In a recent decision (Daimler Chrysler India Pvt. Ltd. v. DCIT), the Pune Bench of the Income Tax Appellate Tribunal elaborately addressed several questions of law related to the application of Double Taxation Avoidance Agreements. This post will highlight the issues involved; while the reasoning and the decision will be looked at in a later post.
The facts:
Under the Indian Income Tax Act, 1961, Section 79 states that where there is any change in shareholding resulting in a change in more than 51% of the voting rights in the company, not being a company in which the public is substantially interested, no losses incurred in a year before the relevant previous year may be carried forward or set off against the income earned in the previous year.
Under Section 2(18) of the Act, a subsidiary of a public company whose shares are listed in a recognized stock exchange in India is treated as a company in which the public is substantially interested. On the other hand, a subsidiary of a public company which is not listed in a recognized stock exchange in India will not be entitled to be treated as a company in which public are substantially interested. Thus, a subsidiary of a company listed on the BSE will be a company in which the public is substantially interested; while that will not be the case with a subsidiary of a company listed on the NYSE.
In the facts of the case before the Tribunal, the assessee was a company incorporated in India. At the beginning of the relevant financial period, 81% of assessee’s share capital was held by a German company, Daimler Benz; and the balance 19% by an Indian company, TELCO. In the relevant financial period, Daimler Benz and the Chrysler Corporation, USA, decided to merge. A new company called Daimler Chrysler was formed in Germany for this purpose. In the process of the merger, all the assets and liabilities of Daimler Benz were transferred to Daimler Chrysler. One of these assets was the shareholding in the assessee company.
In view of these facts, there was a change in shareholding pattern of the assessee company as the shares held by Daimler Benz were transferred to Daimler Chrysler. Daimler Chrysler was not listed on any recognized stock exchange in India; and the assessee was prima facie not a “company in which public are substantially interested”. Therefore, the Revenue sought to apply the provisions of Section 79 to prevent the assessee from carrying forward and setting off its previous losses.
The assessee claimed that the provisions of Section 79 could not be validly invoked in view of the Indo-German DTAA. In particular, the contention of the assessee was that the invocation of Section 79 violated the “non-discrimination” clause in the DTAA. The contention was, effectively, that the assessee was being discriminated against, as other Indian companies were allowed to set off their losses. Notably, it was admitted that there was no double taxation as such.
The Tribunal had to deal with several issues, including the following ones:
In the absence of any double taxation, can the provisions of a DTAA be invoked?
If the provisions of the DTAA do apply, has there been any discrimination against the assessee, resulting in violation of the non-discrimination clause? Should the assessee be compared with any other Indian company (in which case, there would be discrimination as the other company would be treated as a company in which the public were substantially interested, thereby precluding the application of Section 79) or with another Indian company controlled by a foreign company (in which case there would be no discrimination, as Section 79 would continue to apply)?
The decision and the reasoning will be discussed in another post soon.
According to a report in the Economic Times, the Supreme Court has refused to admit Vodafone’s SLP against the order of the Bombay High Court in the Vodafone tax matter. The Court has asked Vodafone to respond to the show cause notice issued by the Revenue. It now appears that the battle will reach the Supreme Court – if at all – only after a long time.
The judgment of the Bombay High Court has been discussed here. While technically decided on the point of maintainability, the High Court did make strong observations on the merits against Vodafone's case. It remains to be seen how far those observations will actually impact the assessment, and the issue is likely to heat up again as Vodafone prepares its reply to the show-cause notice.
The latest digest of Indian income-tax cases is available on the website of the Mumbai ITAT Bar Association. The Digest is available here.
Also (and unrelated to the income tax digest), The Social Blog has now moved to a separate website, ‘Something About the Law’. The Social Blog, started by Aditya Swarup and Arun Mohan [from NALSAR], is one of the leading socio-legal blogs around. Over at the new website, the authors intend to begin deeper coverage of several issues, including guest interviews, podcasts etc. Aditya and Arun say:
“There’s a lot in store for the new platform, right from interview - podcasts to online polls and exclusive reports about socio-legal events. It would be too early to thank our large reader base for inspiring such a decision, because there’s still a long way to go…”
Here’s wishing them all the best with Something About the Law…
Earlier on this blog, I have referred to the concept of the “umbrella clause” in bilateral investment treaties. Another complex issue in BIT arbitration concerns itself with the role of Most-Favoured-Nation clauses (MFN clauses) in dispute settlement under the provisions of the BITs.
MFN clauses typically provide that the host state shall provide treatment no less favourable to investors from the signatory state than to investors from third states. Thus, if countries A and B sign a BIT, an MFN clause in the BIT may provide that state A shall accord investors from state B treatment no less favourable than that accorded to investors from third states.
What exactly is the scope of protection under an MFN clause? In this context, it is generally accepted that MFN clauses are to be construed according to the ejusdem generis principle. Under this rule, an MFN clause can only attract matters as to which the clause relates. For instance, if an MFN clause says that the “no less favourable” treatment is to be granted particularly with respect to the procurement of government contract, it would not be proper to argue purely based on the MFN rule that the host state cannot treat third state investors more favourably in management of investments. While this rule may be a useful general guideline, it is difficult to apply it in some particular situations, particularly in deciding whether the MFN clause applies to dispute resolution procedures.
For instance, one may assume that the MFN clause in the BIT between states X and Y says that the MFN clause applies in respect of “all matters related to commerce”. Most BITs generally contain such broad references (“matters related to commerce” is just one example) regarding the scope of the MFN clause. Let us further assume that the BIT between X and Y provides for a specific dispute resolution procedure. State X then enters into another BIT with state M; wherein investors of state M are granted more favourable dispute settlement procedures than investors under the X-Y BIT. Does such treatment violate the MFN clause in the X-Y BIT? If it does, can an investor from state Y rely on the “more favourable” dispute resolution provisions in the other BIT? Does the protection under an MFN clause relate only to substantive treatment, or does it include treatment in terms of procedural requirements? Does “commerce” in an MFN clause include “administration of justice”?
One might seek to argue that even dispute resolution procedures are within the scope of MFN clauses. The decision of the ICSID Tribunal in the Maffezini case supports such an interpretation. There, the Tribunal noted that there are good reasons to conclude that dispute settlement arrangements are inextricably connected to the protection of foreign investors. The Tribunal in that case went on to say:
“… if a third-party treaty contains provisions relating to dispute settlement procedures which were more favourable than those included in the basic BIT, then those beneficial provisions may be extended to the beneficiary of the MFN clause.”
This view was followed in the Siemens decision. On the basis of these decisions, one may argue that an efficient dispute-resolution system is essential for the development and promotion of commerce, and therefore the phraseology “all matters related to commerce” would include dispute resolution procedures.
On the other hand, one can argue on the basis of the ejusdem generis principle that dispute settlement procedures are outside the ambit of MFN clauses such as the one described above. Other decisions of the ICSID, notably those in the Salini and Plama cases, seem to have narrowed down the broad propositions of Maffezini. In Salini, it was held that in accordance with the ejusdem generis principle, it was not possible to read in to the MFN protection those matters which were not included in the categories specifically enumerated in the MFN clause. Thus, where dispute resolution procedures or the administration of justice was not specifically incorporated into the MFN clause, the same could not be read into the MFN protection. In Plama, the ICSID affirmed the principle that an agreement to arbitrate must be clear and unambiguous. Therefore, if an MFN clause was used to incorporate into a BIT a dispute resolution procedure from a different BIT, the parties’ intentions in this regard must be clear and unambiguous. Ordinarily, it is not possible to incorporate into one BIT (which has an MFN clause) the dispute resolution procedure from another BIT on the basis of the MFN clause.
These questions involve important issues involving elements of investment law, international law and arbitration law. The position surrounding these matters seems rather hazy, considering the conflicting decisions rendered by investment tribunals. Can these decisions be reconciled? Can a general rule be derived from these cases which will help bring about a proper understanding of the legal position? Indeed, what appears to be the best formulation of a rule governing such cases – both in law, and in policy?
Another recent decision of an ICSID tribunal (headed by Fali S. Nariman), Wintershall v. Argentine Republic, revisits this complicated issue. In Wintershall, there were extensive arguments offered on the scope of an MFN clause. The factual situation was somewhat similar to the one in Maffezini. The relevant BIT contained a dispute-resolution clause which mandated a waiting period before submitting the dispute to ICSID arbitration. Another BIT entered into by the Respondent and a third state did not contain this waiting period. It was alleged by the Claimant that the waiting period was therefore “less favourable” treatment, and that the Claimant could submit the dispute to ICSID directly. Maffezni had allowed a similar request. In Wintershall, after considering the previous decisions on the point, the Tribunal refused to follow Maffezini and said:
“In the absence of language or context to suggest the contrary, the ordinary meaning of ‘investments shall be accorded treatment no less favourable than that accorded to investments made by investors of any third State’ is that the investor’s substantive rights in respect to the investments are to be treated no less favourable than under a BIT between the host State and a third State.It is one thing to stipulate that the investor is to have the benefit of MFN treatment but quite another to use a MFN clause in a BIT to bypass a limitation in the settlement resolution clause of the very same BIT…”
Thus, it appears that the broad approach of Maffezini (of treating the MFN clause as governing both substantive and procedural matters) has been narrowed down. This may perhaps be the best course to take – there is no sound reason to disregard a specifically provided dispute settlement provision in a BIT on the basis of a separate BIT. In any case, every waiting period need not be a “less favourable” treatment – a waiting period may well be, viewed from the shoes of the parties at the time of entering into the BIT, a “more desirable” form of conciliation and settlement.
The ICSID decisions cited in this post, except for Wintershall, are available on the ICSID website here. The Wintershall decision is available here on Professor Andrew Newcombe's Investment Treaty Arbitration website - an excellent site for resources on international investment law. A discussion on the nature of a ‘waiting period’ in ICSID arbitrations, although in a different context, is found on this blog here.
A recent decision of the Bombay High Court in Clifford Chance v. DCIT revisits interesting legal issues connected to the taxation of non-residents under Section 9 of the Income Tax Act, 1961. Section 9(1) deals with “income deemed to accrue or arise in India”. The relevant part of the Section read – prior to 2007 – as follows:
Section 9(1). The following incomes shall be deemed to accrue or arise in India:
…
(vii) income by way of fees for technical services payable by –
…
(c) a person who is a non-resident, where the fees are payable in respect of services utilized in a business or profession carried on by such person in India or for the purposes of making or earning any income from any source in India…
This Section was interpreted in a relatively recent decision of the Supreme Court of India, Ishikawajima-Harima. There, the Apex Court held:
“Reading the provision in its plain sense, it can be seen that it requires two conditions to be met – the services which are the source of the income that is sought to be taxed, has to be rendered in India, as well as utilized in India, to be taxable in India. In the present case, both these conditions have not been satisfied simultaneously, therefore, excluding this income from the ambit of taxation in India. Thus, for a non-resident to be taxed on income for services, such a service needs to be rendered within India, and has to be part of a business or profession carried on by such person in India. The petitioners in the present case have provided services to persons resident in India, and though the same have been used here, they have not been rendered in India.”
Thus under Ishikawajima, it was not the mere place where a service was utilized which was determinative of the taxability of the fees received in lieu of the service. For the fees to be taxable, the services rendered to non-residents must have been both rendered in India and utilized in India. In effect, a territorial nexus requirement was sought to be established before the income would come within the purview of the Indian tax net.
Following this decision, the Finance Act 2007 inserted a retrospective amendment which provided:
Explanation. For the removal of doubts, it is hereby declared that for the purposes of this section, where income is deemed to accrue or arise in India under clauses (v), (vi) and (vii) of sub-section (1), such income shall be included in the total income of the non-resident, whether or not the non-resident has a residence or place of business or business connection in India.
A Memorandum explaining the rationale behind the amendment made it clear that the legislative intent behind the provisions was to clarify that the requirement of territorial nexus was irrelevant under clauses (v), (vi) and (vii) of Section 9(1). In considering the effect of this amendment, the Bombay High Court in November 2008 in the case of CIT v.Siemens held that the ratio of Ishikawajima was overcome by the amendment.
The decision in Clifford Chance (decided in December 2008) seems to throw doubt on this latest proposition.
Clifford Chance – a leading UK-based law firm – had provided certain advisory services for certain resident and non-resident clients who were engaged in certain projects in India. The firm had separately billed its clients for the work it had done in India and the work done outside India. The issue was whether the whole of the fees received for the services were chargeable in India, or whether only that part of the fees which was received for the services rendered in India was chargeable to tax in India. Under the law as laid down in Ishikawajima, although the service was utilized in India, part of it was not rendered in India and was therefore not chargeable to tax in India. The Department contended (following the line seemingly adopted in Siemens, although not specifically citing that case) that the ratio of Ishikawajima must be considered as having been overruled. Therefore, only the place where the service was utilized would be relevant, not the place where it was rendered. The non-residents by whom the fees were payable were clearly using the services in India – as such, the fees should have been taxable.
However, the Court relied on Ishikawajima and held that the fees would be taxable only where the service was both utilized and rendered in India. Surprisingly, the Court did not cite Siemens at all (perhaps the arguments in the case were concluded before the decision in Siemens). Furthermore, although the Revenue’s contention (that the 2007 amendment nullified the effect of the Apex Court decision) was noted, the Court did not substantively address this contention except for noting that the provisions of Section 9 were clear and unambiguous.
The Court’s decision may perhaps be read in two ways. The Court might have intended to say that under principles of statutory interpretation, an explanation cannot curtail the clear scope of a section. This reading appears to be a difficult one to sustain on the wording of the particular Section or on the basis of legislative intent.
Alternatively, the Court’s decision may be explained by arguing that the explanation does away only with “residence or place of business or business connection in India”, and not with all the requirements of territorial nexus. Thus, while the nexus need not be as deep as “residence or place of business or business connection in India”, it was still essential that the services should be rendered in India. Thus, the explanation only intended to avoid a strong nexus, but did not entirely do away with nexus.This approach would require a differentiation between rendering services in India and maintaining a business connection in India, and would also require Siemens to be adequately distinguished.
The former should not be too difficult a task. The leading decision on the concept of business connection under the Income Tax Act, CIT v. R.D. Aggarwal (AIR 1965 SC 1526), notes that 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.” Mere rendering of services, it appears, will not lead to a conclusion of the existence of business connection. On the issue of reconciliation with Siemens, it is possible to contend that what that caseactually held in its factual context is that in the post-amendment scenario, it was not necessary for the assessee to have a PE or a business connection. On its facts, however, Siemens was decided in favour of the assessee on the ground that the assessee was entitled to relief under the relevant DTAA in that case. Thus, the interpretation of Section 9 was not actually in question in that case, and the statement that Ishikawajima’s ratio was overturned by the amendment was only tentative statement and not a reasoned conclusion.
Although the purpose behind the 2007 amendment might well have been to get over Ishikawajima, the decision in Clifford Chance shows that this purpose has not actually been achieved by the wording of the explanation. Ishikawajima continues to be good law on the issue, subject to the fact that the territorial nexus requirement need not be as deep as “residence or place of business or business connection in India.” Nonetheless, mere use of services in India will not attract Section 9; the services should be both used and rendered in India.
(I recently posted the following note on Spicy IP. All references to “this blog” in the note refer to Spicy IP.)
The link between traditional IP protection and innovation has been coming under greater scrutiny. On this blog too, there has been a good deal of discussion on this aspect.
To give a few instances, Mr. Basheer in an interview to the Hindu (about which he wrote a post here) articulates concerns about formal IP systems. An anonymous reader had pointed out in the comments to another post dealing with the Indian Bayh-Dole Bill:
“… Thus, heavy-duty IP protection has not apparently lead to any increase in innovation, rather, they have stifled innovation because smaller and nimbler companies which are innovative get gobbled up by the Big Pharma and all their innovative potential is sidelined to protect the big franchises. The software industry, on the other hand, has thrived without much patent protection, at least in the past….”
I had highlighted a report of an international Expert Group on moving towards an era of “New IP”. The report dealt with the problems of traditional IP systems in encouraging innovation, and recommended:
“Working together, governments, universities and industry should develop new measures of the success of technology transfer, and other means of development and social investment that better correspond to desired social and economic return…”
In my post on the Expert Group report, I had been somewhat critical of the report for failing to suggest concrete measures to promote innovation in the framework of the so-called “New IP”. One solution proposed by Mr. Basheer tries to indicate the way forward in conceptualizing “informal” IP norms which would actually promote innovation (also see the comments to that post).
It is in this background that I came across an interesting piece in today’s Business Standard. The article gives instances of what has been done by the government in promoting innovation outside the formal IP framework – namely the setting up of new IITs and IIMs etc. Nonetheless, quite clearly, the mere setting up of newer institutions will not on its own promote greater innovation. One further recommendation made by the Business Standard is worth mentioning here.
“That in turn raises several other problems: the failure to attract a sufficient number of bright students to the science stream at the secondary and higher secondary school level (which is when a fascination with science can be expected to begin), and perhaps linked to this the poor remuneration level and even poorer career advancement prospects for those who opt for a career in science teaching and research...These disadvantages can be neutralised by offering open learning environments in university departments, along good laboratory and other facilities... With the intellectual property regime getting tighter, shutting out the scope for reverse engineering and re-engineering to launch new products, the need for indigenous research to sustain economic development has become even more important than before. Greater investment in research and development, including fundamental research, is imperative. And both the public and private sectors need to contribute.”
Broadly, there seems to be increasing consensus (or at least, there seems to be a movement toward a consensus) that the mere existence of formal structures of IP protection does not go far enough in promoting innovation; indeed, it may well be stifling innovation. How do we go about developing informal norms? What must the content of these informal norms be? Can informal norms be ‘enforced’ in law, or must acceptance of and compliance with such norms necessarily be based on consensus of all participants in the process? If so, how can such consensus be achieved?
A few answers may be found by developing ideas of “social innovation”, as detailed in this post. Additionally, the draft National Innovation Act, 2008 may provide a few leads – Sections 3, 4 and 5 appear to be particularly important. See this post for details on Sections 3 and 4 and for some potential problems in those sections. Section 5 of the Act provides for incentives to angel investors. “Angel investors” are defined in Section 2 to mean:
“(1) ‘Angel Investor’ means a person, or entity that provides risk capital to start up ventures established for facilitating and encouraging Innovation (whether proprietorships, partnerships or cooperatives or companies or any other legal entity), not involved in the management, but adding value through association by providing access to commercialization of products and services through its social, professional or business networks and expertise.”
It remains an open question as to whether these provisions will have any actual impact in promoting innovation. Nonetheless, there is at least a beginning to the process of promoting innovation. On the other hand, the Indian Bayh-Dole seems to be more of a disappointment in terms of promoting innovation. Clearly, much remains to be done in terms of finding more solutions...
I have been unable to regularly post on this blog over the past couple of weeks. I will write on the decisions of the Bombay HC in Clifford Chance and Siemens in another couple of days (see this post). Regular postings will resume in another week or so
Posts
