Serdia Pharmaceuticals (I) Pvt Ltd vs. ACIT (ITAT Mumbai)

TPO entitled to substitute ‘CUP’ for ‘TNMM’ to determine arms’ length price. For generic drugs, CUP is appropriate method despite quality differences

The assessee, a pharmaceutical company, imported ‘Active Pharmaceutical Ingredient’ (‘API’s) from its foreign associated enterprises (‘AE’) and used them for manufacture of drugs. For transfer pricing purposes, the assessee adopted the ‘Transactional Net Margin Method’ (“TNMM”) as the most appropriate method and claimed that its transactions with the AE’s were at arms length on the basis that the assessee’s operating profit at 8.76% on net sales was higher than that of its comparable competitors. However, the TPO held that the assessee had purchased the APIs from the AE’s at prices that were higher than that paid for similar APIs by other companies in India. He rejected the contention of the assessee that the higher prices paid by the assessee were justified owing to their superior quality. The TPO held that the TNMM was not a “reliable” method and that the Comparable Uncontrolled Price (‘CUP’) was, on facts, the most appropriate method and computed the arms’ length price (‘ALP’) on that basis. On appeal, the CIT (A) upheld the stand of the TPO. On further appeal by the assessee, HELD dismissing the appeal:

(i) While innovators of drugs are allowed monopolistic pricing during the period when patents are in force so as to recoup the research and development costs, once the patent period expires, the higher pricing of the drug vis-à-vis prices of generic drugs manufactured by competitors cannot be justified on the ground of heavy R&D costs;

(ii) The argument that as s. 92C does not set out any hierarchy or order of preference amongst the various methods for computing the ALP, the assessee has the unfettered discretion to adopt the TNMM and the TPO is not entitled to reject that method without showing deficiencies/ defects therein is not acceptable. S. 92C r.w. Rule 10C requires the “most appropriate” method to be chosen from amongst those specified. The exercise of selecting the “most appropriate” method implies that the appropriateness of method is to be ranked in some order. Accordingly, it is open to the TPO to reject the TNMM and adopt the CUP method on the basis that the latter is “most appropriate” on the facts of the case;

(iii) Generally, the TNMM is a “method of last resort” and should be adopted only when the standard methods (CUP, Resale Price Method and Cost Plus Method) cannot be reasonably applied. The standard (transaction) methods have an inherent edge over the profit method (TNMM) in most situations and wherever both methods can be applied in an equally reliable manner, the transaction methods should be preferred over TNMM (ACIT vs. MSS India 32 SOT 132 (Mum) followed, OECD Guidelines 2010 considered);

(iv) On merits, the CUP method is the ‘most appropriate method’ to determine the arm’s length price in the cases of generic drug manufacturers so long as comparables are available. As the API imported by the assessee was a generic drug and not patent protected, the CUP method could be used. The argument that the APIs are “unique” on the ground that they are better, of proven effectiveness and manufactured using WHO – GMP practices is not acceptable because while the high quality standards does confer a certain degree of comfort, it does not affect the comparability of the API with the same API manufactured by competitors. (Principles laid down in Glaxo Smith Kline Inc Vs Her Majesty (2008 TCC 324) approved on this point by the Canadian Court of Appeal in 2010 FCA 201 followed – Noted that it was not the argument that the higher prices of API were warranted on account of commercial compulsions arising out of licensing agreement);

(v) The facts also showed that the prices at which the generic drugs were purchased by the assessee from its AEs were not driven by market forces but on considerations which had no role to play in a typical arm’s length transaction. The assessee’s AE had reduced prices of the APIs to compensate the assessee for the low selling price of the drugs which would not have happened in an arms’ length transaction. The price movements and demand sensitivity to the price indicate that the APIs imported by the assessee were are not unique items and that such business models being adopted by pharmaceutical companies leave ample scope for them to manipulate API prices so as to regulate profitability of their controlled entities in the end use jurisdiction;

(vi) The fact that another arm of the Government (customs) considered the price paid by the assessee to be an arm’s length price does not mean the assessee is relieved of the burden of establishing that it is an arm’s length price for transfer pricing purposes;

(vii) Whether the amount paid by the assessee to the AEs in excess of ALP can be re-characterized (in the absence of re-characterization provisions such as s. 247(2) of Canada’s Income-tax Act) left open;

(viii) The technical objection for AY 2002-03 that as the CBDT vide circular dated 20.5.03 directed the AO to make the reference by 30.6.03, the making of the reference on a later date renders it bad in law is not acceptable because the Circular did not impose a bar on the making of references after that date.