When CA Certificates Win Damages Claims in Arbitration, and When They Sink Them: A Practitioner’s Perspective

A CA certificate can help the Arbitral Tribunal understand quantum, but only if it is tied to identified source records and a clear method.

In construction arbitrations, a claim for damages is only as strong as the evidence behind it. Project progress and costs are recorded across large volumes of site records, ledgers, vouchers, and plant and material logs. Parties often try to condense such extensive records in a single document, most commonly a chartered accountant (CA) certificate.

A CA certificate can help the Arbitral Tribunal understand quantum, but only if it is tied to identified source records and a clear method. If it is vague or if the figures cannot be traced back to the underlying books and contemporaneous documents, it becomes an easy target and can weaken an otherwise sound claim.

Two recent decisions of the Delhi High Court show the difference between a certificate that assists, and a certificate that collapses claims for damages. In Delhi Transco Ltd. v. KEC International Ltd.¹, the Delhi High Court dealt with CA certificates relied on to support idling and related losses, and found them too vague to carry the awards made on that basis. The High Court’s concern was not with the idea of taking accounting assistance, but with the absence of particulars, the absence of supporting records, and the fact that the certificate was not properly proved since the CA was never examined as a witness. By contrast, in NHAI v. Hindustan Construction Co. Ltd.², the Delhi High Court observed that certified compilations can legitimately assist an Arbitral Tribunal in deciding claims when they are anchored to project accounts and other relevant records.

The lesson is simple and often ignored: A CA certificate is not a shortcut to proof. Its weight depends on what it certifies, the records it relies on, the method it applies, and how it is introduced and tested in the arbitral record. A certificate that maps each head of claim to ledger extracts and project records is harder to dismiss, and far easier for the Arbitral Tribunal to rely upon with confidence.

Why CA certificates have become the default tool for proving quantum?

Construction disputes create a recurring proof problem. The events said to cause loss are operational and site-based, while the loss itself appears in operational accounts, vouchers, payrolls, plant and machinery logs, hire charges, and other allocations. Tribunals are frequently called upon to convert on-site disruptions into financial consequences (in the form of damages) pertaining to the long periods and multiple work fronts. In that setting, a CA certificate is commonly used to present, in a structured form, what the project books are said to show.

The certificate is most often used for prolongation and disruption claims, including extended site overheads, extended stay of plant and machinery, idling of labour and resources, and financing costs during the extended period. These heads usually require month-wise build-ups, linkage to cost centres, and reconciliation with audited project accounts. Certificates are also sometimes used to quantify loss of profit or contribution, by applying a margin to alleged shortfalls or unexecuted work.

The central mistake is to treat the certificate as evidence of entitlement or causation. An Arbitral Tribunal does not award damages because a professional has signed a document. It awards damages because the claimant proves 1) the existence of a contractual duty, 2) the breach of such duty, and 3) the loss that reliably follows from the breach. A certificate may help with the computation and the presentation, but it cannot fill gaps in causation, attribution, or contractual entitlement.

What a CA certificate should contain, and what it should avoid?

Practitioners often treat CA certificates as documents interchangeable with other founding records/evidence. They are not. Furthermore, a certificate that merely compiles figures from identified records is very different from one that embeds contested assumptions about attribution, causation, or entitlement. The more the certificate moves from extraction to judgment, the greater the need for transparency and testing.

At a minimum, a CA’s certificate used to support damages should answer five questions on its face. First, what exactly is being certified, and for which head of claim. Second, what period it covers and which contract, package, or project identifiers it relates to. Third, which source records were examined. Fourth, what method was applied, including allocation keys, assumptions, inclusions/exclusions, and any adjustments. Fifth, how the figures reconcile to the project books and, where relevant, to audited financial statements.

It is equally important to state what the certificate is not doing. A CA is not a witness of site events. The certificate should not purport to certify that delay was caused by the employer, that a particular period of extension is attributable to one party, or that resources were idle “due to hindrances”. Those are factual and contractual questions, proved through notices, correspondence, progress reports, hindrance registers, engineer’s decisions, and other site records. When a certificate strays into these areas, it invites a strong objection that it is making assertions outside the domain of accounting verification.

The certificate should also disclose the character of the figures it reports. It should state whether the amounts are actuals, accruals, provisions, etc. It should state whether it relied on original records, scanned copies, or internal summaries prepared for arbitration. These points matter because they go directly to reliability.

The case of Delhi Transco

In the Delhi Transco case³, the employer had filed an appeal under Section 37, Arbitration and Conciliation Act, 1996 (Arbitration Act) against a Single Judge’s decision which had upheld an arbitral award dated 29 January 2015. The award had allowed substantial monetary reliefs in favour of the respondent (KEC International), including amounts described as compensation for delay, interest components, and costs, taking the overall figure to a significant total.

A principal focus of the appeal was the quantification of certain heads that were said to have been awarded on the basis of CA’s certificates. The Delhi Transco case, in essence, was that the Tribunal had accepted those certificates without the supporting record that would ordinarily enable a proper assessment of the figures, and that the certificates had not been proved or tested in a manner that justified reliance for the sums awarded.

The High Court examined the relevant certificates and their broad formulation. The certificates in this case stated that figures were “verified” with reference to audited books of account, the contract ledger, and related documents, and then provided tabulated totals under expense heads. The High Court also noted the absence, within the certificates themselves, of details that would show how the amounts under each head were derived or what specific underlying records were used for the compilation.

Against that background, the High Court examined how the Arbitral Tribunal had dealt with the certificates. It noted that the Arbitral Tribunal had accepted the figures and allowed the claim amounts, despite objections to their admissibility and authenticity. The High Court also took note of the fact that the signatory CA was not examined as a witness.

The High Court focused on the lack of specific details and how the certificates were presented and relied upon when evaluating whether the award could be upheld. It also distinguished the respondent’s use of authorities where CA-certified statements were considered acceptable aids, noting that in those cases, the context depended on the contractual and evidentiary setting, and did not directly resolve the issues in the current facts.

The pitfalls identified by the High Court in CA certificates relied in Delhi Transco

In the Delhi Transco case⁴, CA certificates lost persuasive value because of how they were framed, what they contained, and how they were supported (or not supported) by the arbitral record. The first pitfall was vagueness on the face of the certificates. The certificates in this case stated that the figures were “verified” and then provided totals under the expense heads. They did not provide the breakdown within each head, the underlying records/supporting materials, or any mapping that would allow a reader to understand how the totals were compiled. The High Court treated this as a core deficiency because the certificate, by itself, did not show the route from source records to the claimed numbers.

The second pitfall was the lack of/non-identification of, any supporting material accompanying the certificates. When a certificate is presented without the underlying extracts or schedules that it claims to draw from, the decision-maker is left with a bottom-line figure and a general assertion that such figures are derived from records.

The third pitfall is related to how the certificates were introduced and tested. The signatory CA was not examined as a witness, and the certificates themselves were treated as not proved. In a case of contested quantum, these gaps matter because they directly affect reliability.

The fourth pitfall was the Tribunal’s approach to reliance. The High Court read the award as accepting the certificates without a visible process of verification of the figures, despite objections having been raised about admissibility, authenticity, and lack of particulars. In the High Court’s view, where the material placed is a sparse certification with no accompanying trail, uncritical acceptance creates perversity.

The case of NHAI — a counterpoint

In the NHAI case⁵ is a useful counterpoint because it shows when certified compilations may be treated as a legitimate aid in quantification. The dispute arose from a highway project where the contractor sought additional costs for the extended period. The High Court’s focus was on whether the Arbitral Tribunal’s quantification of damages was anchored in identifiable material and approached with scrutiny.

The High Court, while dealing with NHAI‘s challenge to the Tribunal’s reliance on CA-certified material, held that the Arbitral Tribunal’s approach to quantification could not be faulted merely because it drew upon CA certificates. It noted that the certificates were rooted in books of account maintained in the ordinary course of business and audited by statutory auditors, and had in fact been furnished to the project engineer during contract administration along with supporting trial balance material. On that footing, and in the absence of anything shown by National Highways Authority of India (NHAI) to demonstrate why the audited certifications should not be relied upon in arbitration, the High Court accepted that CA certificates could legitimately be treated as reflective of actual overhead expenditure.

The High Court’s decision in NHAI case⁶ is useful because it treats certification as potentially helpful, but not self-proving. Notably, the High Court had set aside the amounts awarded under the head of plant and machinery retention costs, while leaving the rest of the Arbitral Tribunal’s quantification intact.

Principles distilled from Delhi Transco and NHAI

Read together, the Delhi Transco case⁷ and NHAI case⁸ do not set out competing rules. They show a spectrum. At one end, certificates are treated as useful compilations because they are linked to relevant project records and have been thoroughly examined by the Arbitral Tribunal. At the other end, certificates carry little weight when they are broad, untethered to identified records, unsupported by a traceable trail, and presented before the Arbitral Tribunal without being properly explained or proven through witness examination.

The principles that may be distilled from an examination of the judgments in the Delhi Transco case⁹ and NHAI case¹⁰ are as follows:

1. A certificate that aligns with project documentation is generally easier to evaluate and more likely to withstand scrutiny than a certificate produced for the first time in arbitration as a stand-alone statement with no traceable foundation in the arbitral record.

2. A certificate may assist in presenting quantum, but it cannot, by itself, establish the contractual and factual foundations necessary to prove damages.

3. The evidentiary weight of the certificates increases where its author is examined, and where the Arbitral Tribunal has a real opportunity to test and scrutinise the certified contents against the underlying record.

Practitioner’s checklist: making a CA certificate work for quantum

A CA certificate is best viewed as a structured presentation of records, not as a replacement for them. When the certificate is treated as an indexed gateway to underlying material, it can be useful. When it is presented as a stand-alone conclusion, it is vulnerable to being discounted. The following checklist sets out illustrative factors that make a CA certificate usable for the Arbitral Tribunal and defensible on review:

1. Clearly state the exact head of claim, contract/package identifier, and accounting period covered.

2. Identify source records with specificity, including ledger heads, trial balance extracts, voucher numbers/series, payroll registers, plant and hire invoices, fuel logs, and any other relevant documents.

3. Explain how the figures were computed, including the assumptions, inclusions, exclusions, and other adjustments.

4. Provide a reconciliation to audited project accounts or audited financial statements where relevant.

5. Link each sub-head and amount to the relevant pleading paragraph and exhibit or document reference.

6. Make clear what the certificate does not cover, including attribution of delay, entitlement, causation, and responsibility.

7. Ensure that the CA is examined and the underlying basis is available for proper testing before the Arbitral Tribunal.

Conclusion

A CA certificate can either serve as a structured summary of a record that is already coherent and traceable, or be used as a substitute for that record. The Delhi High Court largely accepts the former approach, under which certified compilations are rooted in audited accounts and relevant documentation, and the Arbitral Tribunal’s reasons show that it scrutinised the figures rather than deferring to them. The High Court is sceptical of the latter approach, especially where certificates are broad, unsupported by schedules or primary material, and introduced without a clear trail or meaningful testing on the arbitral record.

The takeaway for practitioners from the decisions in the Delhi Transco case¹¹ and NHAI case¹² is to treat CA certificates as a transparent gateway to the otherwise voluminous underlying books and documents, with the computation methods and assumptions made visible, so that each head of claim can be tested and assessed on its own merits rather than presented as a conclusory bottom line.