The case where a company’s own books beat its best defence. Sit in the Gallery for a ₹6,000-crore claim, a tribunal rebellion, and a 1961 rule that settled everything. Step up to the Bar for the compulsion fallacy and the caveat rule. Take the Bench for the over-readings already loose in the market.
Most readers start in the Gallery — and climb before they realise it.
The plant. 2009: Corporate Power Ltd. builds a thermal power project in Jharkhand on loans from SBI and a consortium of lenders. By July 2013 the account is declared bad; by 2015 the frustrated banks hand their debts to a specialist — an asset reconstruction company.
The late knock. December 2018 — more than five years after the account went bad — the ARC files an insolvency petition for ₹5,997.80 crore. On the surface, hopelessly late. But the company’s own signed balance sheets, year after year, kept recording the debt as owing.
The wall. The NCLT admits the petition — the balance sheets restarted the clock. But the appeals tribunal has a problem: its own Full Bench (V. Padmakumar, decided 4–1) had ruled that balance sheets can never extend limitation — because companies are forced by law to file them, and a forced statement is no admission.
The rebellion. A three-member NCLAT bench reads the Supreme Court’s precedents, doubts its own Full Bench, and refers the question to a larger bench. The five-member bench refuses to even look — the reference itself, it says, is “incompetent”. Stalemate, with thousands of crores hanging on one question.
The library speaks. The Supreme Court walks through six decades of authority — from a 1961 Calcutta case about a silk mill to its own rulings — and finds one line has ruled all along: the law compels a company to prepare a balance sheet, but it never compels it to admit any particular debt. What the company chooses to record, it records.
The verdict. A signed balance-sheet entry can acknowledge a debt and restart the clock — but whether it does is decided case by case, reading the entry with its notes and the auditor’s caveats. V. Padmakumar is swept aside; the matters go back to be decided under the new rule (15 April 2021).
Epilogue — exhibit one, forever. The company’s own books are now the first exhibit in every limitation fight under the Code — and the caveat door stays open for debtors. Three years later, the rule decided Vidyasagar Prasad v. UCO Bank — No. 3 in this series — where a single balance-sheet note handed a bank three more years.
Because the compulsion runs out before the admission begins: the law compels a company to prepare a balance sheet — it never compels it to admit any particular debt in it. What the company chooses to record of its own liabilities, signed by its own directors, is its own statement.
And the balance: a company that genuinely disputes a debt can say so — in the notes, or through the auditor’s caveats — and such a qualified entry may acknowledge nothing. That is why the rule is case-by-case, not automatic.
A judgment is authority only for what it decides. Fix the questions before you read a single answer.
The bridge, again: does s. 18 of the Limitation Act apply to s. 7 IBC proceedings at all — or do the Code’s special features (the Explanation to s. 7, the nature of CIRP) shut it out?
The compulsion problem: can an entry in a balance sheet be an acknowledgment under s. 18 when the law itself compels companies to prepare and file balance sheets?
The limits: if it can — is the acknowledgment automatic, or is it controlled by the notes, the auditor’s caveats, and the facts of each case?
Two layers, so you always know whose voice you are reading: Our reading is this page’s interpretation — we state it and we stand behind it. The Court is the judgment verbatim, para-numbered from the certified copy; where the Court speaks through an earlier case, we name it and the paragraph where this judgment quotes it.
Our readingSection 18 applies to s. 7 proceedings through s. 238A — a question already settled twice over, which no fresh ingenuity reopens (Paras 8–10).
Our readingThe compulsion argument confuses two different things: the law compels the document, never the admission. A signed balance-sheet entry can therefore acknowledge a debt — the 1961 Bengal Silk Mills rule is expressly held correct (Para 22).
Our readingWhether an entry actually acknowledges is a question of fact in every case: the entry is read with the notes (legally part of the statements — s. 134(7)) and the auditor’s report, where caveats can strip an entry of its unequivocal character (Para 22).
Our readingThe acknowledgment need not be addressed to the creditor, and entries under general heads can suffice — six decades of approved authority travel with the rule (Paras 14–16, 23, 32).
Our readingThe NCLAT Full Bench majority is set aside as contrary to the catena of binding authority (the lone dissent vindicated), the five-member order that “rubber-stamped” it goes with it, and the lead matter is remanded to be decided under the law now laid down (Paras 33–35).
1 · Close the settled question fast. Section 238A’s bridge had been crossed twice already — Sesh Nath Singh and Laxmi Pat Surana (quoted at Paras 8–9) — so the attack on s. 18’s applicability is dispatched with a five-judge maxim: novelty is not a ground to reopen what is settled (Para 10).
2 · Let the library do the work. Against one Full Bench majority, the Court assembles sixty years of consistent authority — its own judgments, and High Courts from Calcutta to Telangana (Paras 14–32) — and shows V. Padmakumar considered none of it. Precedent falls to weight, not rhetoric.
3 · Split the compulsion. The Companies Act sections (Paras 21–22) compel preparation, signature, audit and filing — but leave the content of what is admitted to the company, with notes and auditor’s caveats as the safety valve. Compelled form, voluntary substance: the entry can acknowledge, the caveat can save.
“…there is a compulsion in law to prepare a balance sheet but no compulsion to make any particular admission…” — Para 22, adopting Bengal Silk Mills (1961): one sentence, sixty years old, ends a tribunal war.
We read the certified copy end to end so you don’t repeat what the summaries got wrong. Both findings are checkable by anyone with the PDF.
Wrong. The holding is can, never does: each entry is examined “on a case by case basis”, read with its notes and the auditor’s report — and an entry “entered into with caveats” may acknowledge nothing. The debtors’ caveat argument was noted, not rejected: it lives on for the remand and every later case.
The move: never cite this case bare — walk the tribunal through the specific entry, its notes, and the auditor’s words. Defending, those same pages are your battlefield.
✓ verified against certified copy · Para 22Wrong — the companion appeals prove it. Two creditors holding the very evidence this judgment blesses had never pleaded it: one was “completely remiss and deficient in pleading acknowledgement of liability”, and written submissions were held to be no substitute. They survived only by a costs-laden second chance.
The move: plead the acknowledgment chain in the petition itself — each balance sheet, each date, each signature. Opposing, attack the pleadings before you ever reach the entries.
✓ verified against certified copy · CA 3765/2020, Paras 4–6A judgment is a tool with two edges. From the Bench you must see both — you will not always be on the same side of it.
Is the debt reflected in a signed balance sheet of the corporate debtor?Paras 14, 22 — books of account can acknowledge under s. 18
Was that balance sheet signed before the current limitation window expired?Laxmi Pat Surana, quoted at Para 9 — a dead claim cannot be revived
Read with the notes and the auditor’s report, is the entry unequivocal — free of caveats?Para 22 — the case-by-case examination
Has the acknowledgment actually been pleaded in the application (or properly amended in)?CA 3765/2020 — written submissions are no substitute for pleadings
Answer the four questions to see whether the balance sheet saves the petition.