By- Bhavi Pungliya, Phd Scholar, Sri Aurbindo Institute of Law, Indore, Madhya Pradesh
Abstract
For well over a century, the Bankers’ Books Evidence Act, 1891 governed a banking system organised around physical ledgers, and Indian courts proved bank records through certified paper copies accordingly. The Bankers’ Books Evidence Bill, 2026, passed by the Lok Sabha on 5 August 2026, repeals the 1891 Act and, for the first time, brings electronic, digital, virtual and cloud-based records expressly within the statutory meaning of “bankers’ books.” This paper examines whether that expanded definition, together with the Bill’s two-track certification scheme, is enough to establish the authenticity, integrity and reliability of banking records now generated, processed and stored through core banking systems, application programming interfaces, distributed databases and third-party cloud infrastructure, or whether the move from paper to cloud calls for evidentiary safeguards beyond certification alone. Drawing on a doctrinal reading of the Bill’s text, the Bharatiya Sakshya Adhiniyam, 2023, the Information Technology Act, 2000, and relevant Reserve Bank of India directions, alongside case law under Section 65B of the former Evidence Act and Section 63 of the Bharatiya Sakshya Adhiniyam and comparative material from the United Kingdom, the United States and Singapore, the paper contends that the Bill’s certification model performs a valuable but partial function: it mainly confirms that a copy was made and produced, rather than vouching for the integrity of the distributed, third-party-operated systems that generate today’s banking records. The paper’s original contribution is to show, through the comparative lens of the Post Office Horizon prosecutions in the United Kingdom, that a certification-only model creates a structural risk of misplaced trust in system reliability at exactly the point where that reliability is hardest to verify. It sets out a differentiated, risk-based framework built on layered certification, minimum disclosure of audit-trail and access-control information, and a clearer allocation of evidentiary responsibility between banks and third-party cloud operators, weighed against the offsetting costs of compliance burden and litigation delay. The analysis is constrained by the fact that, at the time of writing, the 2026 Bill has not yet been judicially interpreted and public technical detail about individual banks’ cloud architectures remains scarce.
Keywords: Bankers’ Books Evidence Bill, 2026; electronic evidence; digital banking; cloud evidence; Bharatiya Sakshya Adhiniyam, 2023; certification; evidentiary reliability.
Introduction
Indian banking evidence law rested, for 135 years, on an assumption that no longer reflects how banks actually keep records: that a bank’s “books” are physical ledgers, day-books and cash-books, held at one identifiable location, that could be inspected, copied and certified by an officer who had personally seen the original page. The Union Finance Minister introduced the Bankers’ Books Evidence Bill, 2026 in the Lok Sabha on 3 August 2026, and the House passed it two days later, on 5 August 2026, with little extended debate. The Bill repeals the Bankers’ Books Evidence Act, 1891 in full and substitutes a text that keeps the essential architecture of the old law, certified copies as prima facie proof, limited compellability of bank officers, and court-ordered inspection, while adding a new and interpretively demanding layer: the electronic or digital record.
The legislative intent here is plain. Section 2(1)(b) of the Bill defines “bankers’ books” to cover not just ledgers and account books kept in written or physical form, but every record “stored in any form of data storage mechanisms such as electronic or digital form, or otherwise, either onsite or at any offsite or virtual or cloud location, including a back-up or disaster recovery site.” Section 6, through a non obstante clause, states that admissibility of an electronic or digital record of a banker’s book “shall not be denied on the ground that it is an electronic or digital record,” subject to the conditions in Section 7. On its face, this is a thorough and long-overdue update. But a harder question sits beneath the surface, and it is the question this paper takes as its central concern: does naming cloud-based and distributed records within the statutory definition of “bankers’ books,” and attaching a certificate to them, genuinely rebuild the evidentiary assurances that certifying a physical ledger once supplied? Or does the technological distance between a paper ledger, inspected and copied by hand, and a record assembled across core banking systems, application programming interfaces, replicated databases and third-party data centres, demand something beyond an expanded definition and a signature?
.jpeg)
