
1. Is arbitration binding in India?
Yes. Under Section 35 of the Arbitration and Conciliation Act, 1996, an arbitral award is final and binding on the parties. It is enforced as a decree of the court under Section 36, and there is no appeal on the merits, only a narrow challenge to set it aside under Section 34.
2. Can you appeal an arbitration award in India?
There is no merits appeal. You can only apply to set the award aside under Section 34 on limited grounds, such as a violation of public policy or patent illegality, within three months (extendable by 30 days). A further appeal against the Section 34 order is possible under Section 37 on limited grounds.
3. How much does arbitration cost in India?
It varies widely. The Fourth Schedule caps arbitrator fees at Rs 30 lakh per arbitrator, but senior ad hoc arbitrators can charge per sitting and exceed this. Legal fees are usually the largest head, often around three-quarters of the total, and pursuing a Rs 1 crore claim can cost Rs 30 to 40 lakh end to end.
4. Is arbitration cheaper than litigation?
Not always, and often not for small disputes. Ad hoc arbitration between Indian parties has become expensive enough that one court called it “luxury litigation.” Institutional arbitration is more cost-disciplined, but court litigation still carries lower fees for low-value claims.
5. How long does arbitration take in India?
Section 29A requires the award within twelve months of the completion of pleadings, extendable by six months with party consent, and beyond that only by a court. In practice, extensions are granted in the large majority of cases, so ad hoc arbitrations often run well past the statutory clock.
6. Is the Fourth Schedule arbitrator-fee cap mandatory?
The Fourth Schedule is a model fee scale that caps fees at Rs 30 lakh per arbitrator, but it is not mandatory for international commercial arbitration or where an institution’s rules apply. In ad hoc domestic arbitration, courts increasingly treat it as the benchmark, though disputes over fees still arise.
7. Who pays the arbitrator’s fees?
The parties do, usually sharing the tribunal’s fees during the proceedings. Section 31A gives the tribunal power to allocate costs, so the final award can direct the losing party to bear or reimburse a share of the fees and legal costs.
8. Is arbitration confidential in India?
Yes. Section 42A, inserted by the 2019 Amendment, imposes a statutory duty of confidentiality on the arbitrator, the parties and the institution, except where disclosure is necessary to enforce or challenge the award. This is a genuine advantage over open-court litigation.
9. How is a foreign arbitral award enforced in India?
Foreign awards from notified New York Convention countries are enforced under Sections 44 to 49. The award-holder files in the appropriate court, which checks the narrow resisting grounds; once satisfied under Section 49, the court treats the award as a decree. The limitation period is three years.
10. On what grounds can an award be set aside under Section 34?
The main grounds are incapacity of a party, an invalid arbitration agreement, lack of proper notice, the award exceeding the scope of the reference, an improperly constituted tribunal, non-arbitrability, and conflict with the public policy of India (including fraud, or patent illegality for domestic awards). Courts cannot reappreciate evidence or review the merits.
11. Does arbitration create binding precedent?
No. An award binds only the parties to that dispute, and most awards are confidential and unpublished. Two arbitrations on similar facts can reach different outcomes, which is one reason a business seeking a settled legal rule may prefer litigation.
12. Can a non-signatory be joined to arbitration?
Only in limited circumstances. In Cox and Kings Ltd. v. SAP India Pvt. Ltd., the Supreme Court recognised the “group of companies” doctrine, so a non-signatory can be bound where there was a mutual intention to bind it. A genuinely unrelated third party cannot be forced into arbitration.
13. Arbitration vs litigation: which is better for a business?
It depends on the dispute. Arbitration suits cross-border, confidential, technical or high-value disputes, especially through an institution. Litigation suits small domestic claims, matters needing precedent or third-party relief, and non-arbitrable subject matter.
14. Ad hoc vs institutional arbitration: which is cheaper?
Institutional arbitration is usually more cost-predictable because the institution fixes fees, timelines and administration. Ad hoc arbitration, with per-sitting fees and no timeline discipline, is where India’s cost and delay problems concentrate.
15. What are the pros and cons of an arbitration clause?
The pros are privacy, expert decision-makers, finality and cross-border enforceability. The cons are cost, potential delay, limited appeal and difficulty involving third parties. Most of the downside is controlled by how well the clause is drafted, particularly by naming an institution, a seat and a fee model.
16. What is the status of the Arbitration Amendment Bill 2024?
The draft Arbitration and Conciliation (Amendment) Bill, 2024 was released for public consultation in October 2024 and, as of mid-2026, has not been enacted. The operative law remains the 1996 Act as amended through 2021. Key proposals include an emergency arbitrator, a contested appellate arbitral tribunal, tighter timelines, and deletion of the Fourth Schedule fee grid.
References
Case Law
- Associate Builders v. Delhi Development Authority, (2015) 3 SCC 49
- Bharat Aluminium Co. v. Kaiser Aluminium Technical Services Inc., (2012) 9 SCC 552
- Central Organisation for Railway Electrification v. ECI SPIC SMO MCML (JV), 2024 INSC 857
- Cox and Kings Ltd. v. SAP India Pvt. Ltd., (2024) 4 SCC 1
- Gayatri Balasamy v. ISG Novasoft Technologies Ltd., 2025 INSC 605
- Hindustan Construction Company Ltd. v. Union of India, (2020) 17 SCC 324
- In Re: Interplay Between Arbitration Agreements under the Arbitration and Conciliation Act 1996 and the Indian Stamp Act 1899, (2024) 6 SCC 1
- Oil and Natural Gas Corporation Ltd. v. Saw Pipes Ltd., (2003) 5 SCC 705
- Oil and Natural Gas Corporation Ltd. v. Western Geco International Ltd., (2014) 9 SCC 263
- PASL Wind Solutions Pvt. Ltd. v. GE Power Conversion India Pvt. Ltd., (2021) 7 SCC 1
- Perkins Eastman Architects DPC v. HSCC (India) Ltd., 2019 SCC OnLine SC 1517
- Ssangyong Engineering and Construction Co. Ltd. v. National Highways Authority of India, (2019) 15 SCC 131
- TRF Ltd. v. Energo Engineering Projects Ltd., (2017) 8 SCC 377
- Vidya Drolia v. Durga Trading Corporation, (2021) 2 SCC 1
Statutes
- The Arbitration and Conciliation Act, 1996 (as amended in 2015, 2019 and 2021), sections cited: 5, 7, 8, 9, 10, 11, 19, 20, 29A, 31A, 34, 35, 36, 37, 42A, 44 to 49, and the Fourth Schedule
Legal disclaimer
This article is for informational and educational purposes only and does not constitute legal advice. The law on arbitration in India continues to evolve through amendment and judicial interpretation, and the application of any provision depends on the specific facts of a dispute. For advice on a particular matter, consult a qualified legal professional.
