2025 (9) TMI 374 - HC - Income Tax

Revenue appeal dismissed: additions based on speculative 89% yield and rejected accounts set aside for lack of evidence

HC dismissed Revenue's appeal, upholding CIT(A) and ITAT findings that additions for alleged unaccounted production and sales based on an estimated 89% ... Addition of suppressed its yield and had indulged in unaccounted production and sales - addition on account of unaccounted sales based on an estimated production yield of 89% in the assessee’s SMS Division - AO adopted an estimated yield ratio and proceeded to calculate alleged unaccounted production and consequential sales, resulting in substantial additions over multiple years - ITAT deleted addition HELD THAT:- CIT(A) and the ITAT, both, after objectively analysing the factual situation, found complete absence of any adverse material against the assessee which can support the allegation of the AO towards unaccounted production presumed on the basis of alleged low yield declared by the assessee. Thus, in complete absence of any adverse material, both the authorities have concurrently reached to the conclusion that the addition made by the AO is baseless and without any evidence, therefore, the rejection of books of accounts is invalid and addition made by the AO on account of alleged suppression of yield is based upon mere guess work. Yield declared by the assessee is neither low nor the books maintained by the assessee could be impeached by some tangible evidence/material on record and therefore the ITAT has rightly confirmed the order of the CIT (Appeals) and proceeded to dismiss the appeal filed by the Revenue. In our considered opinion, the concurrent finding recorded by the two authorities holding that the addition made is baseless and without any evidence/material, is a pure and simple finding of fact based on the evidence available on record, which is neither perverse nor contrary to the record. Decided against the Revenue. ISSUES PRESENTED AND CONSIDERED 1. Whether the addition made by the Assessing Officer on account of alleged unaccounted production and sales-computed by adopting an estimated production yield (89%)-could be sustained in the absence of tangible material linking seized information to the assumed yield discrepancy. 2. Whether reliance on mathematical/statistical estimation and presumed low yield, arrived at from search-seized material, suffices to reject books of account and make additions under the relevant assessment provisions, or whether something more than suspicion is required. 3. Whether concurrent factual findings of the first appellate authority and the Tribunal that there was no adverse material to impeach the books of account are susceptible to interference by the High Court as per the record before it. ISSUE-WISE DETAILED ANALYSIS Issue 1 - Validity of addition based on estimated production yield (89%) Legal framework: The Assessing Officer made additions under the assessment provisions applicable post-search, estimating unaccounted production and sales by applying a standard yield percentage (89%) to the SMS Division's inputs/outputs. Rejection of books and making additions by inference is governed by the requirement that an assessment must be based on material and not mere conjecture. Precedent treatment: The Court invoked the principle articulated in Dhakeswari Cotton Mills (Constitution Bench) that an income-tax authority cannot make a 'pure guess' or act on 'bare suspicion'; there must be something more than suspicion to support an assessment founded on estimation. Interpretation and reasoning: The Assessing Officer's methodology rested on statistical variations in consumption (electricity, raw materials) vis-à-vis declared production, presumed burning/processing losses being estimated, and the alleged non-correlation of inputs and outputs. The appellate authorities examined seized material, excise records, loose slips, comparative industry yields and financial results, and found (i) absence of disclosure of the source/basis for the 89% figure, (ii) inability to reconcile AO's statistical approach to tangible proof of undisclosed sales, and (iii) that the assessee's declared yields were comparable or better than industry averages in available data. The AO did not point to any specific defect in books or contra documentary evidence; instead he relied on statistical inference without establishing causal/material nexus to unaccounted sales. Ratio vs. Obiter: Ratio - Where an Assessing Officer adopts an estimated standard yield to infer unaccounted production/sales, the estimate must be supported by tangible material linking the estimate to undisclosed income; absent that, the assessment based on such pure estimation is unsustainable. Obiter - Observations about capacity utilization and periodic shutdowns affecting yield are factual considerations applied to the instant record. Conclusions: The addition based solely on the adopted 89% yield, unsupported by demonstrable material showing suppression of production/sales or infirmity in books, was held to be based on suspicion and guesswork and therefore invalid. Issue 2 - Permissibility of mathematical/statistical inferences from search material to reject books of account Legal framework: Authorities are not strictly bound by courtroom standards of evidence in tax assessments but must act on material that is more than mere conjecture; Section provisions permitting assessment after search require factual foundation for adverse conclusions and rejection of books. Precedent treatment: The Court applied Dhakeswari Cotton Mills to reiterate that estimation without material is impermissible. The appellate authorities also relied on co-ordinate decisions in similar factual matrices where identical standardized yields were struck down (noted by Tribunal as persuasive/parallel decisions). Interpretation and reasoning: Statistical variations per se (wide month-to-month fluctuation in power/raw material consumption vs production) do not ipso facto demonstrate suppression unless the AO identifies specific discrepancies in primary records (excise returns, invoices, production registers) showing misstatement. The AO's failure to disclose the derivation of the 89% benchmark or show how seized loose slips contradicted excise/regular books meant the mathematical exercise lacked the necessary evidentiary foundation. Comparative analysis of peer yields and financial performance further undermined the premise that a lower declared yield necessarily equated to unaccounted sales. Ratio vs. Obiter: Ratio - Mathematical/statistical exercises cannot replace tangible corroborative material; statistical inference must have an evidential anchor before books can be rejected. Obiter - The specific methodology of the AO (e.g., reliance on highest/lowest consumption months) was criticized as over-reliance on routine statistics without showing how those statistics, when correlated, establish undisclosed sales. Conclusions: The Court affirmed that while authorities may use non-technical material, the use of statistics must be accompanied by tangible linking material; absent such linkage, rejection of books and consequent additions are unsustainable. Issue 3 - Scope for interference with concurrent factual findings of appellate authorities Legal framework: High Court's interference with concurrent findings of fact recorded by the Commissioner (Appeals) and the Tribunal is limited to instances of perversity or error apparent on record; findings based on evidence, even if differenced in evaluation, are generally respected. Precedent treatment: The judgment treats the appellate and Tribunal conclusions as factual assessments and applies settled standards for judicial interference with such concurrent findings. Interpretation and reasoning: Both first appellate authority and Tribunal conducted detailed factual analyses: they examined seized documents, excise returns, industry comparisons, witnesses' statements, financial results, and capacity-utilization explanations. They found no adverse material pointing to suppression; the AO did not identify defects in books, nor disclose basis for 89% standard. The Tribunal also noted co-ordinate bench decisions in an identical search context where identical yield-based additions were set aside. Given these concurrent, reasoned findings, the High Court found no perversity or legal error warranting interference. Ratio vs. Obiter: Ratio - Concurrent findings deliberately and objectively based on evaluation of evidence will not be disturbed unless perverse or unsupported by record. Obiter - Observations about uniformity of yields across peers and the sufficiency of excise records as corroboration are factual amplifications supporting that ratio. Conclusions: The concurrent determinations that additions were baseless and founded on conjecture were factual findings not vitiated by perversity; therefore the Court declined to interfere and dismissed the Revenue's appeal. Cross-references Refer to Issue 1 and Issue 2: The impermissibility of basing additions on the 89% yield (Issue 1) flows from the principle that statistical/mathematical inference absent tangible linking material is insufficient (Issue 2), and both inform the limited scope for disturbing concurrent factual conclusions (Issue 3).