NOKIA NET WORKS NOKIA GROUP FINLAND vs. JCIT, NEW DELHI.,

We, the undersigned company, hereby guarantee in your favour the due and timely discharge and performance, in accordance with the said services contract, of all the obligations and liabilities of Nokia India arising from and pursuant to the said services contract. This guarantee constitutes an independent and legally valid undertaking in your favour. We represent to you that it is duly approved by the Directors, Nokia Telecommunications OY 3. The Assessing Officer was of the view that the above arrangement shows that the Indian subsidiary of the assessee company is a permanent establishment of the assessee company. His reasoning, as set out at page 21 of the assessment order for the assessment year 1997-98, was as follows: (a) The Indian subsidiary company is a dependent agent of the assessee company because (a) even though the Indian subsidiary company has concluded contracts with various cellular companies for installation of equipment and services, the entire responsibility rests with the assessee company; (b) the assessee company has given written guarantee to the effect that it will be responsible for proper discharge of obligations by the Indian subsidiary; and (c) the assessee company has even given undertaking that, except with the prior written approval of the vendor with which the Indian subsidiary has entered the contract, the assessee company will not dilute its shareholding in the Indian company below 51%. The Assessing Officer thus pointed out that the assessee company was in full control and the Indian subsidiary was no more than an agent of the assessee company. (b) The contract entered into by the Indian companies with the end customer specifically provided, such as in Article 19 of the contract, that “any notice sent by the cellular operator to Nokia India must be sent to Nokia Telecommunications OY also”. The AO was of the view that such an arrangement was a clear evidence of the fact that there was close business connection between the assessee company and its Indian subsidiary. The suggestion apparently was that the business connection was so close that the Indian company was no more than an agent of the assessee company for all practical purposes. (c) The assessee company has provided a twelve month warranty to the purchaser of equipment and during this warranty period, the Indian subsidiary is providing services to the purchaser of equipment. The AO was of the view that clearly all the responsibilities under the warranty clause of the agreement between the assessee and the customer have been discharged by the assessee company as is evident from the following provision in contract with Tata Cellular: The purchaser shall carry out the necessary work to identify and locate the defect. The supplier agrees to provide the technical assistance reasonably required for such fact finding work upon request or the supplier will procure that an affiliate of the supplier will provide such assistance to the purchaser and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 4 of 41 under such terms and conditions as my be mutually agreed between the said affiliate and the purchaser under a technical support agreement. Infact an attempt was thus made to demonstrate that the technical support arrangement was an integral part of the arrangement between the assessee company and the end buyer and that it could not be considered in isolation with assessee’s business. (d) It was also noted that “expatriate employees of the Indian company (Nokia India), who were responsible for installation work etc, were employees of the assessee company or its associated enterprises” The Assessing Officer was of the view that “there was no reason for the assessee company to incur costs on behalf of the Indian company”. It was noted that “the Indian company has not been paid any compensation by the foreign company during the year”. He thus concluded that, for this reason, it is clear that the Indian subsidiary is acting as a dependent agent of the assessee company. (e) The Assessing Officer then noted that the Indian subsidiary was “economically dependent” on the assessee company, as “all its shares are held by Nokia Telecommunications OY and all its receipts are from contracts executed by it for supplies made by Nokia Telecommunications OY” and as “there is total control over the management and affairs of Nokia Telecommunications Pvt Ltd since its is a 100% subsidiary and further the foreign company is giving guarantee on behalf of Indian that they (the assessee company) will see to it that contracts are properly executed”

Having so analysed the facts of the case, the Assessing Officer proceeded to form his opinion about these arrangements, at page 25 of the assessment order for the assessment year 1997-98, as follows:

The Assessing Officer once again resumed his analysis of facts and highlighted the fact that it is only upon the issuance of certificate by the Indian subsidiary, that the equipment is properly installed and is properly functioning, that sale gets concluded. He was thus of the and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 5 of 41 view that the work of the Indian subsidiary was in conjunction with the work of the assessee company, and for this reason, the Indian subsidiary should be treated as a PE of the assessee company. While he did refer to the expression ‘Dependent Agent Permanent Establishment’ most of the time, the reference was for the Indian subsidiary being treated as PE of the assessee company on account of closely related commercial activity. It was then observed that, in any case, the assessee also had a fixed place of business in India. It was noted that “the local office, which was opened in March 1994, was in fact providing a fixed place of business for assessee in marketing its products”, that “infact, this office provided an interface between the customer and the assessee company” and that “this office was collocated with Indian company and all the expenditures may have been shifted to the Indian company, whereas the activity (of the assessee company) was carried through this office”. It was also noted that “before a contract was signed, a number of expatriates would came to India, stayed in India and carried out network planning” and “they were also involved in negotiating the deal with various customers and were interacting with them on regular basis” which would not have “been possible without the assessee having a fixed place of business from which it carried out these operations” and “this fixed place of business was the liaison office together with Indian company”. It was also noted that the visiting staff of the assessee company were given all the administrative support at this location, that “the Indian company was maintaining stocks of spare parts for replacement under the warranty period” and that “the Indian company provided the helpline facility for any type of problems accruing to the customer”. He then referred to certain dates and the signatories of the contracts, an aspect on which facts have been set out fairly elaborately in the lead order, and I, therefore, need not really supplement the same. Suffice to say that conclusion of the Assessing Officer was that the assessee company had a PE in India, and then, after taking note of the fact that gross profit of the assessee company was 40.87% in 1997 and 39.4% in 1996, the Assessing Officer adopted 40% as the gross profit earned by the assessee for the assessment year 1997-98. The net sales, converted into INRs, being Rs 146,61,61,361, the gross profit was worked out at Rs 41,05,25,180 by assuming that hardware sale was 70% of total sales. Allowing a deduction of 5% under section 44C and noting that there was no claim of expenses incurred in India operations, the Assessing Officer computed the profits attributable to the PE at Rs 38,99,98,921 for the assessment year 1997-98. Using the same methodology, but with total sale figure of Rs 84,51,41,736 for the assessment year 1998-99, the taxable profit attributable to the PE was worked out at Rs 10,97,07,697. Aggrieved by the stand so taken by the Assessing Officer, assessee carried the matter in appeal before the CIT(A) but without complete success. While the CIT(A) did reduce the profit attributable to the Indian PE to 5% of total sales, he confirmed the conclusion of the Assessing Officer with respect to existence of a PE. We may, in this regard, refer to the following portion of his operative order for the assessment year 1997-98 which was also adopted, by a separate order, for the assessment year 1998-99 as well:

Coming to the issue of the computation of income on sale of hardware, the ld. AO observed that the gross profit of the assessee was 28.7% while the operating profit was 10.8% in the year ended on 31.12.96. However, thereafter, without assigning any reason he estimated the profit margin on sale of hardware at 10% of the total sale. The view of the ld. Counsels was that the Indian and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 7 of 41 Telecom Industry had been passing through a bad phase in that period and Indian operators were incurring heavy losses. The assessee was a new entrant in this highly competitive market and it had to make significant initial investment for gaining market share. Copies of P&L a/c for Indian tax purposes for financial years 1996-97 and 1997-98 were filed in the paper book as item 6 of section 3. This showed a loss of U.S. dollars 2,37,52,669 for the year ended 31.3.97. The sales were shown at US Dollars 1,57,51,391. On the face of it, there appears to be something wrong with the P&L a/c as direct costs were shown at U.S. Dollar 2,10,24,054. This is in the context worldwide gross profit of 28.7%. This P&L a/c was not substantiated with any documents. Therefore, it is held that these accounts are not reliable for the purpose of computing income from sale of hardware. Accordingly, assistance of Rule 10 of the I.T. Rules is taken to compute profit on the basis of global accounts. The global accounts showed net profit of 10.8% as mentioned by the Ld. AO in the assessment order. Therefore, the net profit is taken at 10.8%. The whole of this profit cannot be attributed to Indian operations as activities regarding manufacture and development of products etc. was undertaken outside India. Therefore, the profits attributable to operations in India are taken at 5% of the sales to the Indian Parties.

Aggrieved by the stand so taken by the CIT(A), assessee carried the matter in appeal before this Tribunal, and, in the first round of proceedings, a coordinate special bench of the Tribunal concluded, in a rather brief operative portion of the order on this point, as follows:

Their Lordships noted the contention of the assessee that the Special Bench “did not go into the issue of how much income can be attributed to the activities carried out in India because that analysis was only made in respect of the subsidiary constituting a PE”, that “it was necessary to ascertain as to whether any income was attributable to the PE”, and that “no such income could be attributed to PE in India and these aspects were not correctly appreciated by the Tribunal”. It was also contended before Their Lordships that “various factual errors which has crept in the orders of the lower authorities”. Hon’ble High Court, in this background, observed, inter alia, as follows: “Mr. Parasaran, learned ASG appearing for the Revenue could not controvert the aforesaid pleas……. We find that the aforesaid errors on facts have crept in. It is primarily for the reason that the Tribunal had taken the facts in the case of Ericsson case and on the presumption that those facts were common the case of Nokia as well and the legal questions in the appeals of Nokia were decided therefore the actual inaccuracy has crept in the fact findings of the Tribunal………..We would like to record that the CIT(A) proceeded on the basis that Indian subsidiary incurred huge loss and the parent assessee was aware of its profitability. The CIT(A) also observed that since NPL was 100% subsidiary and the assessee had wide experience in this area of business, it is logical that a transaction between the assessee and the Indian subsidiary did not occur at arm's length. Mr. Syali argued that there was no basis for drawing such inference and at the time of arguments, the learned ASG conceded that there was no evidence to support that losses were absorbed by the Indian company. Again, pertinently, the Tribunal also observed that NIPL could be considered PE of assessee in India being subsidiary as it is the virtual projection of the company in India. Further, the accounts of the Indian subsidiary show that the company incurred huge losses as it was not compensated properly for the installation work carried on by it. In the opinion of the ITAT since it was a wholly owned subsidiary, the assessee would have direct and complete control over the activities of this subsidiary. The learned ASG also conceded that it was not correct”.

It was in this backdrop of these factual errors, as noted by Their Lordships, that Hon’ble Delhi High Court referred the matter back to this Tribunal by observing as follows:

Let me, at this stage, make careful note of the factual mistakes, as pointed out by Hon’ble high Court, to have crept in the order of the Special Bench: - The special bench did not at all deal with the quantification aspect of the profits attributable to the PE. [This is an error in the nature of error of omission committed by the Tribunal] - The special bench proceeded on the basis that “since it (the Indian subsidiary) was a wholly owned subsidiary, the assessee would have direct and complete control over the activities of this subsidiary” [This is an error in the nature of error of misconception of facts by the Tribunal] - The special bench “had taken the facts in the case of Ericsson case and on the presumption that those facts were common the case of Nokia as well and the legal questions in the appeals of Nokia were decided therefore the factual inaccuracy has crept in the fact findings of the Tribunal”. [This is an error in the nature of error of misconception of facts by the Tribunal] - The Special bench erred in not appreciating “that the CIT(A) had proceeded on the basis that Indian subsidiary incurred huge loss and the parent assessee was aware of its profitability” and that “the CIT(A) also observed that since NPL was 100% subsidiary and the assessee had wide experience in this area of business, it is logical that a transaction between the assessee and the Indian subsidiary did not occur at arm's length” but it was argued before Their Lordships that “there was no basis for drawing such inference and at the time of arguments, the learned ASG conceded that there was no evidence to support that losses were absorbed by the Indian company” [These are errors in the nature of error of not appreciating incorrectness of the findings of the CIT(A) which, as learned ASG agrees, are not based on any material ] - The conclusion arrived at by the special bench “was erroneous as it was based on various factual errors which has crept in the orders of the lower authorities (emphasis supplied by me now)”……..and “the factual errors of the orders of the AO were specifically pointed out in the submissions to the CIT (A) and specific grounds were also taken before him (emphasis supplied by me now) which are as under:-

As I proceed to deal with this aspect of the matter, I must deal with a preliminary issue raised by the learned counsel. He submits that this question regarding existence of PE is to be determined in the light of certain observations made by the Hon’ble High Court. All the findings in the orders of the authorities below, as indeed whole of the order of the Tribunal in the first round, stand disapproved, and, therefore, unless revenue brings out some new and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 13 of 41 material in support of the contention that the assessee had a PE in India, the assessee company cannot be said to have a PE in India. He, nevertheless, accepts that there is no finding by Hon’ble High Court on the existence of the dependent agency permanent establishment, and that is an aspect, therefore, which may be adjudicated upon by the Tribunal now. Learned counsel’s basic plea thus is that so far as the issue of existence of permanent establishment under the basic rule is concerned, Hon’ble High Court has comprehensively decided the issue in favour of the assessee, and all that can, therefore, be examined is whether or not the assessee company had a DAPE (dependent agent permanent establishment) by way of the Indian subsidiary (i.e. Nokia India Pvt Ltd).

I do not share the perceptions of the learned counsel. Undoubtedly, Hon’ble Delhi High Court has pointed out an error of omission (i.e. not dealing with quantification of PE profit attribution), certain errors of misconception of facts (i.e. mixing up facts of Ericson with the facts of this case, proceeding on the basis that since NIPL was a wholly owned subsidiary, the assessee will have direct and complete control over the subsidiary and proceeding on the basis that there was evidence to support the plea that losses of the Indian subsidiary were absorbed by the assessee company) and the error of reaching erroneous conclusions on the basis of factual errors in the orders of the AO which were duly pointed out to the CIT(A). Yet, none of these errors, either on standalone basis or taken together, cannot lead us to the conclusion that the finding of the earlier Special Bench is to be reversed. The errors so pointed out by Their Lordships do lead to the conclusion that, based on the material on record, findings of the Tribunal could not be sustained, but then there is a difference, and a vital difference at that, in the findings of the Tribunal not being sustained and the findings of the Tribunal being reversed. Take, for example, Their Lordships’ observation that “In the opinion of the ITAT, since it (the Indian subsidiary) was a wholly owned subsidiary, the assessee would have direct and complete control over the activities of this subsidiary”- as assertion which learned ASG conceded to be incorrect. The cause and effect relationship mentioned by Their Lordships is between a company being “wholly owned subsidiary” and the assumption that “the assessee would have complete control over the activities of the subsidiary”. What Their Lordships have observed about incorrectness of this approach is certainly valid on plain first principles, apart from being binding law on the subject, and the same principle is well enshrined in Article 5(8) of the applicable Indo Finnish tax treaty which states that “The fact that a company which is a resident of a Contracting State controls or is controlled by a company which is a resident of the other Contracting State, or which carries on business in that other State (whether through a permanent establishment or otherwise), shall not of itself constitute either company or a permanent establishment of the other”. All that the principle recognized by Their Lordships, which is also well enshrined in Article 5(8), means is that a subsidiary cannot be said to be PE of the foreign company solely on the ground that it is controlled by a company resident of the other contracting state. The wordings employed in the observations of Their Lordships are different but implications are the same, because once it is concluded that the foreign company has direct and complete control over the activities of a subsidiary, essentially that subsidiary ends up being treated as a permanent establishment of the foreign company on the ground of its being a “wholly owned subsidiary”. However, that does not mean that a subsidiary of the assessee company cannot be held to be its PE at all. It is only elementary that there can be, and there are, situations in which a subsidiary can be treated as PE of a company fiscally and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 14 of 41 domiciled in the treaty partner jurisdiction. Clearly, therefore, while there is no bar on the subsidiary of a foreign company being treated as a PE of the parent company, mere existence of the subsidiary of a company resident in the treaty partner country would not imply that such a foreign enterprise has a PE in India. Going even by the OECD Commentary (which has been adopted in the UN Commentary as well), with which I have some issues on this point and I will discuss that in detail a little later, there can be situations in which a subsidiary can be a permanent establishment of the parent company, and even vice versa. The following extracts from the current OECD Commentary will throw light on the same: 115. It is generally accepted that the existence of a subsidiary company does not, of itself, constitute that subsidiary company a permanent establishment of its parent company. This follows from the principle that, for the purpose of taxation, such a subsidiary company constitutes an independent legal entity. Even the fact that the trade or business carried on by the subsidiary company is managed by the parent company does not constitute the subsidiary company a permanent establishment of the parent company.

A parent company may, however, be found, under the rules of paragraph 1 or 5 of the Article, to have a permanent establishment in a State where a subsidiary has a place of business. Thus, any space or premises belonging to the subsidiary that is at the disposal of the parent company (see paragraphs 10 to 19 above) and that constitutes a fixed place of business through which the parent carries on its own business will constitute a permanent establishment of the parent under paragraph 1, subject to paragraphs 3 and 4 of the Article (see for instance, the example in paragraph 15 above). Also, under paragraph 5, a parent will be deemed to have a permanent establishment in a State in respect of any activities that its subsidiary undertakes for it if the conditions of that paragraph are met (see paragraphs 82 to 99 above), unless paragraph 6 of the Article applies.

The same principles apply to any company forming part of a multinational group so that such a company may be found to have a permanent establishment in a State where it has at its disposal (see paragraphs 10 to 19 above) and uses premises belonging to another company of the group, or if the former company is deemed to have a permanent establishment under paragraph 5 of the Article (see paragraphs 82 to 99 above). The determination of the existence of a permanent establishment under the rules of paragraph 1 or 5 of the Article must, however, be done separately for each company of the group. Thus, the existence in one State of a permanent establishment of one company of the group will not have any relevance as to whether another company of the group has itself a permanent establishment in that State.

Whilst premises belonging to a company that is a member of a multinational group can be put at the disposal of another company of the group and may, subject to the other conditions of Article 5, constitute a permanent establishment of that other company if the business of that other company is carried on through that place, it is important to distinguish that case from the frequent situation where a company that is a member of a multinational group provides services (e.g. management services) to another company of the group as part of its own business carried on in premises that are not those of that other company and using its own personnel. In that case, the place where those services are provided is not at the disposal of the latter company and it is not the business of that company that is carried on through that place. That place cannot, therefore, be considered to be a permanent establishment of the company to which the services are provided. Indeed, the fact that a company’s own activities at a given location may provide an economic benefit to the business of another company does not mean that the latter company carries on its business through and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 15 of 41 that location: clearly, a company that merely purchases parts produced or services supplied by another company in a different country would not have a permanent establishment because of that, even though it may benefit from the manufacturing of these parts or the supplying of these services.

It will, therefore, be wholly inappropriate to proceed on the basis that just because Their Lordships have observed that the Special Bench was incorrect in proceeding on the basis that merely because the NIPL was a wholly owned subsidiary, the assessee had direct and complete control over the activities of subsidiary, we have to now proceed on the basis that NIPL cannot be a PE of the assessee company. The direct control over subsidiary simply by the virtue of ownership is just one of the aspects of the matter, but there can be a direct control over subsidiary on account of several other factors as well. Those aspects of the matter, in my humble understanding, are open issues and this Tribunal is duty bound to deal with those aspects of the matter, to the extent clearly discernible from material on record, as well.

Similar is the position with respect to the mistake on account of mixing up the facts of this case with the facts of Ericson’s case and proceeding on the basis that the assessee company absorbed the losses of the Indian subsidiary. All that these mistakes require is that while proceeding to draw our conclusions, we must remain confined to our facts, without mixing up with the facts of Ericson’s case- or, for that purpose, the case of any other assessee, and that while deciding whether or not Nokia India Pvt Ltd is a PE of the assessee company, we must not proceed on the basis that the losses of Nokia India Pvt Ltd are absorbed by the assessee company. That does not, however, mean that Nokia India Pvt Ltd cannot now be treated as PE of the assessee company when it can be so established on its own facts and irrespective of the fact as to whether or not the looses of Nokia India Pvt Ltd were absorbed by the assessee company. On the same lines, while it is indeed held by Hon’ble Delhi High Court that the Special Bench did err in not appreciating that “that the CIT(A) had proceeded on the basis that Indian subsidiary incurred huge loss and the parent assessee was aware of its profitability” and that “the CIT(A) also observed that since NPL was 100% subsidiary and the assessee had wide experience in this area of business, it is logical that a transaction between the assessee and the Indian subsidiary did not occur at arm's length” and it was argued before Their Lordships that “there was no basis for drawing such inference”, this finding cannot come in the way of this Tribunal’s holding that there the Nokia India Pvt Ltd was a PE of the assessee company, or even the transactions not being arm’s length transactions, as long as it can be so held without any influence of the holding- subsidiary relationship and rich experience of the assessee company in its line of business. The situation with regard to the conclusions of the Tribunal being vitiated by incorrectness of findings in the orders of the authorities below is no different either. No doubt, as pointed by Their Lordships, the conclusions arrived by the Tribunal “was erroneous as it was based on various factual errors which has crept in the orders of the lower authorities”……..and “the factual errors of the orders of the AO were specifically pointed out in the submissions to the CIT (A) and specific grounds were also taken before him”, but then that’s not the end of the road for the case of the Revenue. Their Lordships noted that “the order of the Tribunal is based on many factual errors” but then referred the matter back to the Tribunal “for fresh consideration on the issues as to whether the subsidiary of the assessee …….is Permanent Establishment…”. What essentially implies is that the Tribunal has to take a fresh call on this question, and, while doing so, the conclusions of the Tribunal must not be vitiated by the and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 16 of 41 factual mistakes that it had committed in the first round of proceedings. In the light of the discussions above, it is also clear that typically subsidiary companies, by default, do not constitute permanent establishments of the parent company though but then in a situation in which facts of a case warrant or justify such a conclusion, there is no bar in treating an Indian subsidiary as PE of the foreign company in India. That is the reason as to why, in a situation in which the foreign parent company and the Indian subsidiary company are operating in India simultaneously, any finding about the Indian subsidiary constituting a PE of the foreign company essentially requires a careful analysis about operations of both the companies in India.

It is in the light of the above discussions that I have to take my call on the question as to whether the assessee company could be said to have a business connection in India so as to bring it income within deeming fiction of Section 9(1)(i) of the Income Tax Act, 1961, and whether the assessee company can be said to have a PE in India, in the form of its subsidiary company- Nokia India Pvt Ltd.

Let me now turn to the case of the Revenue so far as existence of a business connection and existence of the PE is concerned. The stand of the Assessing Officer, as noted at page 24 of the assessment order for the assessment year 1997-98, is that “the Indian company is economically dependent on the foreign company” inasmuch as “all it’s shares are held by the Nokia Telecommunications OY and (emphasis, by underlining, supplied by me) all its receipts from contracts are executed by it for the supplies made Nokia Telecommunications OY” as also the fact that “there is total control over the management of affairs of Nokia Telecommunications Pvt Ltd since it’s a 100% subsidiary and further (emphasis supplied by me) the foreign company is taking guarantee on behalf of Indian company that they (the foreign company) will see to it that contracts (entered into by the Indian subsidiary company) are properly executed”. The Assessing Officer also noted that awarding of technical support services contracts, in respect of the equipment supplied by the assessee company, involved a specific undertaking to the end customer to the effect that “as long as any part of the commitments under the technical support agreement remain outstanding, we will continuously and diligently monitor business affairs of Nokia India with the aim of ensuring that the company at all times is in a position to meet its commitments to you” and that “as long as any part of technical support agreement are not performed, we shall not dispose of our ownership of Nokia India Pvt Ltd below 51% without your prior written permission”. He was of the view that in the light of this position, “the assessee company has a permanent establishment in India in the form of its Indian subsidiary, which is a dependent agent permanent establishment”. It is on the basis of this reasoning that the Assessing Officer, in the immediately succeeding paragraph, concludes that “the Indian company is nothing but an extension of the foreign company”, that “the Indian company was providing the marketing support and coordination on behalf of the foreign company” and that these factors, along-with other facts of the case, make it clear “that the incorporation of Indian company was a veil to avoid the taxability of the foreign company on profits earned through the supplies made by the assessee company”. Essentially, therefore, unmistakable pointer of the Assessing Officer is that the Indian subsidiary company of the assessee, because of the activities carried out by the subsidiary and the manner in which these activities are carried out, constitutes permanent establishment of the assessee company. Yet, he has stated, clearly unmindful of the correct and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 17 of 41 implications of the expression ‘dependent agent permanent establishment’, “the assessee company has a permanent establishment in India in the form of its Indian subsidiary, which is a dependent agent permanent establishment”. There is not even whisper of a reasoning in support of any of the ingredient of the DAPE under article 5(5). The existence of a DAPE, under Article 5(5), comes into play when a person “has, and habitually exercises in that state ( i.e. India, in this case), an authority to conclude contracts in the name of the enterprise (i.e. the assessee company)” in certain circumstances, or when such a person “has no such authority but habitually maintains in the first mentioned state (i.e. India, in this case) a stock of goods or merchandise from which he regularly delivers goods or merchandise on behalf of that enterprise (i.e. the assessee company)”. No case has been made out for satisfaction of these conditions, and, therefore, it is not even case of the Assessing Officer, in terms of the legal connotations of ‘dependent agent permanent establishment”, that the Indian subsidiary constituted PE of the assessee company. As a matter of fact, given the reasoning adopted by the Assessing Officer and the broad thrust of his case, the natural corollary of this reasoning, when held to be correct, is that the Indian subsidiary is held to be a permanent establishment of the assessee company under the basic rule on interdependence and interplay of activities of the assessee company and its Indian subsidiary. This aspect is even more clear from the order of the CIT(A), whose powers are coterminous with that of the Assessing Officer, which, inter alia, states as follows: The assessee had also a wholly owned subsidiary in India. It has been pointed out earlier that representation was made to the Indian operators that the assessee would to ensure that the installation contract was carried out fully by the IC, and the assessee would fully support the IC in discharge of its obligations under the contract. Not only that it was also represented that the assessee will not dilute its activity below 51% in the IC without the written permission of the Indian operator. In view of these representations and the counter undertakings given by the assessee and the IC, it cannot be said that the IC acted independently in discharge of its obligations under the contracts. ……………….. In the context of these facts, it will be difficult to hold that the assessee and the IC acted independently in so far as their businesses are concerned and it will more appropriate to hold that the IC merely acted at the instructions of the assessee in respect of installation and marketing contracts. …………... We have also seen that the appellant himself has taken up the responsibility on behalf of the subsidiary company and has gone to the extent of holding out that its equity will not be diluted below 51% till installation contract is completed, except with written permission of the Indian operator. This strengthens the view that the assertions of the assessee in the matter are not correct. Accordingly, it is held that the ld. AO was right in holding that the appellant had a PE in India through the office of the IC.

The findings of the CIT(A), which are called into question in appeal before us, adopt and approve the same reasoning as was adopted by the Assessing Officer and come to the conclusion that the assessee company had a PE in India through the office of the Indian subsidiary company as the Indian subsidiary was acting, not because of the ownership of the Indian subsidiary company but because of the nature of, and manner in which, business activities are carried out by the assessee company and the Indian subsidiary company. and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 18 of 41

Verbosity of the assessment order apart, there can hardly be any doubt about underlying thrust of the assessment order. There is no mention about satisfaction of the conditions of the DAPE under article 5(5) and yet the Assessing Officer has referred to the dependent agent at several places, even though, as I have noted earlier in the order, the unmistakable thrust of the assessment order is that the Indian subsidiary, because of the nature of, and the manner in which, business activities of the assessee company and its subsidiary are carried out, the PE is established. Such a reference to the PE does not refer to, and must not be construed as referring to, the existence of PE under the article 5(5) which relates to certain fact situation, with respect authority to conclude contracts, since it was not even the case of the Assessing Officer than such a situation existed on the facts of this case. I cannot be so pedantic in my approach that just because the Assessing Officer has used the expression ‘DAPE’, maybe incorrectly, I confine myself to article 5(5), and decline to deal with natural corollaries of the reasoning adopted by the Assessing Officer. In any event, the order passed by the Assessing Officer stands merged in the CIT(A)’s order, and, in the absence of any specific reference to article 5(5), such a specific reference need be inferred. The unmistakable thrust of the case of the Assessing Officer is that the Indian subsidiary company, on account of the nature of business activities of the Indian subsidiary, and the manner in which these business activities are carried out, constitutes a PE, and we must deal with that. Let us not forget that the point of time when the impugned assessment order was framed was the point of time when Indian economy had just opened its doors to the global businesses and international taxation, as a field of study in India as also as a specialized area of work in the field offices of income tax department, was still in its infancy. It was much later that a separate wing was established for dealing with international taxation matters. Given these ground realities, it would perhaps only be appropriate not to be pedantic and hyper technical in our approach and concentrate on the substance of the findings of the Assessing Officer and natural corollaries thereof.

The first question that I must, however, deal with is whether the assessee can be said to have a business connection in India. 22. The expression ‘business connection’ is not a defined expression under the statute but observations made by Hon’ble Supreme Court’s landmark judgment in the case of CIT Vs R D Agarwal & Co [(1965) 56 ITR 20 (SC)] give ample guidance about its connotations: ……….The expression "business" is defined in the Act as any trade, commerce, manufacture or any adventure or concern in the nature of trade, commerce or manufacture, but the Act contains no definition of the expression "business connection" and its precise connotation is vague and indefinite. The expression "business connection" undoubtedly means something more than "business". A business connection …….involves a relation between a business carried on by a non-resident which yields profits or gains and some activity in the taxable territories which contributes directly or indirectly to the earning of those profits or gains. It predicates an element of continuity between the business of the non- resident and the activity in the taxable territories: a stray or isolated transaction is normally not to be regarded as a business connection. Business connection may take several forms: it may include carrying on a part of the main business or activity incidental to the main business of the non- resident through an agent, or it may merely be a relation between the business of the non-resident and the and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 19 of 41

I have also noticed that though an impression is given that the contracts awarded to the IC by Nokia’s customers are given on independent principal to principal basis but a little probe in the facts of the case would clearly show that these are not independent contracts inasmuch as the assessee company has given significant and decisive representations on the basis of which the contacts are awarded to the IC. While on this aspect of the matter, I may once again refer to the letter dated 15th June 1996 whereby the assessee company has confirmed to Tata Communications Ltd as follows: This is to confirm that we, Nokia Telecommunications OY, a company duly registered and existing under the laws of the Republic of Finland, are fully aware that you have awarded a contract to Nokia Telecommunications Pvt Ltd (Nokia India) for installation, testing and commissioning of GSM Digital Cellular Mobile Telephone Network of Tata Telecommunications Ltd and for the performance of various other services and activities in connection therewith (the service contract). We, the undersigned company, hereby guarantee in your favour the due and timely discharge and performance, in accordance with the said services contract, of all the obligations and liabilities of Nokia India arising from and pursuant to the said services contract. This guarantee constitutes an independent and legally valid undertaking in your favour. We represent to you that it is duly approved by the Directors, Nokia Telecommunications OY 27. The services being rendered by the IC to the Indian customers are on the basis of assurances given by the assessee company and it is on the basis of the services rendered by the IC to the Indian customers that the assessee company is selling its infrastructure projects in India. As I have noted earlier also, the assessee company has also given a specific undertaking to the customers to the effect that “as long as any part of the commitments under the technical support agreement remain outstanding, we will continuously and diligently monitor business affairs of Nokia India with the aim of ensuring that the company at all times is in a position to meet its commitments to you” and that “as long as any part of technical support agreement are not performed, we shall not dispose of our ownership of Nokia India Pvt Ltd below 51% without your prior written permission”. On a realistic note, the role played by these undertakings and arrangements cannot be ignored in the association of the IC with the customers of the assessee company. One cannot be so naïve so as to ignore the role played by the assessee company is ensuring business for its subsidiary and the role played by the subsidiary in furtherance of the business interests of the assessee company. These two entities, even though hypothetically and legally independent of each other, have carried out their respective business activities in tandem with each other. On these facts, the work done by the IC cannot be viewed on a standalone basis. It has to be little more than coincidence that all the India based customers of the assessee company have awarded the erection and after sales service contracts to its Indian subsidiary. As a matter of fact, it is not even in dispute that the assessee company plays the decisive role in deciding as to who should be awarded the erection contract. In the submissions dated 14th February 2000, and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 23 of 41 the assessee has accepted that position by stating, at page 5- paragraph 7(1), that “the clause in agreements (for sale of equipment by the assessee company) merely provides for installation to be done by person approved by the supplier of the equipment (i.e. the assessee company)”. Essentially, therefore, none could have been given the installation contract without the blessings of the assessee company. One also has to bear in mind the fact that control by the assessee company was essence of the approval as evident from the fact that not only the assessee had given performance guarantee about the work performed by the IC but also an undertaking to the end customer to assure him that even ownership control will not be diluted during the currency of agreement. No reasonable person with elementary understanding of trade and commerce cannot be so naïve as to actually buy the theory that the contacts between the IC and the end customers were independent commercial contracts with no involvement of the assessee company. The relationship between the assessee company and its Indian subsidiary is thus clear and vital. The assessee company is able to sell its infrastructure equipment in India as the erection, after sales, roll out services are rendered by its Indian subsidiary and the Indian subsidiary is able to get these contracts on the basis of assurances given by the assessee company not only with respect to the quality of work done by the Indian subsidiary but also on the basis of a specific assurance that the ownership in India subsidiary, during the currency of arrangements of the assessee company’s agreements with the Indian subsidiary, will not be diluted below 51%. All this, at the minimum, shows the interconnection and interdependence of the assessee company and its Indian subsidiary so far as business operations of the assessee company in India are concerned.

In my considered view, therefore, the assessee did indeed have a business connection in India by way of its Indian subsidiary which was acting in a manner which was, at the minimum, as much, even if not more, for the furtherance of the business interests of the assessee company in India as much, if not more, for its own economic and business interests. As I hold so, I must also point out that just because the manner in which the assessee company has acted is in its own interests, even if we hypothetically assume so, it does not cease to be a business connection for its parent company because a business connection, to quote the words of Hon’ble Supreme Court in R D Agarwal’s case (supra), “may merely be a relation between the business of the non-resident and the activity in the taxable territories, which facilitates or assists the carrying on of that business” of the non-resident. That is precisely what the IC, at the minimum, does in the present case. I donot share the perception of the majority that “the marketing activities and installation contract undertaken by NIPL has been on principal to principal basis; and in the case of the former agreement between assessee and NIPL the payment has been made to NIPL on cost plus basis which has not been disturbed; and in the later agreement there….(are) independent contracts with Indian customers which has nothing to do with the assessee”. These contracts, in my opinion, have so much to do with the assessee that these contracts would not have been possible but for the indulgences shown by the assessee and these contracts cannot be viewed as independent contracts between the IC and the Indian customers of the assessee company, in isolation with the assessee company. The sale agreements clearly provide that the installation of equipment is to be done by a person approved by the assessee company, and, as I have noted earlier, it is certainly little more than coincidence that all the installation contracts have been awarded to the IC. As regards the services contracts between the assessee company and the IC, things are no better either. As I have discussed earlier, there is no contemporaneous evidence in support of pre 1997 contract for services, there are conflicting statements made by the assessee, and that, in any case, the contract dated 19th and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 24 of 41 April 1996 cannot be justified as a commercial contract in bonafide interests of the IC in ordinary course of business. As for the contracts between Indian customers and the IC, the assessee company has, on record and by way of specific undertakings, played a crucial role for these contracts being awarded to the IC vitiating the character of these contracts as independent contracts between the IC and the assessee company’s Indian customers which have, to quote the words of the majority, “nothing to do with the assessee”. As for the profit earned by the IC on these contracts having been offered to tax anyway, on which so much of emphasis has been placed by the learned counsel, that aspect of the matter is not relevant at this stage for determining the existence of business connection.

Let me now briefly touch upon Hon’ble jurisdictional High Court’s judgment in the case of Nortel Network India International Vs Director of Income Tax [(2016) 386 ITR 353 (Del)]. Incidentally, this is a case primarily on profit attributable to the business connection rather than existence of business connection itself, and nothing much therefore really turns on this case on the question that I am addressing at present, but let me deal with it nevertheless.

In Nortel’s case, the factual background, as noted by Their Lordships, was this. The assessee before Their Lordship was a US based company (Nortel USA, in short), which was a step down subsidiary of Nortel Network Limited Canada (Nortel Canada, in short). Nortel Canada, through a network of companies based in Luxemburg, the Netherlands and Mauritius, had a subsidiary in India by the name of Nortel Network India Ltd (Nortel India, in short). Nortel Canada also had a liaison office in India. On 8th June 2002, Nortel India negotiated three separate contracts with Reliance India Ltd (RIL, in short) – namely Optical Equipment Contracts, Optical Services Contract and Software Contract, and, on the same date- with Nortel Canada and RIL being parties to the arrangement, assigned the equipment contract to the Nortel USA. Nortel Canada guaranteed the performance of equipment contract by Nortel USA. On these facts, the case of the Assessing Officer was that Nortel USA was a shadow company of Nortel Canada and was inserted as an intermediary only to avoid taxes, and that “in substance, the contracts were performed by Nortel Canada along with its LO and Nortel India, who acted in unison to identify, negotiate, appraise, secure, execute, manufacture, supply, install, commission and provide warranty and after sales service in respect of the Optical Fibre project of Reliance”. Rejecting the financial results filed by the assessee, which were unaudited, the Assessing Officer proceeded to estimate the profits of the assessee company on the basis of book results shown by the Nortel Canada. When matter travelled in appeal before the CIT(A), it was held, as noted by Their Lordships, that “ (a) that the Assessee was assigned the contract for supply of hardware to Reliance Infocom days after its incorporation; (b) this is the only business that appellant had done during the relevant period under consideration; (c) the Assessee did not have any financial or technical capability of its own; (d) the equipment supplied was manufactured by Nortel Canada and Nortel Ireland and shipped directly from Canada/Ireland; (e) that the Assessee had supplied the equipment at approximately half its purchase price, thus, incurring huge trading loss in the transaction”. The CIT(A) held that the transactions were to be viewed as a whole and not merely in the form of the agreement. On the basis of the aforesaid findings, the CIT(A) upheld the conclusion of the AO that the assessee was a paper company incorporated only with a motive to evade income tax liability on the income arising out of the supply contract in India and, therefore, Nortel Canada and the assessee were to be considered as a single entity. The CIT(A) further rejected the assessee's contention that it did not have a business and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 25 of 41 connection in India”. In appeal, a division bench of this Tribunal “rejected the assessee's contention that the sale of equipment was completed overseas and the installation was done under a separate contract” and held that "the assessee through Nortel India and LO approached the customer, negotiated the contract, bagged the contract, supplied equipment, installed the same, undertook acceptance test after which the system was accepted” and that “The equipment remained in the virtual possession of Nortel Group till such time the equipment is set up and acceptance test is done." The income in the hands of Nortel USA was thus attributed on the basis of profitability of the Nortel Canada.

It was in this backdrop that Hon’ble Delhi High Court held that Nortel USA did not have PE in India and the profit on sale of equipment by Nortel USA to RIL could not be brought to tax in India. Interestingly, however, it was not even the case of the income tax authorities that Nortel India was acting as an extension of Nortel USA who was assessee in this case. That fact takes it out of comparability with the present case. The guarantee for performance of Indian subsidiary was given by Nortel Canada and not the assessee in this case which was Nortel USA. There could not have been, therefore, any consideration attributable to Nortel USA for the services rendered in India by Nortel India. It was on these facts that Hon’ble High Court observed that “if it is accepted that the assessee (i.e. Nortel USA) has received only the consideration for equipment manufactured and delivered overseas, it would be difficult to uphold the view that any part of assessee’s income is chargeable to tax under the Act as no portion of the said income could be attributed to operations in India”. There can be no quarrel with this proposition, not only in law as this is the binding law for us, but even on the first principles because this is exactly what the unambiguous scheme of the Act is. In sharp contrast with this case before Their Lordships, in which performance guarantee was given by an entity other than the assessee before Their Lordship, the performance guarantee for Indian subsidiary, and commitment not to dilute ownership and control in the Indian subsidiary, is given by the assessee company itself and no separate consideration is received by the assessee for the risk inherent in such a performance guarantee and commitments. Essentially, therefore, reward for this risk, which is wholly undertaken in respect of India operations, is prima facie embedded in the sale of equipment by the assessee to the Indian customers. The observations made in Nortel’s case, therefore, have no application in the case before us.

As regards Explanation 2 and 3, which has been dealt with in paragraph 44 of the order- as reproduced below, it is important to bear in mind the fact that since Explanation 2 only explains what is included in the definition of ‘business connection’ and is certain not exhaustive, it would be relevant only when the case being made out for existence of business connection hinges on the scope of Explanation 2 alone, which is not the case here. The relevant observations are, nevertheless, reproduced below: 44. There is little material on record to hold that Nortel India habitually exercises any authority on behalf of the Assessee or Nortel Canada to conclude contracts on their behalf. There is also no material on record which would indicate that Nortel India maintained any stocks of goods or merchandise in India from which goods were regularly delivered on behalf of the Assessee or Nortel Canada. Thus, by virtue of Explanation 2 read with Explanation 3 to Section 9(1)(i) of the Act, no part of Assessee's income could be brought to tax under the Act. It is only when a non-resident Assessee's income is taxable under and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 26 of 41 the Act that the question whether any benefit under the Double Taxation Avoidance Treaty is required to be examined.

It is not, and it cannot be, even the contention of the assessee that when conditions of Explanation 2 to Section 9(1)(i) are not fulfilled, there cannot be a business connection, and, rightly so, since the definition of business connection under this statutory provision is only an inclusive, and not exhaustive definition, to cover the dependent agency situations. The wordings of Explanation 2 are on the same lines as DAPE definition in Article 5(5). Therefore, if someone is to proceed on the basis that since the conditions of the Explanation 2 to Section 9(1)(i) are not fulfilled, there cannot be a business connection at all, it will almost be like saying that when there is no DAPE, there cannot be a PE at all. DAPE is only a particular type of the PE, just as much as Explanation 2 definition, which is parallel to Article 5(5) definition is most tax treaties, is a particular type of the business connection. The observations made by Their Lordships in the above paragraph are in the particular context before Their Lordships in a situation in which the general scope of section 9(1)(i) was held to be inadmissible on the particular facts of the case in the immediately preceding paragraph, and these observations cannot be construed as authority for the proposition that when the condition under Explanation 2 to Section 9(1)(i) are not satisfied in any fact situation, even if the provisions of Section 9(1)(i) are satisfied in general, there cannot be a business connection at all.

In my humble understanding, therefore, Hon’ble jurisdictional High Court’s judgment in the case of Nortel Network (supra) cannot be viewed as an authority for the proposition that income embedded in sale of equipment supplied overseas cannot be taxed under the Act under any circumstances, such as in a situation when a part of the consideration paid for such equipment can be reasonably attributed to the risks undertaken by the assessee in India for which he is not separately or adequately rewarded. The sweeping generalizations, as learned counsel seeks to make by relying upon this precedent to draw a proposition which will hold good de hors the peculiarities of this case, cannot be permitted. I, therefore, decline to be so overawed by some similarities in Nortel Network’s case vis-à-vis the case of this assessee as to, rather than examining this case on its own merits, jump to treat this as a covered matter.

As I take this stand, I am reminded of the oft quoted words of Hon’ble Supreme Court, in the case of Mumbai Kamgar Sabha vs. Abdulbahi Faizullbhai [AIR (1976) SC 1455], wherein their Lordships have, in their inimitable and felicitous words observed thus, "It is trite, going by anglophonic principles that a ruling of a superior Court is binding law. It is not of scriptural sanctity but of ratio-wise luminosity within the edifice of facts where the judicial lamp plays the legal flame. Beyond those walls and de hors the milieu we cannot impart eternal vernal value to the decisions, exalting the precedents into a prison house of bigotry, regardless of the varying circumstances and myriad developments. Realism dictates that a judgment has to be read, subject to the facts directly presented for consideration and not affecting the matters which may lurk in the dark". It is, therefore, indeed duty of every subordinate judicial forum to apply the rulings of Hon’ble Courts above in such a manner so as to enforce the true legal principles emerging from the same, by putting the words and expression used in the ruling in the right perspective and by taking a holistic legal view of the matter. Viewed thus, I am not inclined to treat this as a matter “squarely covered” by the decision of Hon’ble jurisdictional High Court, in the case of Nortel Network (supra), in favour of the assessee.

I am, in the light of the above discussions, of the considered view that the assessee had a business connection in India which brings the assessee within the ambit of taxability in India.

The next question that I must address is whether the assessee, on account of its Indian subsidiary i.e. Nokia India Pvt Ltd, can be said to have a PE in India in terms of the provisions applicable Indo Finnish tax treaty. 38. As I proceed to deal with the question as to whether the Indian subsidiary of the assessee company can be treated as its PE and analyse facts of the case in that light, I must first discuss, in some detail, as to under what circumstances can normally a subsidiary be treated as PE of the parent company. 39. ‘The Law & Practice of Tax Treaties- An Indian Perspective’ by Nilesh Modi (Second Edition, 2014; ISBN-13: 978-81-8473-531-4; at page 447), refers to, what it considers to be, “true test for constitution of a PE” by the subsidiary, in respect of a foreign company, as “whether: • The business of the foreign enterprise is carried out by its local affiliate; or • The local affiliate is the alter ego of the foreign enterprise or, speaks his masters voice only or, is a mere façade; or • The foreign enterprise carries on a business in State S, using the premises or personnel, of its local subsidiary”. 39. The reference to ‘alter ego’ of the foreign enterprise, as is used in Nilesh Modi’s book, is more of a colloquial and factual expression rather than a legal term. This term is now also increasingly used in the tax literature worldwide. 40. The expression alter ego companies seems to be rather appropriate expression for dealing with a particular type of subsidiaries which constitute PEs of the parent foreign enterprise, and that is how I intend to use it in this analysis. It does refer to the situations in which a subsidiary should be treated as a permanent establishment of its non-resident parent company for the reason that the way and manner in which carries out its business activities, it is nothing but an avtar of, a virtual projection of, or an extension of, its non-resident parent company in the country in which the subsidiary is domiciled. 41. As I make this observation, I cannot but be reminded of the inimitable words of Hon’ble Justice Jagannatha Rao in the case of CIT Vs Vishakahapatnam Port Trust Vs CIT [(1983) 144 ITR 146 (AP)] had observed, well ahead of the time, as follows: The words ‘permanent establishment’ postulates existence of a substantial element of an enduring or permanent nature of a foreign enterprise in another country which is attributed to a fixed place of business in that country. It should be of such a nature that it would amount to a virtual projection of the foreign enterprise of one country into the soil of the another country. 39. These words of Hon’ble Justice Rao, who later adorned the Hon’ble Supreme Court, are not only always quoted in international tax literature in many parts of the world, Hon’ble and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 28 of 41 Supreme Court has referred to the same, with approval, in the relatively recent case of Formulae One World Championship Limited Vs CIT [(2017) 394 ITR 80 (SC)]. A coordinate special bench of this Tribunal, in the first round of proceedings in this case, has also referred to the virtual projection concept and upheld the existence of the permanent establishment for that reason alone. While the order so passed by the coordinate bench has been remitted to us for fresh adjudication, it has been done so specifically on account of certain factual mistakes creeping in the order and without disturbing the legal principles laid down therein. It is in this backdrop that I believe that virtual projection of the non-resident enterprise into the soil of the source country does result in creation of a permanent establishment as long as it is of enduring and permanent character and it is attributed to a fixed place of business in the source jurisdiction. If such a virtual projection of the foreign enterprise is by way of a subsidiary, which is nothing but an alter ego company of the non- resident parent company, that too would also result in creation of a permanent establishment.

Such alter ego companies without any significant and independent activities in their own right have always been, by default, treated as permanent establishments of their parent companies. One ruling, rendered by the Authority for Advance Ruling over two decades back, illustrates this point. In the case of ABC In Re (Application No. P- 8) [(1997) 223 ITR 416 (AAR)], the Authority for Advance Ruling, speaking through Justice S Ranganathan- one of the most illustrious former Presidents of this Tribunal, who later adorned high judicial officers including that as a judge of Hon’ble Supreme Court, had observed that it “is of the opinion that the subsidiary will have to be considered to be a permanent establishment of ABC unless it has significant independent activities on its own or on behalf of persons other than ABC and unconnected with it”. It is to be noted that, in this case, there is no reference to the conclusion of contracts, on behalf of the principal, or to the conditions precedent for invoking article 5(5), and yet the dependent agent, in the form of the subsidiary, has been held to be a PE. I am in most respectful and considered agreement with this approach which essentially leads to the conclusion that when a subsidiary company is merely an alter ego, or virtual projection, of its parent company, in the sense that it has no significant activities of its own or on behalf of persons other than the non-resident parent company, it must be treated as a permanent establishment of the non- resident parent company for that reason alone.

While dealing with the alter ego companies, Arvid A Skaar’s book of “Permanent Establishment- Erosion of a Tax Treaty Principle (South Asian Reprint Edition, 2009; ISBN: 978-81-89960-81-0; at page 544) states, inter alia, as follows: Alter Ego companies- extensive business cooperation For a subsidiary to be constitute a PE, the subsidiary is required the business activity of the parent company. This applies under the basic rule as well as under agency clause and the construction clause. In US administrative practice, the use of alter ego companies, i.e. a subsidiary which merely performs the business of the parent in another country, has been considered …………… The IRS stated, that if “the subsidiary is merely an alter ego of the parent….then a permanent establishment would result under the Convention and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 29 of 41

Referring to the work by Arvid Skaar, Klaus Vogel, in his oft quoted book ‘Klaus Vogel on Double Taxation Conventions’ (Indian Reprint Edition, 2005; ISBN-13: 978-81- 89960-62-9; at page 353) recognizes a situation in which a PE is created by the parent company’s assumption of the economic risk of fulfilment of contract by the subsidiary, and, inter alia, as follows: (The situation is, however, different)…..if the parent assumes economic risk of the contract’s fulfilment in relation to main customer. In this situation, the parent company and the subsidiary have in fact established a company of which they are partners. This will lead to permanent establishment of the partners if the general pre-conditions are fulfilled 43. This principle also finds recognition in a rather recent judicial precedent in India. In the case of Ansaldio Energia SPA Vs CIT [(2009) 310 ITR 337 (Mad)], Hon’ble Madras High Court had an occasion to deal with a case in which the assessee, a foreign company, was engaged in the business of selling and setting of power plants, and this foreign company ensured the installation contract being given, by the buyer of power plant i.e. Neyveli Lignite Corporation Ltd (NLC) to its Indian subsidiary, namely Ansaldo Services Pvt Ltd (ASLP). It was in this context that Hon’ble High Court had observed as follows: 15. Let us look at this contract…………... The assessee, and not NLC, selected ASPL to execute Contract Nos. III & IV. Therefore, though NLC (i.e. the Indian customer) entered into Contract Nos. III and IV with ASPL (i.e. the Indian subsidiary) it was only at the instance of the assessee (i.e. the foreign enterprise and the parent company). ASPL was the assessee's subsidiary company. At least as far as this Project was concerned ASPL (i.e. the Indian subsidiary) is virtually the 'assessee's presence' (i.e. virtual presence of the foreign enterprise and the parent company) in India. The assessee controlled and managed ASPL for quality ensuring, maintenance of time schedule, quality control, progress of work etc. 44. It was in this backdrop that Hon’ble Madras High Court upheld the stand of this Tribunal to the effect that the foreign company had a PE in India in the form of its subsidiary company which was assigned, at the instance of the foreign company, the installation work and which constituted “virtually the assessee’s presence”. That is in effect the same thing as an alter ego company. To this extent thus, the concept of alter ego companies is not really alien to Indian tax jurisprudence. 45. Undoubtedly, the question as to whether a subsidiary company is an alter ego company of its parent company or not, in the sense that its presence constitutes “virtually the presence of the assessee” on standalone basis and in view of the activities being carried out in a dependent relationship with the parent company, it is a question of fact which is to be determined on the facts of each case. 46. On the first principles, taxation will infringe neutrality if source taxation of an enterprise is to depend on the form, and not substance, of its business presence in the source jurisdiction- such as rather than doing a thing itself, getting it done through a subsidiary when the essence of arrangement is a performance guarantee by the non-resident parent enterprise and its undertaking to continue to exercise effective control over the subsidiary. and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 30 of 41

A situation in which the non-resident parent company not only exercises control over the company with a view to control its operational functioning but also to ensure an agreed level of performance, and undertakes to continue to have such controlling equity and control for that purpose till the contractual obligations of the subsidiary vis-à-vis its customers are discharged, presumption about independence between parent and subsidiary company, as envisaged in Article 5(7) of the OECD Model Convention which is exactly the same as Article 5(8) of the UN Model Convention- replicated in Article 5(8) of Indo Finnish tax treaty, are nullified by the peculiar circumstances of such arrangements. While Article 5(8) does state that “The fact that a company which is a resident of a Contracting State controls or is controlled by a company which is a resident of the other Contracting State, or which carries on business in that other State (whether through a permanent establishment or otherwise), shall not of itself (Emphasis supplied by me now) constitute either company or a permanent establishment of the other” but that does not mean that a subsidiary of the assessee company cannot be held to be its PE at all. The underlying rationale of this provision is the presumption about independence of the principal and subsidiary in day to day operations and management of their business as separate entities but once this presumption is demolished, the very raison d'etre for exclusion of subsidiaries from being permanent establishments of the overseas parent companies and vice versa ceases to hold good. Explaining this point, Arvid A Skaar, in his book “Permanent Establishment- Erosion of a Tax Treaty Principle (South Asian Reprint Edition, 2009; ISBN: 978-81-89960- 81-0; at page 540), observes as follows: The treaty based protection of related companies recognizes the legal independence of related companies for tax purposes as a material reality until the opposite is proved [OECD Comm. 1977 art. 5 no. 39; cf. Debatin, Systematik IV, in Korn/Debatin, 1 Doppelbeseteuerung ann. 183 (looseleaf)]. This affects both the constitution of PE, and the allocation of income to a separate entity. [Emphasis, by underlining, provided by me] 48. I am in considered agreement with the school of thought that the presumption of independence of the subsidiary holds good only till the contrary is proved, and that is precisely the proposition my analysis in the preceding paragraphs seeks to justify. 49. It may, at this stage, also be noted that while the OECD Model Commentary invariably confine the role of subsidiaries as PE only as a dependent agent- and UN Model Commentary does no better as it simply reproduces and accepts the same, there is no conceptual justification for the same. On this point also, I would like to refer to a very thought provoking observation made by Arvid A Skaar, in his book referred to earlier in this order- at page 541-542, as follows: The conventional position of the OECD based tax treaty doctrine is that a subsidiary PE can only be based on the agency PE clause [OECD Comm. 1977 art. 5 no. 39]. However, the tax treaties aim at allowing the source state to tax business profits with certain economic allegiance to the country, expressed through the enterprise’s PE. This intention must also apply when parent company’s business income is earned by intermediation of a subsidiary. Thus from a de lege ferenda point of view, PE taxation of the parent company is justified in cases where residence state taxation of subsidiary does not adequately and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 31 of 41 attribute taxing jurisdiction to the state. The commentaries to OECD Model treaty donot de lege lata give conclusive reasons for the conventional wisdom with regard to this question.

The OECD approach has not, however, found favour with the Authority for Advance Ruling in the case of ABC In re (supra) when the Authority opined, and I agree with that approach, that it “is of the opinion that the subsidiary will have to be considered to be a permanent establishment of ABC unless it has significant independent activities on its own or on behalf of persons other than ABC and unconnected with it”. If the OECD view was to be followed in entirety and in its literal sense, the subsidiary could not have been a PE unless the conditions set out in Article 5(5) were to be satisfied or if the disposal test was not satisfied so as to being the case within ambit of basic rule PE under article 5(1); that was not, however, the case in the pre transfer pricing legislation era. To this extent, in my view, departure from the convention OECD approach, which is rightly questioned by Arvid A Skaaar and implied rejected by our own Authority for Advance Ruling, is fully justified. Of course, the vital question is whether there is anything to prove that there is “independence of the related companies” and whether “the opposite was proved”. In the post transfer pricing legislation era, the determination of arm’s length price fortifies the “independence of related companies” and nullifies the impact of the intra AE association. That situation is materially different from the situation in pre transfer pricing legislation era, particularly when there is clear prima facie evidence, as in this case- as discussed earlier in this order, that the transactions were not at arm’s length in the sense that the reimbursement mark-up was not even equal to interest factor for the time period involved in incurring the expenditure and reimbursement of the same, and that there were certain risks assumed by a party which were not rewarded at all. The presumption of independence is thus clearly demolished on the facts of this case. The next question then is to what extent OECD Commentaries bind us. I find guidance from the observations of Hon’ble Supreme Court, in the case of CIT Vs PVAL Kulandagan Chettiar [(2004) 267 ITR 654 (SC)], wherein Their Lordships have, inter alia, observed that, “Taxation policy is within the power of the Government and s. 90 of the IT Act enables the Government to formulate its policy through treaties entered into by it and even such Treaty treats the fiscal domicile in one State or the other and thus prevails over the other provisions of the IT Act, it would be unnecessary to refer to the terms addressed in OECD or in any of the decisions of foreign jurisdiction or in any other agreements”. In effect thus, when OECD Commentary is the same as the judicial interpretation, one can refer to same, with approval, but when a fair and judicious interpretation takes you to some other conclusion, the OECD Commentary cannot come in the way. In other words, the settled legal position in India is that the judicial forums are not fettered by the OECD Commentaries. That apart, when Philip Moris decision was given by judicial forum of an OECD Country, i.e. Italy, and the subsequent OECD Commentary amendments sought to nullify the same, even Italy added a reservation in the commentary to the effect that “ As for the subsidiary being a PE under the basic rule, it is only by the virtue of conditions imposed by the OECD Commentary position, and not by the text of the treaty provisions, that the subsidiary cannot ordinarily be a PE under the basic rule since it cannot ordinarily satisfy the disposal test vis-à-vis the foreign parent enterprise unless it business model consists of providing place to the foreign enterprise for functioning in the source country. Even without so providing the place to the foreign enterprise, the subsidiary can nevertheless be an extension or virtual projection of the foreign parent enterprise in the light of manner in which, and predominant object for which, its business activities are carried out, and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 32 of 41 and that is clearly evident from the approach adopted by the AAR in the case of ABC In re (supra). In the light of Hon’ble Supreme Court’s guidance extracted earlier, however, the OECD Commentary, or for that purpose any Commentary, does not bind our judicial interpretation, Therefore, it is not even a question of de legeal ferenda vs de lege lata from the point of view rejecting or accepting OECD commentary on this point; the OECD approach does not anyway provide legal basis for treaty interpretation in India and it is only as persuasive as any other relevant aid to interpretation of the tax treaties. I donot, therefore, concur with the OECD theory, if it can be construed to be that OECD approach permits only subsidiary being treated as a PE only under article 5(5).

In Philip Baker’s book on Double Taxation Conventions, there is an interesting discussion about the nature of the permanent establishments, which is beautifully captured, with approval, in paragraph 24 of Hon’ble Supreme Court’s judgment in the case of Formula One World Championship Ltd (supra) as follows: Emphasising that as a creature of international tax law, the concept of PE has a particularly strong claim to a uniform international meaning, Philip Baker discerns two types of PEs contemplated under Article 5 of OECD Model. First, an establishment which is part of the same enterprise under common ownership and control – an office, branch, etc., to which he gives his own description as an ‘associated permanent establishment’. The second type is an agent, though legally separate from the enterprise, nevertheless who is dependent on the enterprise to the point of forming a PE. Such PE is given the nomenclature of ‘unassociated permanent establishment’ by Baker. He, however, pointed out that there is a possibility of a third type of PE, i.e. a construction or installation site may be regarded as PE under certain circumstances. In the first type of PE, i.e. associated permanent establishments, primary requirement is that there must be a fixed place of business through which the business of an enterprise is wholly or partly carried on. It entails two requirements which need to be fulfilled: (a) there must be a business of an enterprise of a Contracting State (FOWC in the instant case); and (b) PE must be a fixed place of business, i.e. a place which is at the disposal of the enterprise. It is universally accepted that for ascertaining whether there is a fixed place or not, PE must have three characteristics: stability, productivity and dependence. Further, fixed place of business connotes existence of a physical location which is at the disposal of the enterprise through which the business is carried on.

When I read the above observations in the context of the subsidiary company being the PE, the subsidiary company can only fall in the second category of PEs i.e. of “unassociated permanent establishment” as against the first category of PEs which consists of parts of the enterprise i.e. office or branch etc, i.e. of “associated permanent establishment”. In the light of the risk of expression “associated permanent establishments” being mixed up the connotations of “associated enterprises”, and natural corollaries thereof, affecting this analysis at a rather subliminal level, I would rather rephrase these two types of permanent establishments, for the ease of discussions, as “direct permanent establishments” and “indirect permanent establishments”. Now, in the light of the observations made by Hon’ble Supreme Court approving the path taken by Philip Baker, the fixed place of business test and disposal test is relevant only for, what I have termed as, ‘direct permanent establishments’ or and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 33 of 41 what Baker has termed as ‘associated permanent establishments’. In other words it is only in the situations of direct PEs or associated PEs that the fixed place of business and the disposal tests are to be satisfied vis-à-vis the foreign enterprise. These twin tests are not really relevant, vis-à-vis foreign enterprise, for the second category, i.e. unassociated or indirect PEs, which has been described by Hon’ble Supreme Court, taking a clue from Baker’s work, by observing that “the second type (of a PE) is an agent, though legally separate from the (foreign) enterprise, nevertheless who is dependent on the (foreign) enterprise to the point of forming PE”. The only other category of PE visualized is a construction or installation site being regarded as PE under certain circumstances. If a subsidiary, considered to be a permanent establishment on account of, to borrow the expression employed by the AAR, not having “significant independent activities on its own or on behalf of persons other than ..(foreign parent company)…and unconnected with it” is to fit in these three types of PEs, it can only fit in the second category i.e. unassociated permanent establishment or as indirect permanent establishment, and that is the category for which the requirement of fixed place of business and disposal test vis-à-vis the foreign enterprise does not, even going by Hon’ble Supreme Court’s analysis, does not apply. These tests cannot be applied in such cases for the elementary reason that when the work of a foreign enterprise is being carried out by a separate legal entity in the source jurisdiction, there is no question of the foreign enterprise using the place of business in the source jurisdiction or having such a place at its disposal since the work of the foreign enterprise is carried out by a separate legal entity and it is this separate legal entity which must have the fixed place of business at its disposal for performing actions in furtherance of the business interests of the foreign enterprise.

I may, however, add that so far as the second category of PEs in the preceding discussion is concerned, Hon’ble Supreme Court has only referred to “an agent, though legally separate from the foreign enterprise, nevertheless dependent on the foreign enterprise to the point of forming PE”, the conventional OECD approach, which has been referred to by Baker as well, is confining it to Article 5(5)- an approach which, for the detailed reasons recorded earlier in this order and following the approach of the AAR, I have rejected. The OECD approach, which has not found favour even with well known western treaty experts like Arvid A Skaar, a Professor in the University of Oslo, does not, in my considered opinion, merit acceptance.

There is no point in adding much to the discussions, in the lead order, on the basic rule PE that the Indo Finnish tax treaty, like possibly all other tax treaties, provides by stating that “the term ‘permanent establishment’ means a fixed place of business through which the business of an enterprise is wholly or partly carried on”. On the facts of this case, in my considered view, there is no dispute that the business of the foreign enterprise was, at least partly, carried out by the IC by the manner, as I have discussed in detail, IC was carrying on its business- as much, if at all, in the interest of the IC as much, even if not more at the minimum, in the interest of the foreign enterprise. The issue really is with respect to disposal test, i.e. the premises being at the disposal, qua the foreign enterprise, but then, for the detailed reasons I have set out earlier, taking things forward from certain observations in Baker’s work- which is quoted with approval by Hon’ble Supreme Court’s judgment in the case of Formula One (supra), in the case of ‘indirect PEs’ or ‘unassociated PEs’ as a subsidiary PE inherently is, such a disposal test can only be satisfied qua the agent who acts as proxy of the assessee in conducting the business. When business activities are through an agent, representative or proxy, the disposal test cannot be taken up qua the principal. There is and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 34 of 41 no conceptual justification for the confining the role of the subsidiary as a PE under article 5(5) and denying the status of PE under the basic rule, i.e. under article 5(1), when business of the parent foreign enterprise is being carried partly through a fixed place, though at the disposal of such subsidiary- in a capacity as agent, representative or proxy. The theory of ‘disposal test’ vis-à-vis the foreign enterprise remains confined to only the ‘associated PEs’, or ‘direct PEs’ as I put it, and cannot extend to the subsidiary PEs or, for that purpose, any other ‘unassociated PE’ or ‘indirect PE’. To this extent, I am of the view that the approach adopted in the OECD Model Commentary, which is accepted in the UN Model Commentary as well, is not a legally acceptable position. My parallel discussions on this point, at other places in the same order and which are somewhat repetitive in that way, also lead me to the same conclusion 55. Learned counsel suggests that even if a subsidiary company is a virtual projection of the foreign enterprise, and it does business wholly or mainly for its foreign parent company, even then it cannot be treated as a PE of the foreign enterprise under the basic rule. That statement does hold good, as the position now in settled in law is by the binding judicial precedents from Hon’ble jurisdictional High Court, in a situation in which the transactions between the subsidiary and the foreign enterprise are at arm’s length prices, but right now we are dealing with the cases in pre transfer pricing legislation era and in cases where the transactions between the subsidiary and the foreign enterprise are, as evident from the material on record, not on arm’s length basis. That apart, the impact of Hon’ble Supreme Court on the question of applicability of disposal case in the case of indirect PEs, or, as Baker puts it, ‘unassociated PEs’ is to be examined nevertheless which may, to a considerable extent, end up diluting the broad case against the subsidiaries not becoming PEs of the parent foreign enterprise. His argument is that the question of ‘virtual projection’ is irrelevant unless all the tests of fixed place PE, including disposal test, are satisfied. What is canvassed before us is that all the elements of a normal fixed place PE are to be satisfied and, it is only when these tests are satisfied, one has to see whether there is a virtual projection or not. In support of this proposition, a lot of emphasis is laid on the observations made by Hon’ble Supreme Court to the effect that “We are of the opinion that the test laid down by the Andhra Pradesh High Court in Visakhapatnam Port Trust case fully stands satisfied. Not only the Buddha International Circuit is a fixed place where the commercial/economic activity of conducting F-1 Championship was carried out, one could clearly discern that it was a virtual projection of the foreign enterprise, namely, Formula-1 (i.e. FOWC) on the soil of this country”. It is thus contended that, in the esteemed opinion of Hon’ble Supreme Court, in order to be a PE, there has to be a fixed place of business, which satisfied all the tests of the fixed place PE, and it must amount to the virtual projection of the foreign enterprise as well. His suggestions is, which have been accepted by the majority, that “the concept of virtual projection does not mean that even without a fixed place, virtual projection will lead to an inference of PE” and that “the concept of virtual projection cannot be in vacuum dohors any other parameter of the PE”. It is on the basis of this reasoning that he contends that since the conditions precedent for existence of a fixed place PE, i.e. right to disposal, stability and productivity, are not satisfied, there cannot be a PE even if there is a virtual projection of the foreign enterprise by the NIPL.

This line of reasoning proceeds on the fallacious assumption that the concept of “virtual projection” has the same ramifications whether it is in respect of, to use the expression employed by Baker, “associated permanent establishment” (direct PE) and in and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 35 of 41 respect of “unassociated permanent establishment” (indirect PE). No doubt, when the virtual projection theory is applied in terms of “associated PEs” (or, as I would prefer to put it- “direct PEs”), the disposal test must be met, but this position cannot hold good for any form of “unassociated PEs” (or, as I would prefer to put it- “indirect PEs”) by way of, in the words of Hon’ble Supreme Court, “is an agent, though legally separate from the (foreign) enterprise, nevertheless who is dependent on the (foreign) enterprise to the point of forming PE”. The reason is simple. In the case of unassociated PEs, the legally separate enterprise steps into, because of its conduct, the shoes of the foreign enterprise and, in a way, acts as a proxy. In such a situation, the disposal test, vis-à-vis the foreign enterprise, is irrelevant as the separate legal entity, representing the foreign enterprise, performs most of, if not all, the actions in the proxy capacity, and, therefore, the disposal test, if all, must be vis-à-vis such a separate legal entity acting as a proxy. That is the reason that when a business is carried out through the agent (including a subsidiary or any other separate legal entity) in such a manner as it amounts to a virtual projection of the foreign enterprise, and provides a fixed place of business though which business of the foreign enterprise is carried out, the disposal test vis-à- vis the foreign enterprise has no application in determination of a PE. As I have pointed out above, the business is carried out through the agent and not necessarily by the non-resident directly and in entirety. It is important to appreciate that indirect PEs are hypothetical PEs and these are the situations in which, on account of a deeming fiction, the legal independence of the entities is relegated into insignificance by other factors and a legally independent enterprise is treated as hypothetical PE of a foreign enterprise. These situations need not necessarily, and cannot always, satisfy the disposal tests vis-à-vis foreign enterprises.

I may also point out that Hon’ble Supreme Court has, in the case of Formula One (supra), has not stated, as is being projected, that even in the case of virtual projections by subsidiaries, which was not even the case before Their Lordships, virtual projection must also satisfy the disposal test. Quite to the contrary, what Their Lordships have noted is that the case before Their Lordships was a case in which “Not only the Buddh International Circuit is a fixed place where the commercial/economic activity of conducting F-1 Championship was carried out, one could clearly discern that it was a virtual projection of the foreign enterprise, namely, Formula-1 (i.e. FOWC) on the soil of this country (all emphasis supplied by me now)”. While reading this sentence, the importance of expression “not only”, and the preceding discussions on the basis rule PE, cannot be ignored, and, if that is taken into account, logical conclusion is that it was a case in which the conditions precedent for basic rule PE were satisfied, and, in addition, it met the virtual projection test as well. Its almost like saying “whichever way one looks at it, it constitutes a PE nevertheless”. This statement cannot read in the condition of ‘virtual projection’ as a sine qua non for the existence of basic rule PE. It is not, and it cannot be, the case of the assessee that even though all the preconditions of basic rule PE are satisfied, whatever constitutes PE cannot be treated as a basic rule PE because it does not amount to “virtual projection” of the foreign enterprise. To me, this is clearly an incongruous result. In any case, judgments of Hon’ble Supreme Court cannot be read like a statute and cannot be interpreted as answering the questions which did not even fall for the consideration of Their Lordships. If authority is needed even for this elementary proposition, one can usefully refer Hon’ble Supreme Court’s judgment in the case of CIT Vs Sun Engineering Works Pvt Ltd [(1992) 198 ITR 297 (SC)].

As regards the decision of Hon’ble Delhi High Court, in the case of DIT Vs E Funds IT Solutions [(2014) 364 ITR 256 (Del)], which is now approved by Hon’ble Supreme and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 36 of 41 Court in the judgment reported as [(2017) 399 ITR 234 (SC)], that was a case in which Hon’ble jurisdictional High Court had specifically observed that “…… The international transactions between the assessees and e-Fund India and the income of e-Fund India, it is accepted, were made subject-matter of "arm's length pricing" adjudication by the Transfer Pricing Officer (TPO, for short) and the Assessing Officer (AO, for short) in the returns of income filed by e-Fund India. We are not primarily concerned with the merits of the computation of income declared and assessed in the hands of e-Fund India in the present appeals, though the factum that e-Fund India was assessed to tax on its global income as per law or on ‘arms length pricing’ in relation to associated transactions and the basis of the said computation of income earned by e-Fund India, as noticed below, is a relevant and an important fact. Revenue has not disputed the said legal position” (Emphasis supplied by me). The fact of transactions between the parent company and subsidiary company being at arm’s length price was thus not in dispute and, the assessments of income having been made after arm’s length price adjustments, as may have been necessary, the impact of intra AE association stood nullified. Clearly, therefore, the presumption of independence not only was left intact by the Assessing Officer but was further fortified by the transfer pricing assessments of the parties. In such a situation, there cannot indeed be any occasion for assuming dependence in the parent-subsidiary relationship which is the very foundation of a subsidiary being treated as a PE of the parent company. The backdrop and the context in which the observations were made by Their Lordships cannot therefore, by default, hold good in pre transfer pricing legislation era- particularly when there are clear indicators to the position that the transactions between the parent subsidiary were not at arm’s length, that the mark-up on reimbursements was far less than interest compensation for the period of incurring the expenditure and receiving reimbursements for the same and that certain functions and risks assumed by the foreign enterprise were completely unrewarded. In the backdrop of these facts, and when I bear in mind the oft quoted words of Hon’ble Supreme Court, in the case of Sun Engineering Works (supra), that “It is neither desirable nor permissible to pick out a word or a sentence from the judgment of this Court, divorced from the context of the question under consideration and treat it to be the complete "law" declared by this Court”. When Their Lordships themselves observe that application of arm’s length prices by the transfer pricing authorities was “a relevant and an important fact”, it would not really be possible that the observations made in that case will, by default, apply on a fact situation in which not only the transfer pricing provisions were not applicable and thus arm’s length price were not the basis of assessment income but there were clear indicators about the transactions being not an arm’s length basis. In my humble understanding, the question whether the same principles will apply to pre transfer pricing legislation, particularly when these parent-subsidiary transactions are clearly not on arm’s length basis, is not concluded by the said judgment. The other important aspect of the matter is that, after expressing the agreement with the findings of the Hon’ble High Court by stating that “We agree with the findings of the High Court in this regard (that there is no fixed place PE on the facts of the present case), Hon’ble Supreme Court has also observed that in the said case the Indian subsidiary was not an interface of the foreign company with the Indian customers, and in that sense it was not a place of the business of the assessee in India. That is the only specific reasoning given by Hon’ble Supreme Court. Hon’ble Supreme Court has noted that “ ……. no part of the main business and revenue earning activity of the two American companies is carried on through a fixed business place in India……”. Hon’ble Supreme Court upheld the stand of the assessee by expressing observing that “……. and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 37 of 41 It is clear from the above that the Indian company only renders support services which enable the assessees in turn to render services to their clients abroad. This outsourcing of work to India would not give rise to a fixed place PE and the High Court judgment is, therefore, correct on this score”. These observations cannot be taken as approval, as well of all other propositions laid down by Hon’ble jurisdictional High Court, by Hon’ble Supreme Court. There was no occasion for Hon’ble Supreme Court to examine those aspects of the matter. In sharp contrast with this fact situation, right now the case before us is that Indian company was the face of the non-resident parent company in India, where the end customers are situated, and such an association would indeed give rise to the place of business in India. By no stretch of logic, these observations can be seen as subscribing to the theory that, as per law settled by Hon’ble Supreme Court, a subsidiary company can never be the PE of the parent non-resident company- as is being canvassed before us. On the other hand, the observations made in Formula One (supra) seem to support the plea that the disposal test is relevant only for ‘associated PEs’ which are integral part of the non-resident enterprise and are not separate legal entities.

There is one more factor to be borne in mind. The underlying presumption of the path taken by another special bench of equal strength in the first round of proceedings, was that the basic rule PE tests, as canvassed by the commentary, are not necessarily the tests to decide whether there is a “virtual projection” or not, in a situation in which the transactions between the related are not at arm’s length, and when virtual projection can be held de hors these tests, it will be a PE nevertheless. This approach is clear from the fact that the coordinate special bench held the IC to be PE of the assessee company only on the basis of twin factors of transactions not being at arm’s length and virtual projection of the foreign enterprise. Even though this approach was specifically in challenge before Hon’ble Delhi High Court, in question no. 2 i.e. “Whether the Tribunal was right in law in holding that a perception of virtual projection of the foreign enterprise in India results in a permanent establishment?”, Their Lordships have, without expressing any opinion on the same, remitted the matter for reconsideration on factual aspects. As a matter of fact, the question framed by the assessee for the consideration of Hon’ble High Court hardly leaves any doubt about the legal proposition laid down of the earlier special bench, and that legal proposition still holds good. The decision of the earlier special bench on this approach, therefore, continues to be a binding judicial precedent for this special bench as well. The position will of course be difference when a bench of greater strength, or a higher judicial forum, specifically disapproves the path so taken by the earlier special bench. There was no request at all for a reference being made to the larger bench. There is no occasion thus to deviate from the earlier special bench decision on this issue.

Let’s now deal with the mistakes pointed out by Hon’ble high Court, and impact thereof on our conclusions. In the course of the hearing, a specific question was asked as to what were the facts in Ericsson’s case which were, as pleaded by the assessee before Hon’ble High Court, wrongly taken as of the case of this assessee. Learned counsel, however, could not throw any light on this aspect of the matter. Anyway, all that could be done, and that is what has been done, is that only the facts of the present case are being taken into account. I have also ignored the findings, being devoid of any material to support these findings, of the CIT (A) that (a) the Indian subsidiary has incurred huge loss and the parent company was aware of the profitability; (b) the Indian subsidiary being a one hundred percent subsidiary or parent company which having wide experience, the natural inference was that the and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 38 of 41 transactions were not at arm’s length. As regards the long list of factual errors in the order of the Assessing Officer, which were specifically challenged before the CIT(A) but no dealt with by the CIT(A), I keep aside all these findings of the Assessing Officer as well.

Let us now quickly recapitulate facts of the case. Here is an assessee company which has a subsidiary Indian company, by the name of Nokia India Pvt Ltd (IC), and this IC provides marketing and administration support services on cost plus 5% mark-up basis and the mark-up of 5% is less than adequate inasmuch as it does not even factor for the interest element for the period for which its funds are blocked in incurring the expenses. The IC provides, under approval of the assessee company, installation services for which assessee company undertakes performance guarantee and commitment not to dilute the shareholding below 51%, so as to ensure the control over operations and resultant agreed performance level to the IC’s customers, but the assessee company is not at all rewarded for these functions and risks at all. Such an arrangement cannot, therefore, be justified on the commercial considerations at all. It may also be mentioned that the services for which, the IC is so inadequately rewarded that if the IC was to park its funds employed in this activity in a fixed deposit with any Indian bank, it would have earned much more than what it has earned under this agreement, the IC has rendered a number of services, in marketing support function. These services are listed at page 2 and 3 of the agreement dated 19.4.1996, a copy of which is placed before the bench in the paperbook filed by the learned Departmental Representative.

In effect thus the entire marketing and administrative support work is done by the assessee in India, through the Indian subsidiary and without adequate arm’s length consideration, at a fixed place in India. This is carrying on the business of the assessee in India through a fixed place of business. The visiting employees of the assessee company also use the premises of the assessee and carry out important core business functions from the place of the IC. It is important to bear in mind the fact that at no stage, including in the course of the proceedings before us, the assessee has not submitted the details about names and duration of stay of the expatriate employees who availed such support from the IC. The IC renders these important and vital services to the assessee company on a non arm’s length consideration. That aspect of the matter has already been discussed in detail earlier in this order. As for the disposal test vis-à-vis the foreign enterprise, as the assessee company has done the said work inserting a separate legal entity which is working wholly and predominantly for the assessee, on consideration other than arm’s length consideration, it is a case of ‘unassociated PE’ or ‘indirect PE’ which, in the light of detailed discussions earlier in this order, are not required to meet the disposal test vis-à-vis the foreign enterprise. As the Indian subsidiary, a separate legal entity, virtually works as a proxy or virtual projection of the foreign enterprise, the disposal test has to be vis-à-vis the Indian subsidiary, and there is no dispute, nor can there be any dispute on the facts of this case, that the disposal test vis-à- vis the Indian subsidiary is satisfied.

It is also beyond any doubt any controversy that the assessee company was also responsible, without any remuneration or reward, for ensuring certain level of quality of work in respect of the installation work undertaken by the IC. Earlier in this order, this aspect of the matter has been dealt with in considerable length. It is also not in dispute that the assessee has given specific undertaking to the end customers of the IC that, during the currency of the and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 39 of 41 agreements of the end customers with the IC, the assessee company will not dilute its equity ownership below 51%. There is no separate reward or remuneration for this function and risk as well which is again unjustifiable on commercial considerations. This non arm’s length situation also indicates that, as was held in the first round of proceedings, that the distinction between the assessee company and the IC is so blurred that the IC was virtually a projection of the foreign enterprise and the IC is a PE for this reason as well.

An equally undisputed position is that all the installation work generated for the IC is on account of specific approval of the assessee company. The work done by the IC is thus entirely in the control of the assessee company and the work being obtained by the IC is entirely at the mercy of the assessee company. The people at the operational level in the IC also include a number of expatriates on deputation, secondment or assignment from the assessee company. The role of the assessee company was omnipotent in all the operations of the IC, and it was not only because of the ownership of the IC but also because of the business module adopted by the assessee company. There is also no dispute that the installation and other post sale services rendered by the IC were complementary to the core business operations of the assessee company. The IC, in substance and in effect, was acting as a proxy of the assessee company’s interest in performance of commercial activities as well. Viewed thus, the office of the IC is a PE of the foreign enterprise which constitutes the fixed place of business through which the business of the assessee company is wholly or partly carried out. In this case also, since IC is acting in a proxy capacity, and as an agent, the disposal test has to be vis-à-vis the IC, as a proxy or as an agent, and not the foreign enterprise directly.

Quite interestingly, at page 6 in a written submissions filed before us, signed by one Tarandeep Singh, it is stated that “one important aspect which is not considered by the lower authorities was that no adverse inference was drawn in the assessment proceedings of the Indian subsidiary with regard to compensation received on account of installation activities” and that “the learned Assessing Officer of the Indian subsidiary had duly accepted the compensation to be at arm’s length which is evident from the fact that Section 92 was not invoked”. The fact that the assessee accepts application of arm’s length provisions in this case implies that even though the transaction is with unrelated parties, the prices thereof are influenced by the parent company. What this plea, however, overlooks is that the assessment years before us are 1997-98 and 1998-99, whereas Section 92 was brought to the statute vide Finance Act 2001.

It is thus clear that even after taking into account the factual mistakes as pointed in the order of Hon’ble High Court, there is no change in the situation. The reasons are different but the conclusions, on the issue of the permanent establishment, remain the same as were arrived at by the coordinate special bench in the first round of the proceedings. In other words, these mistakes did not affect the ultimate outcome of the appeal on this point. In the light of the detailed discussions above, I am of the considered view that not only that Indian subsidiary of the assessee company provided business connection to the assessee company, the Indian subsidiary of the assessee company also constitutes permanent establishment of the assessee company under article 5(1) of the Indo Finnish tax treaty.

That takes me to the question as whether any profits are to be attributed to the signing, networking, planning and negotiations of the offshore supply contracts in India. These are and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 40 of 41 core marketing functions and core support technical functions which are vital to the business of sale of equipment. At the minimum, these services can be treated at par with marketing services rendered by the assessee, through its PE, in India. As has been discussed earlier, all the crucial marketing and support functions have been rendered by the Indian PE, by way of the IC, and the IC has not been adequately compensated for the same. There are judicial precedents to hold that 35% of the overall profits on sale can be attributed to the marketing function. In the case of Rolls Royce plc Vs DCIT [(2011) 339 ITR 147 (Del)], while upholding the stand of this Tribunal in allocating 35% of global profit to the marketing function, Their Lordships have observed as follows: We are in agreement with the aforesaid view of the Tribunal. While restricting the attribution to 35 per cent, the reason given by the Tribunal was that profit attributed to manufacturing activity and research and development activities, i.e., 50 per cent and 15 per cent respectively had to be excluded. Thus the expenses on research and development were already taken care of when remuneration @ 35 per cent was attributed to marketing activities in India on which global profits was apportioned and there was no question of setting off the research and development expenses again in respect of marketing activities. We, thus, answer question No. 3 against the assessee.

Learned counsel’s submission is that the assessee company has, on the basis of a separate profit and loss account prepared for the equipment sales in India, incurred losses of US Dollars 2,37,52,669 for the year ended 31.3.1997 and US Dollars 86,14,672 for the year ended 31.3.1998. In view of these losses, according to the learned counsel, there cannot be any question of any profits being allocated to the sale of equipment in India.

I have noted that the India specific financials donot make it clear as to on what basis the allocations of expenses have been done. All that these financials show are the broad heads like Direct Costs, R& D Costs, A&G Costs, Sales and Marketing Costs and BoA advisory and financial expenses. As these financials are apparently based on sweeping generalizations and the related evidences have not been produced before any of the authorities below, I would approve the following approach of the CIT(A) in the impugned order: …………On the face of it, there appears to be something wrong with the P&L a/c as direct costs were shown at U.S. Dollar 2,10,24,054. This is in the context worldwide gross profit of 28.7%. This P&L a/c was not substantiated with any documents. Therefore, it is held that these accounts are not reliable for the purpose of computing income from sale of hardware. Accordingly, assistance of Rule 10 of the I.T. Rules is taken to compute profit on the basis of global accounts. The global accounts showed net profit of 10.8% as mentioned by the Ld. AO in the assessment order. Therefore, the net profit is taken at 10.8%. The whole of this profit cannot be attributed to Indian operations as activities regarding manufacture and development of products etc. was undertaken outside India……… and 1964/Del/2001 Assessment years: 1997-98 and 1998-99 Page 41 of 41

The CIT(A) has, out of the 10.8% global profits, allocated 5% of global profits to the Indian operations. While I uphold the approach of the CIT(A) in principle, I also hold that, on the lines of Rolls Royce decision (supra), only 35% of the total profits can be allocated to the services provided by the PE, as marketing function. Accordingly, in my view, 3.78% of sales, i.e. 35% of 10.8% global profit on sales, can be reasonably allocated to the PE. I round it off to 3.75% for the purpose of computing profits attributable to the specified functions of the PE. My conclusions: 71. In view of the above discussions, I hold that the assessee company had a PE in India, by way of the premises and existence of its Indian subsidiary Nokia India Pvt Ltd, and that the profit attributable to the specified operations of this PE are 3.75% of total sales of the equipment in India. In the result, while I uphold the action of the CIT(A) in principle, I marginally reduce the quantum of profits attributable to the PE. As against profit @ 5% of sales held to be attributable to the Indian PE, I hold the profit on 3.75% of sales to be attributable to PE in respect of the specified activities. 72. In the result, in my considered view, the plea of the assessee against the existence of business connection and the existence of permanent establishment is to be rejected, and plea of the assessee on the attribution of profit is to be partly accepted in the terms indicated above. 73. To this extent, even as I humbly bow to the majority so far results of these appeals are concerned, I disassociate myself with the order as finalized by the majority. Save on the above points, I am in considered agreement with the conclusions arrived at in the lead order and I respectfully endorse the same. Pronounced in the open court today on the 5th day of June, 2018.