CUP method preferred for rice export transfer pricing; TNMM rejected and computation issues remanded for fresh consideration.

For rice export transfer pricing, CUP was preferred over TNMM because reliable customs-based export data and product/geographical comparables were available, and no change in the factual matrix justified departing from the assessee's earlier year decision. The rejection of CUP and adoption of TNMM were therefore set aside. The assessee's additional ground on set-off of above- and below-arm's-length transactions under CUP was admitted and restored for reconsideration by the transfer pricing authority along with fresh computation, leaving the recomputation question open on remand.

---

---

Issues: (i) Whether, on the facts of the case, the Comparable Uncontrolled Price method or the Transactional Net Margin Method was the most appropriate method for determining the arm's length price of the assessee's international transactions in rice exports. (ii) Whether, while applying the CUP method, the assessee was entitled to seek set-off of transactions priced above and below the arm's length price and have the matter reworked accordingly.

Issue (i): Whether, on the facts of the case, the Comparable Uncontrolled Price method or the Transactional Net Margin Method was the most appropriate method for determining the arm's length price of the assessee's international transactions in rice exports.

Analysis: The dispute concerned selection of the most appropriate transfer pricing method for export of rice. The assessee relied on CUP using customs-based export data, while the transfer pricing authorities rejected CUP and applied TNMM. The earlier decision in the assessee's own case for an earlier year had already held that the customs data and product/geographical categorisation could be used reliably under CUP, and that traditional transaction methods have an inherent edge where they can be applied with reasonable reliability. Following that binding factual matrix and finding no change in circumstances, the same approach was applied.

Conclusion: The CUP method was held to be the most appropriate method and the rejection of CUP in favour of TNMM was set aside, in favour of the assessee.

Issue (ii): Whether, while applying the CUP method, the assessee was entitled to seek set-off of transactions priced above and below the arm's length price and have the matter reworked accordingly.

Analysis: The additional ground raised a legal question requiring no fresh factual inquiry. The matter was linked to how the CUP computation should be performed, including whether individual transactions could be aggregated or benchmarked in a manner that reflected the overall package of closely linked transactions. Since the entire transfer pricing exercise was being sent back for fresh determination under CUP, the question of set-off and manner of recomputation was left to be examined by the transfer pricing authority after hearing the assessee.

Conclusion: The additional ground was admitted and restored to the transfer pricing authority for reconsideration under CUP, in favour of the assessee for statistical purposes only.

Final Conclusion: The transfer pricing adjustment based on TNMM did not survive, and the matter was sent back for recomputation under CUP with liberty to the assessee to raise all computation-related contentions.

Ratio Decidendi: Where reliable comparable market data is available for a commodity transaction, CUP is to be preferred over TNMM if it can reasonably determine the arm's length price, and computation issues incidental to CUP may be reconsidered on remand.

---