

Section 28(2) of the Arbitration and Conciliation Act, 1996 is invoked often and analysed rarely. The accusation that a tribunal has decided ex aequo et bono comes easily; the work of establishing it does not. This article attempts to draw the line that the provision actually draws.
What is equity, and what is it not? What exactly are the contours of Section 28(2)? More importantly, can an arbitral tribunal remain practical, realistic, and commercially sensible without stepping outside the contract, the governing law, and the evidentiary record?
The problem is not unusual. Commercial disputes rarely arrive with perfect measurements, complete records, and one immaculate formula waiting to be adopted. They arrive untidily. They arrive with gaps. They arrive with rival computations, each claiming mathematical virtue and each carrying some distortion of its own. Arbitration is then asked to do what commerce itself often fails to do: impose order without losing realism. Section 28(2) is the fence that separates the two.
The statutory starting point is clear. Section 28(1) deals with the law applicable to the substance of the dispute. Section 28(2) permits the arbitral tribunal to decide ex aequo et bono or as amiable compositeur only if the parties have expressly authorised it to do so. Section 28(3) requires the arbitral tribunal, while deciding and making an award, in all cases to take into account the terms of the contract and applicable trade usages. The structure reflects Article 28 of the UNCITRAL Model Law: equity-based adjudication is exceptional and consent-based.
But Section 28(2) is often invoked too loosely. A practical quantification is described as equity. A rough estimate is labelled ex aequo et bono. A plausible middle course is denounced as amiable composition. The accusation comes easily. The analysis often does not. The former needs only dissatisfaction. The latter requires reading the award.
The Indian case law develops along two parallel lines. One may be called the discipline line. The other, the practicality line. The first insists that an arbitral tribunal cannot abandon the contract and the law. The second recognises that an arbitral tribunal need not pretend to impossible precision where the record permits a rational estimate.
The discipline line begins early. In ONGC Ltd. v. Saw Pipes Ltd., the Supreme Court noticed Section 28(2) and recognised that ex aequo et bono decision-making requires express authorisation. That signal became sharper in National Buildings Construction Corpn. Ltd. v. P. Radhakrishna Murthy, where the Court stated the matter in clear terms: an arbitral tribunal is not a conciliator; it must decide according to the legal rights of the parties, not according to what appears fair and reasonable.
Associate Builders v. Delhi Development Authority then gave the principle its doctrinal structure. It treated contravention of Section 28(1)(a) and Section 28(3) as part of the patent illegality framework, while also protecting the arbitral domain by holding that interpretation of the contract is primarily for the arbitral tribunal. Ssangyong Engineering & Construction Co. Ltd. v. NHAI carried that logic into the post-2015 amendment framework. The wording of Section 28(3) shifted. The discipline did not. The contract still matters. It may be interpreted. It may not be rewritten. McDermott International Inc. v. Burn Standard Co. Ltd. is useful in the same way. Interpretation of contract and inferential reasoning remain within the arbitral domain. Not every imperfect or estimated award becomes a Section 28 problem.
That is where the practicality line becomes equally important. A rough and ready quantification rooted in the contract, the governing law, and the evidentiary record remains secundum legem. It does not become ex aequo et bono merely because it is not mathematically exact. The law has long recognised that damages and commercial loss do not always yield exact arithmetic. In Good Value Engineers v. M.M.S. Nanda, the Delhi High Court upheld “honest guess work” where the record contained sufficient material. In National Highway Authority of India v. ITD Cementation India Ltd., the Delhi High Court held that once reasonable material is available, the arbitral tribunal is entitled to make an honest guesstimate, and the court cannot substitute another percentage merely because it might have chosen differently. In Bharat Heavy Electricals Ltd. v. Delkon India Pvt. Ltd., the Delhi Division Bench upheld a mean approach in a fact-specific setting because the quantification remained anchored in the Local Commissioner's report, contractual material, and evidence on record. That is not equity. That is adjudication doing its job when the evidence refuses to be tidy.
Once those two lines are read together, the confusion begins to lift.
Section 28(2) is not triggered merely because the result appears fair. Courts must ask a more precise question: did the tribunal determine entitlement under the contract and law, and then quantify on the available material; or did it bypass entitlement itself and substitute a fairness-based compromise? The former is adjudication. The latter is unauthorised equity.
What Section 28(2) forbids, absent express authorisation, is not practical adjudication but a contra legem departure from the contract and governing law in the name of fairness. It prohibits the arbitral tribunal from deciding the merits on pure fairness, good conscience, compromise, or a private sense of commercial morality in substitution of legal and contractual rights. It catches the case where the legal question is never answered because a middle path feels fairer. It catches the award that overrides an express contractual bar because its operation appears harsh. It catches the disguised compromise dressed up as adjudication.
The arbitral tribunal is the creature of the contract. It may not wander outside the bargain because another result appears kinder. The contract is not clay in the hands of adjudication.
Recent cases also show why courts must be careful before loosely accepting a Section 28(2) objection. In TDI International India Ltd. v. Delhi Metro Rail Corporation, the Delhi High Court rejected the argument that partial remission of licence fee amounted to unauthorised ex aequo et bono adjudication, holding instead that the relief flowed from contractual interpretation, factual appreciation, and a finding of shared responsibility.
The distinction is not between a fair result and an unfair one. It is between an award that remains secundum legem and one that becomes ex aequo et bono or contra legem without consent. Commercial arbitration often demands practical judgment. That is not a weakness of the process. It is part of its design. Section 28(2) does not prohibit an arbitral tribunal from being realistic, practical, or commercially sensible. It only insists that those qualities must operate within the discipline of the contract, the governing law, and the evidentiary record.
There is, perhaps, a quiet elegance in that balance. Arbitration is not asked to ignore commercial reality. It is asked to confront commercial reality without deserting legal principle. An arbitral tribunal may estimate. It may infer. It may choose between rival methods. It may adopt a practical computation rooted in evidence. What it may not do, without express consent, is abandon law and contract for a private sense of justice. Those are the contours of Section 28(2). Not a ban on practical adjudication. A ban on unauthorised equitable adjudication.
About the author: Rishabh Gandhi is a former judge, arbitrator, and arbitration counsel specialising in commercial and construction disputes. He regularly appears in EPC and delay-related arbitrations. Gandhi is also the founder of Rishabh Gandhi and Advocates.
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.