Natural justice in tax assessment requires full disclosure of adverse material before prejudicial findings are drawn.

An assessment based on import-discrepancy material was held unsustainable where the Department relied on incomplete CBIC data and did not share all material with the assessee. The Court applied the principle of natural justice, holding that adverse tax conclusions cannot be drawn from material affecting liability unless the assessee is given full disclosure and a meaningful opportunity to rebut it. Any further material received by the Revenue could be used only after being shared and after fresh opportunity to respond. The assessment order was therefore set aside and the matter remitted for fresh adjudication.

---

---

Issues: Whether the assessment order was vitiated for violation of principles of natural justice on account of reliance on incomplete CBIC data and non-sharing of material with the assessee.

Analysis: The assessment was founded on import-related discrepancy data, but the record showed that complete invoice-wise CBIC material was not available with the Department and that the assessee had not been given all information relied upon for the adverse inference. Where the authority proceeds on material affecting the tax liability, fairness requires that the material in its possession be disclosed to the assessee and that adverse conclusions not be drawn on incomplete data without affording an effective opportunity to meet the case. The Court also noted that any further material received by the Revenue could be used only after being shared with the assessee and after granting appropriate opportunity.

Conclusion: The assessment order was vitiated for breach of natural justice and was set aside, with the matter remitted for fresh adjudication after sharing the material with the assessee and granting opportunity to respond.

Ratio Decidendi: An assessment based on adverse material cannot be sustained unless the material is fully disclosed to the assessee and a meaningful opportunity is given to meet it; incomplete data cannot be used to draw prejudicial inferences.

---