Important law laid down on the principles for identifying comparables for benchmarking an international transaction and determining the ALP in the context of whether KPO services are comparable to BPO services. Law also laid down on whether for TNMM method, broad functionality is sufficient and whether supernormal profits indicate that there is functional dissimilarity.
Aug 26, 2015
The High Court had to consider the principal question whether a Knowledge Process Outsourcing Services (KPO Services) provider could be considered as a comparable for benchmarking international transactions entered into by an entity rendering voice call services – such as the Assessee –with its associated enterprise by using TNMM and taking operating profit margin as the PLI .
The Tribunal held that the activities of Vishal and eClerx, entities engaged in KPO Services such as data processing and analytics services were functionally similar to those of assessee. The Tribunal concluded that voice call services and KPO services were essentially ITeS and, therefore, entities rendering the aforesaid services could be considered as comparables for the purpose of benchmarking international transactions by using TNMM. The Tribunal held that further sub-division of ITeS was not permissible. The Tribunal followed its decision in Willis Processing Services (I) (P.) Ltd. v. Dy. CIT 30 ITR (Trib)129 (Mumbai) 2014. On appeal by the assessee to the High Court HELD reversing the ITAT:
(i) It is not disputed that voice call services are considered to be the lower-end of ITeS. KPO on the other hand are ITeS where the service providers have to employ advanced level of skills and knowledge. Notification No. SO2810(E) dated 18th September 2013 issued by the CBDT notifying Safe Harbour Rules also indicates the above. Rule 10TA(g) of the said Rules defines KPO Services. Whilst Voice Call Center represents the lower-end of ITeS, KPO represents services involving a higher level of skills and knowledge. India has vast human resources and a large number of highly-skilled technical professionals. The expression “KPO” indicates the involvement of domain knowledge in providing ITeS. Typically, KPO includes involvement of advance skills; the services provided may include analytical services, market research, legal research, engineering and design services, intellectual management etc. On the other hand, Voice Call Centers are normally involved in customer support and processing of routine data. In the case of Maersk Global Centers (India) Pvt. Ltd. v. ACIT (supra) a Special Bench of the Tribunal had referred to a report prepared by National Skill Development Corporation (NSDC) on Human Resource and Skill Requirements in IT and ITES Sector (2022) and noted that the KPO sector has been described as “a value play”. The said report also indicates that KPO services are likely to span activities such as “patent advisory, high-end research and analytics, online market research and legal advisory”.
(ii) A Knowledge Process is understood as a high value added process chain wherein the processes are dependent on advanced skills, domain knowledge and the experience of the persons carrying on such processes. KPO services are understood as the higher-end of ITeS in terms of value addition.
(iii) While entities rendering Voice Call Center services for customer support and a KPO service provider may be employing IT-based delivery systems, the characteristics of services, the functional aspects, business environment, risks and the quality of human resource employed would be materially different. It plainly follows that benchmarking international transactions on the basis of comparing the PLI of high-end KPO service providers with the PLI of Voice Call Centers would be unreliable and possibly flawed.
(iv) In order to determine the ALP in relation to a controlled transaction, the analysis must include comparables which are similar in all aspects that have a material bearing on their profitability. Paragraph 1.36 of the “OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations” published in 2010 (hereafter ‘OECD Guidelines’) indicates the “comparability factors” which are important while considering the comparability of uncontrolled transactions/entities with the controlled transactions/entities. Sub-rule (2) of rule 10B of the Income Tax Rules, 1962 also mandates that the comparability of international transactions with uncontrolled transactions would be judged with reference to the factors indicated under clauses (a) to (d) of that sub-rule, which are similar to the comparability factors as indicated under the OECD Guidelines. These include characteristics of property or services transferred and functions performed.
(v) The Tribunal’s view that “once a service falls under the category of ITeS, then there is no sub-classification of segment” is difficult to accept as it is contrary to the fundamental rationale of determining ALP by comparing controlled transactions/entities with similar uncontrolled transactions/entities. ITeS encompasses a wide spectrum of services that use Information Technology based delivery. Such services could include rendering highly technical services by qualified technical personnel, involving advanced skills and knowledge, such as engineering, design and support. While, on the other end of the spectrum ITeS would also include voice-based call centers that render routine customer support for their clients. Clearly, characteristics of the service rendered would be dissimilar. Further, both service providers cannot be considered to be functionally similar. Their business environment would be entirely different, the demand and supply for the services would be different, the assets and capital employed would differ, the competence required to operate the two services would be different. Each of the aforesaid factors would have a material bearing on the profitability of the two entities. Treating the said entities to be comparables only for the reason that they use Information Technology for the delivery of their services, would be erroneous.
(vi) Whilst the Tribunal in Willis Processing Services (India) Pvt. Ltd. v. DCIT (supra) held that no distinction could be made between KPO and BPO service providers, however, a contrary view had been taken by several benches of the Tribunal in other cases. In Capital IQ Information System India (P.) Ltd. v. Dy. CIT, (IT)  32 taxmann.com 21 and Lloyds TSB Global Services Pvt. Ltd. v. DCIT, (ITA No. 5928/Mum/2012 dated 21th November 2012), the Hyderabad and Mumbai Bench of the Tribunal respectively accepted the view that a BPO service provider could not be compared with a KPO service provider.
(vii) The Special Bench of the Tribunal in Maersk Global Centers (India) Pvt. Ltd. struck a different cord. The Special Bench of the Tribunal held that even though there appears to be a difference between BPO and KPO Services, the line of difference is very thin. The Tribunal was of the view that there could be a significant overlap in their activities and it may be difficult to classify services strictly as falling under the category of either a BPO or a KPO. The Tribunal also observed that one of the key success factors of the BPO Industry is its ability to move up the value chain through KPO service offering. For the aforesaid reasons, the Special Bench of the Tribunal held that ITeS Services could not be bifurcated as BPO and KPO Services for the purpose of comparability analysis in the first instance. The Tribunal proceeded to hold that a relatively equal degree of comparability can be achieved by selecting potential comparables on a broad functional analysis at ITeS level and that the comparables so selected could be put to further test by comparing specific functions performed in the international transactions with uncontrolled transactions to attain relatively equal degree
(viii) We have reservations as to the Tribunal’s aforesaid view in Maersk Global Centers (India) Pvt. Ltd. (supra). As indicated above, the expression ‘BPO’ and ‘KPO’ are, plainly, understood in the sense that whereas, BPO does not necessarily involve advanced skills and knowledge; KPO, on the other hand, would involve employment of advanced skills and knowledge for providing services. Thus, the expression ‘KPO’ in common parlance is used to indicate an ITeS provider providing a completely different nature of service than any other BPO service provider. A KPO service provider would also be functionally different from other BPO service providers, inasmuch as the responsibilities undertaken, the activities performed, the quality of resources employed would be materially different. In the circumstances, we are unable to agree that broadly ITeS sector can be used for selecting comparables without making a conscious selection as to the quality and nature of the content of services. Rule 10B(2)(a) of the Income Tax Rules, 1962 mandates that the comparability of controlled and uncontrolled transactions be judged with reference to service/product characteristics. This factor cannot be undermined by using a broad classification of ITeS which takes within its fold various types of services with completely different content and value. Thus, where the tested party is not a KPO service provider, an entity rendering KPO services cannot be considered as a comparable for the purposes of Transfer Pricing analysis. The perception that a BPO service provider may have the ability to move up the value chain by offering KPO services cannot be a ground for assessing the transactions relating to services rendered by the BPO service provider by benchmarking it with the transactions of KPO services providers. The object is to ascertain the ALP of the service rendered and not of a service (higher in value chain) that may possibly be rendered subsequently.
(ix) The transfer pricing analysis must serve the broad object of benchmarking an international transaction for determining an ALP. The methodology necessitates that the comparables must be similar in material aspects. The comparability must be judged on factors such as product/service characteristics, functions undertaken, assets used, risks assumed. This is essential to ensure the efficacy of the exercise. There is sufficient flexibility available within the statutory framework to ensure a fair ALP. Applying the aforesaid principles to the facts of the present case, it is clear that both Vishal and eClerx could not be taken as comparables for determining the ALP.
(x) The Assessee had also sought the exclusion of eClerx and Vishal on the ground that both the companies had returned supernormal profits. Whereas the operating margins (operating margin over total cost) in case of Vishal and eClerx were 50.68% and 65.88% respectively, the PLIs of all other comparables were in the range of 2.2% to 24%. In our view, it would not be apposite to exclude comparables only for the reason that their profits are high, as the same is not provided for in the statutory framework. The OECD Guidelines suggest that a quartile method be adopted which excludes entities that fall in the extreme quartiles for comparability. However, neither Chapter X of the Act nor the Rules made by CBDT provide for exclusion for such statistical reason.
(xi) Having stated the same, it may be necessary to bear in mind that supernormal profits may in certain cases indicate a functional dissimilarity or dissimilarity with respect to a feature that has a material bearing on the profitability. In such circumstances, it would be necessary to undertake further analysis to eliminate the possibility of the high profits resulting on account of any material dissimilarity between the tested party and the chosen comparable. A wide deviation in the PLI amongst selected comparables could be indicative that the comparables selected are either materially dissimilar or the data used is not reliable. The Tribunal proceeded on the basis that an adjustment could be made only in cases where supernormal profits resulted from the factors indicated in Rule 10B of the Income Tax Rules, 1962. In our view, the factors mentioned in Rule 10B are not exhaustive. The principal object of benchmarking international transactions against uncontrolled transactions is to impute an ALP to those transactions. This exercise would fail if a factor, which has a material bearing on the value or the profitability, as the case may be, depending on the method used, is ignored.
(xii) The Tribunal proceeded on the basis that while applying TNMM method, broad functionality is sufficient and it is not necessary that further effort be taken to find a comparable entity rendering services of similar characteristics as the tested entity. The DRP held that TNMM allows flexibility and tolerance in selection of comparables, as functional dissimilarities are subsumed at net margin levels, as compared to Resale Price Method or Comparable Uncontrolled Price Method and, therefore, the functional dissimilarities pointed out by the Assessee did not warrant rejection of eClerx and Vishal as comparables. In our view, the aforesaid approach would not be apposite. Insofar as identifying comparable transactions/entities is concerned, the same would not differ irrespective of the transfer pricing method adopted. In other words, the comparable transactions/entities must be selected on the basis of similarity with the controlled transaction/entity. Comparability of controlled and uncontrolled transactions has to be judged, inter alia, with reference to comparability factors as indicated under rule 10B(2) of the Income Tax Rules, 1962. Comparability analysis by TNMM method may be less sensitive to certain dissimilarities between the tested party and the comparables. However, that cannot be the consideration for diluting the standards of selecting comparable transactions/entities. A higher product and functional similarity would strengthen the efficacy of the method in ascertaining a reliable ALP. Therefore, as far as possible, the comparables must be selected keeping in view the comparability factors as specified. Wide deviations in PLI must trigger further investigations/analysis.
(xiii) Consideration for a transaction would reflect the functions performed, the significant activities undertaken, the assets or resources used/consumed, the risks assumed. Thus, comparison of activities undertaken/functions performed is important for determining the comparability between controlled and uncontrolled transactions/entity. It would not be apposite to ignore functional dissimilarity only for the reason that its impact may be reduced on account of using arithmetical mean of the PLI. The DRP had noted that eClerx was functionally dissimilar, but ignored the same relying on an assumption that the functional dissimilarity would be subsumed in the profit margin. As noted, the content of services provided by the Assessee and the entities in question were not similar. In addition, there were also functional dissimilarities between the Assessee and the two entities in question. In our view, these comparability factors could not be ignored by the Tribunal. While using TNMM, the search for comparables may be broadened by including comparables offering services/products which are not entirely similar to the controlled transaction/entity. However, this can be done only if (a) the functions performed by the tested party and the selected comparable entity are similar including the assets used and the risks assumed; and (b) the difference in services/products offered has no material bearing on the profitability.