Transfer pricing comparability requires reliable year-specific data, consistent loss filters, scale analysis, and LIBOR-based receivables benchmarking

Transfer pricing comparability must use reliable financial data and functional, economic, and scale-based analysis. A company following a different accounting year cannot be excluded where published quarterly results permit construction of the relevant year and the FAR profile is comparable. Rheal Software Pvt. Ltd. could not be rejected under the persistent loss filter without a defined and consistently applied criterion. Very high-turnover, branded companies were not comparable to a smaller captive software service provider because scale and brand advantages affected margins. Delayed foreign-currency receivables had to be benchmarked using an appropriate LIBOR-based rate rather than SBI PLR, requiring recomputation of the transfer pricing adjustment.

---

---

Issues: (i) Whether a software development comparable could be excluded solely because it followed a different accounting year where quarterly financial data was available for constructing the relevant year results; (ii) Whether Rheal Software Pvt. Ltd. was rightly excluded by applying the persistent loss filter; (iii) Whether very high-turnover and branded companies were comparable to the assessee in the software development services segment; (iv) Whether interest on delayed receivables denominated in foreign currency could be computed using SBI PLR instead of LIBOR-based foreign currency rates.

Issue (i): Whether a software development comparable could be excluded solely because it followed a different accounting year where quarterly financial data was available for constructing the relevant year results.

Analysis: The transfer pricing analysis under Section 92CA of the Income-tax Act, 1961 required selection of functionally comparable companies for the relevant financial year. A March year-end filter was applied to reject a listed company following a different accounting year. Since a listed entity publishes quarterly financial results, meaningful year-specific data could be reconstructed from those published results. Where such data is available in the public domain and the FAR profile is otherwise comparable, mere difference in accounting year does not justify exclusion.

Conclusion: This issue was decided in favour of the assessee. R Systems International Ltd. could not be excluded merely for following a different accounting year; if the relevant financial results are produced and found correct, it must be included as a comparable.

Issue (ii): Whether Rheal Software Pvt. Ltd. was rightly excluded by applying the persistent loss filter.

Analysis: The persistent loss filter must rest on a defined and consistent criterion. The record showed losses in two financial years but profit in the intervening year. In the absence of a stated rule that losses in two out of three years would lead to exclusion, the company could not be treated as a persistent loss-making concern. The earlier acceptance of the company as functionally comparable also supported reconsideration of its exclusion.

Conclusion: This issue was decided in favour of the assessee. Rheal Software Pvt. Ltd. was wrongly excluded under the persistent loss filter and was directed to be included.

Issue (iii): Whether very high-turnover and branded companies were comparable to the assessee in the software development services segment.

Analysis: Comparability in transfer pricing requires regard to scale, market position and brand-driven advantages. The assessee's software development revenue was about Rs. 70.44 crores, whereas several selected comparables had turnovers running into many multiples and also carried substantial brand value. Such giant entities operate in materially different economic circumstances, affecting margin comparability. Their inclusion would distort benchmarking for a comparatively small captive service provider.

Conclusion: This issue was decided in favour of the assessee. Mindtree Ltd., L&T Infotech, Wipro Ltd., Tata Elxsi Ltd., Infosys Ltd., Tata Consultancy Services and Cybage Software Pvt. Ltd. were directed to be excluded from the software development comparability set, with fresh computation of margin and adjustment thereafter.

Issue (iv): Whether interest on delayed receivables denominated in foreign currency could be computed using SBI PLR instead of LIBOR-based foreign currency rates.

Analysis: The disputed adjustment related to delayed realization of receivables treated as a separate international transaction. Where invoices are raised in foreign currency, the benchmark interest rate must correspond to foreign currency lending conditions rather than domestic rupee lending rates. Even the transfer pricing analysis acknowledged that foreign currency invoices require adoption of prevailing LIBOR with appropriate mark-up. Application of SBI PLR in such circumstances was therefore not appropriate.

Conclusion: This issue was decided in favour of the assessee. Interest on overdue foreign currency receivables was directed to be recomputed using an appropriate LIBOR-based rate and not SBI PLR.

Final Conclusion: The transfer pricing comparability exercise for the software development segment was modified by directing inclusion of two companies, exclusion of multiple giant companies, and recomputation of margins; the receivables adjustment was also directed to be recomputed on a foreign currency benchmark basis.

Ratio Decidendi: In transfer pricing analysis, a comparable cannot be excluded solely for a different accounting year where reliable quarterly data permits construction of the relevant financial year results; persistent loss exclusion requires consistent application of a defined criterion; giant high-turnover branded companies are not comparable to a smaller captive software service provider; and delayed foreign currency receivables must be benchmarked using an appropriate LIBOR-based rate rather than SBI PLR.

---