Transfer pricing benchmarks were revised: domestic items excluded, comparables filtered, LIBOR applied to receivables, and rupee debentures linked to domestic rates.

In transfer pricing, operating margins must be computed by confining the analysis to international transactions alone, so domestic revenue and expenditure were excluded from the margin computation. The related party transaction filter was required to be applied consistently on an aggregate basis, and comparables failing that correctly computed filter or lacking functional comparability were excluded, while entities supported by a valid database search could be retained subject to verification. Interest on delayed receivables from associated enterprises was benchmarked at LIBOR plus 200 basis points, whereas interest on Indian-currency compulsorily convertible debentures had to be benchmarked with SBI Prime Lending Rate rather than SIBOR. The assessment was therefore recomputed on these transfer pricing issues.

---

---

Issues: (i) whether domestic revenue and domestic expenditure were to be excluded while computing the operating margin for benchmarking international transactions; (ii) whether the related party transaction filter was required to be applied on an aggregate basis across revenue and expenditure; (iii) whether the disputed comparables were liable to be excluded or included on the basis of related party transactions, functional comparability, and database search results; (iv) whether interest on delayed receivables from associated enterprises was to be benchmarked by applying LIBOR plus 200 basis points; and (v) whether interest on compulsorily convertible debentures denominated in Indian currency was to be benchmarked by applying SBI Prime Lending Rate instead of SIBOR.

Issue (i): whether domestic revenue and domestic expenditure were to be excluded while computing the operating margin for benchmarking international transactions

Analysis: The transfer pricing exercise had to be confined to the international transactions alone. Mixing domestic transactions with international transactions distorted the profitability analysis and diluted comparability. As no specific reasoning had been given for including domestic items in the operating margin computation, the entity-level approach was found unsustainable for this purpose.

Conclusion: The issue was decided in favour of the assessee, and the operating margin was directed to be recomputed after excluding domestic revenue and domestic expenditure.

Issue (ii): whether the related party transaction filter was required to be applied on an aggregate basis across revenue and expenditure

Analysis: The methodology had been consistently followed in earlier years on an aggregate basis, and no change in facts or law justified a different approach in the year under consideration. In the absence of any distinguishing feature, consistency in the computation of the filter was required.

Conclusion: The issue was decided in favour of the assessee, and the related party transaction filter was directed to be computed on an aggregate basis.

Issue (iii): whether the disputed comparables were liable to be excluded or included on the basis of related party transactions, functional comparability, and database search results

Analysis: Companies failing the correctly applied related party transaction filter could not be retained as comparables. A company with a low related party transaction percentage on correct computation was directed to be included. A company excluded merely because it did not appear in one database could not be rejected if it appeared in another recognised database, subject to functional verification. Certain entities engaged in segments such as accounting, health, e-publishing, or content development were held not comparable to an assessee rendering IT enabled services.

Conclusion: The issue was decided partly in favour of the assessee. Exclusion was directed where the related party filter or functional comparability failed, while inclusion was directed for companies passing the filter or requiring fresh verification through a valid database search.

Issue (iv): whether interest on delayed receivables from associated enterprises was to be benchmarked by applying LIBOR plus 200 basis points

Analysis: For receivables from foreign associated enterprises, the internationally recognised rate was held to be the proper benchmark. The domestic banking rate applied by the transfer pricing authorities was not accepted. The request to compute interest on a weighted average credit period had already been rejected in an earlier year and was not accepted again.

Conclusion: The issue was decided in favour of the assessee on the benchmark rate, and interest was directed to be recomputed by applying LIBOR plus 200 basis points.

Issue (v): whether interest on compulsorily convertible debentures denominated in Indian currency was to be benchmarked by applying SBI Prime Lending Rate instead of SIBOR

Analysis: Where the debentures were denominated in Indian currency, they were to be benchmarked with reference to domestic lending rates. Foreign interbank rates were not appropriate for such rupee-denominated instruments, and the domestic prime lending rate was the correct benchmark.

Conclusion: The issue was decided in favour of the assessee, and SBI Prime Lending Rate was directed to be applied instead of SIBOR.

Final Conclusion: The assessment was interfered with on several transfer pricing issues, resulting in recomputation of the operating margin, the related party filter, the comparable set, and the benchmarks for receivables and CCD interest, while the appeal succeeded only to that extent.

Ratio Decidendi: In transfer pricing, international transactions must be benchmarked independently of domestic transactions, comparable selection must follow a consistent and correctly computed filter, foreign receivables from associated enterprises may be benchmarked by LIBOR-based rates, and rupee-denominated debentures are to be benchmarked by domestic lending rates.

---