Friends, on 15th January 2026, we all witnessed a six years old seemingly routine withholding tax dispute, transforming into one of the most consequential ‘General Anti Avoidance Rules’ (GAAR) related adjudications in Indian tax jurisprudence.
What started as a non-resident investor’s request for a nil withholding certificate under section 197 of the Income Tax Act, 1961, ultimately culminated in a judgment of the Hon’ble Supreme Court of India examining the primacy of GAAR over treaty benefits and grandfathered protections, even though ironically, GAAR was never formally invoked by the Revenue authorities, in the said case.
The hon’ble Apex Court’s judgement in the case of ‘Tiger Global International Holdings’ [TS-38-SC-2026], decision thus transcends the confines of the original dispute and invites closer reflection on process, proportionality, and the evolving contours of anti-avoidance jurisprudence in India.
From a Threshold Jurisdictional Bar to a Substantive GAAR Verdict
The applicants namely Tiger Global International II/III/IV Holdings, were regulated by the Financial Services Commission in Mauritius, held a Category 1 Global Business Licence under section 72(6) of the Financial Services Act, 2007, and were tax residents of Mauritius both under domestic law and under the India Mauritius Double Taxation Avoidance Agreement. They were set up with the primary objective of undertaking investment activities with the intention of earning long term capital appreciation and investment income.
The applicants had acquired shares of Flipkart Private Limited, a company incorporated in Singapore, during the period from October 2011 to April 2015. These acquisitions were undisputedly made prior to 1 April 2017, the cut-off date introduced for capital gains taxation and treaty grandfathering as per India-Mauritius Double Taxation Avoidance Agreement (DTAA). The Singapore company, in turn, held investments in Flipkart India Pvt Ltd, an Indian company, with the value of its shares being derived substantially from assets located in India.
On 18 August 2018, the applicants transferred part of their shareholding in the Singapore company to Fit Holdings S.A.R.L., a Luxembourg entity, as part of a larger global transaction involving the acquisition of a controlling stake in the Singapore company by Walmart Inc. from several shareholders including the applicants.
Prior to consummation of the transfer, the applicants approached the Indian tax authorities on 2 August 2018 under section 197, seeking a certificate for nil withholding of tax. By communication dated 17 August 2018, the tax authorities declined the request, taking the view that the applicants were not entitled to treaty benefits on the ground that they were not independent in their decision making and that effective control over the transaction did not rest with them. Certificates were accordingly issued prescribing withholding tax at varying rates.
Aggrieved by this determination, the applicants approached the Authority for Advance Rulings (AAR) on 19 February 2019 under section 245Q(1), seeking a ruling on whether the capital gains arising from the transfer of shares of the Singapore company to the Luxembourg buyer were chargeable to tax in India under the Income tax Act read with the India Mauritius DTAA.
What followed thereafter transformed a limited withholding tax dispute into one of the most consequential GAAR related adjudications in Indian tax jurisprudence.
Admissibility before the Authority for Advance Rulings (AAR)
Before the Authority for Advance Rulings, the controversy was narrow and jurisdictional. The AAR did not examine the merits of treaty entitlement or capital gains taxation. It confined itself to the admissibility bar under clause (iii) of the proviso to section 245R(2), forming a prima facie view that the transaction appeared to be designed for the avoidance of tax. GAAR and Chapter X-A were neither pleaded nor invoked. There was no reference to Rule 10U or Rule 10UA, nor any discussion on impermissible avoidance arrangements under sections 95 to 102.
At this stage, the case was not about GAAR. It was about the statutory limits of the AAR’s jurisdiction.
The High Court Stage: GAAR Enters the Discourse Theoretically
The narrative shifted before the hon’ble Delhi High Court. It was at this stage that the Revenue’s standing counsel, for the first time, sought to rely upon Rule 10U(2) of the Income Tax Rules, to contend that GAAR could still apply notwithstanding grandfathering.
This submission was made despite the fact that no GAAR notice had been issued, no approving panel under section 144BA had been constituted, and no proceedings under Chapter X A had been initiated. The High Court itself acknowledged that it was dealing only with the correctness of the AAR’s rejection under section 245R(2)(iii). Nevertheless, it undertook an interpretative examination of GAAR and Rule 10U.
Significantly, the High Court rejected the Revenue’s contention that Rule 10U(2) could override the grandfathering protection under Rule 10U(1)(d). Relying on the Supreme Court’s decision in ITO v. Gwalior Rayon Silk Manufacturing (Weaving) Co. Ltd., the Court held that the expression ‘without prejudice to’ contained in Rule 10U(2) cannot be read so as to nullify the express grandfathering granted by the immediately preceding provision in Rule 10U(1)(d). GAAR thus entered the High Court judgment only as a theoretical construct, not as an applied law.
The Supreme Court Stage: GAAR Verdict Without GAAR Proceedings
The most consequential transformation occurred before the Supreme Court. The hon’ble Apex Court moved well beyond the threshold jurisdictional issue and made observations suggesting that by introducing section 90(2A), the Legislature curtailed the absolute benefit of Treaty override u/s 90(2) by making it subject to GAAR provisions, and that CBDT circulars and Treaty induced grandfathering protection, issued in the pre GAAR regime, cannot survive subsequent statutory amendments.
What is striking, however, is that these conclusions emerged without any formal invocation of GAAR by the Revenue. The statutory preconditions for GAAR were never examined because no GAAR proceedings existed. Instead, GAAR was treated as an already applicable and operative principle, drawing heavily from the theoretical discussion before the High Court and elevating it into a determinative doctrine.
This raises a fundamental jurisprudential question. How can a threshold issue of admissibility of an NRI’s application under section 245R(2)(iii) of the Act, which requires only a prima facie satisfaction, be transformed into a binding declaration on GAAR supremacy in the absence of any initiation of Chapter X-A proceedings.
The core reasoning of the hon’ble Supreme Court judgment, particularly as reflected in paragraph 27, proceeds on the premise that the subsequent insertion of Chapter X-A comprising sections 95 to 101 and the corresponding Rules, including Rule 10U, has rendered some of the ratios of the earlier Apex Court’s judgements in the cases of ‘Azadi Bachao’ and ‘Vodafone’ and the CBDT Circular No. 789, redundant and inoperative, and that treaty benefits, including those otherwise grandfathered, are now expressly subject to GAAR scrutiny. On this foundation, the Court justified the subordination of treaty protection and circular based assurances to the GAAR regime.
While this proposition may hold relevance in an appropriate factual and procedural setting, the difficulty in the present case lies elsewhere. The entire discussion on GAAR, its supremacy, and its overriding effect on treaty benefits, was purely theoretical, divorced from the statutory framework that governs the actual invocation and application of GAAR.
Chapter X-A does not permit GAAR to operate in abstraction. It is a procedure intensive regime, triggered only through a carefully sequenced statutory process. First, GAAR can be examined only in the course of pending assessment proceedings, where the Assessing Officer, upon forming an opinion, is required to make a reference to the Principal Commissioner under section 144BA(1) of the Act. Thereafter, if the Principal Commissioner is not satisfied with the explanation of the assessee, a further reference must be made to the GAAR Approving Panel under section 144BA(4) of the Act, which alone is empowered to take a final view on the applicability of GAAR, after granting the assessee an effective opportunity of being heard.
In the present case, none of these mandatory statutory steps had even commenced, let alone been fulfilled. The subject matter before the appellate forums, including the Supreme Court, had consistently remained the rejection of an advance ruling application by the AAR. At no stage did the Revenue initiate GAAR proceedings, constitute an approving panel, or subject the assessee to the procedural safeguards expressly embedded in Chapter X-A. Indeed, GAAR was not even pleaded by the Revenue before the AAR.
It is in this context that the hon’ble Supreme Court’s extensive reasoning on GAAR assumes an unusual and somewhat anomalous character. The Court effectively pronounced upon the applicability of GAAR, the impermissible nature of the arrangement, and the subordination of treaty and circular based protections, without the statutory GAAR machinery ever having been set in motion, in the present case.
This leads to an ironic and consequential situation. If, pursuant to the judgment, the Revenue were now to initiate GAAR proceedings, the assessee would arguably stand precluded from meaningfully contesting the invocation. Equally, the Principal Commissioner or even the Approving Panel may find themselves constrained in independently evaluating the assessee’s explanation, given that the Supreme Court has already, in substance, characterised the arrangement as impermissible under GAAR, albeit in advance and without following the statutory process.
In effect, the elaborate safeguards built into Chapter X-A risk being rendered illusory, not by legislative design, but by a prior judicial determination delivered in a proceeding where GAAR was never invoked at all. This inversion of sequence, where conclusion precedes invocation, and adjudication precedes initiation, raises serious concerns about procedural discipline in the application of GAAR.
It is also noteworthy that, after extensively anchoring the reasoning in the applicability of GAAR and the role of Rule 10U(2) in addressing the limits of grandfathering, the judgment, in its concluding observations in paragraph 48, records that even if GAAR were to be regarded as inapplicable, the transaction may still be examined under the Judicial Anti Avoidance Rule (JAAR), drawing upon the substance over form doctrine.
This alternative reference to JAAR, while undoubtedly, well established in tax jurisprudence, is made without a detailed exposition on how such a doctrine would operate in the presence of an express statutory grandfathering provision, particularly when the preceding analysis had rested primarily on legislative amendments and GAAR specific provisions to address that very question.
Rule 10U and the Question of Grandfathering
A particularly nuanced aspect of the judgment arises from its engagement with Rule 10U(2). Rule 10U(1)(d) expressly provides grandfathering protection in respect of investments made prior to the prescribed cut-off date of 1.4.2017. Rule 10U(2), however, begins with the formulation “without prejudice to the provisions of clause (d) of sub rule (1)”, a phrase that assumes central importance in the interpretative exercise.
Under settled principles of statutory interpretation, as authoritatively explained by the Supreme Court in Gwalior Rayon Silk Manufacturing (Weaving) Co. Ltd., the expression “without prejudice to” is intended to preserve and coexist with the provision to which it refers. It does not operate as an overriding or non obstante clause. Reading Rule 10U(2) as having the effect of overriding Rule 10U(1)(d) would amount to sub rule (2) immediately taking away what stood expressly saved in the immediately preceding provision, namely clause (d) of sub rule (1). Such an interpretation would inevitably lead to a wholly irreconcilable conflict between the two aforenoted provisions.
An allied concern emerges from the distinction noticed in the judgment between an “investment” referred to in Rule 10U(1)(d) and an “arrangement” contemplated under Rule 10U(2). While the distinction may have been drawn to facilitate scrutiny under anti avoidance principles, its deployment to effectively override and nullify a bilaterally accepted grandfathering commitment, through subsequent unilateral domestic law interpretation, raises legitimate concerns. Such a distinction could perhaps have been avoided, lest the grandfathering assurance be reduced to a largely empty ritual, particularly when, in substance, every investment can potentially be characterised as an arrangement.
The more fundamental point is that the Indian Government and the Legislature consciously agreed to grandfathering protection for investments made prior to 1 April 2017, both as a matter of treaty obligation and as a matter of investor assurance. Revisiting that commitment through a GAAR or substance over form lens, long after the inflow of funds in Indian businesses, does not sit comfortably with the objectives underlying the grandfathering protocol. From a broader perspective, such an approach may not augur well for India’s positioning as an investor friendly jurisdiction, particularly in an environment where policy certainty and credibility of commitments assume heightened importance.
Equally, the evolving geopolitical landscape and contemporary assertions of economic sovereignty, however legitimate, should not operate to retrospectively dilute or re-interpret assurances that were consciously extended and relied upon by investors at the time of making their investments. Grandfathering, by its very nature, is a commitment to the past, and its strength lies in the confidence that such commitments will be honoured, notwithstanding subsequent shifts in policy perspective.
GAAR was introduced as an exceptional anti-abuse measure, not as a default weapon. Its objective was to target impermissible avoidance arrangements, not legitimate tax planning undertaken within the framework of law. The judgment, however, blurs this distinction by facilitating GAAR’s entry even into grandfathered and otherwise compliant structures through conceptual distinctions between an investment and an arrangement.
This approach risks unsettling investor confidence and undermining policy certainty, particularly in a jurisdiction that actively seeks long term capital inflows.
Treaty Policy, Capital Inflows and the GIFT City Analogy
A useful way to contextualise the debate on treaty shopping and GAAR is to revisit the original policy objective underlying India’s DTAA with Mauritius. That treaty was consciously negotiated and sustained for decades with a clear economic purpose: to facilitate and channel foreign capital into India at a time when domestic capital was scarce and India was positioning itself as an attractive investment destination. That this objective was substantially achieved is also borne out by official data released by the Department for Promotion of Industry and Internal Trade, which indicates that nearly 25 percent of total foreign direct investment inflows into India over the last decade have come through the Mauritian route.
Viewed from that lens, a more pragmatic question arises. What real economic difference does it make if investments into India are made by a native Mauritian entity or by a US based investor who has chosen to route investments through a Mauritius incorporated entity for tax optimisation purpose, so long as the investment ultimately flows into India, creates economic value, and complies with the law, as it stood at the time of investment. In both cases, the end result is identical: foreign capital entering India under a treaty framework consciously designed to attract such capital, with grandfathered protection expressly promised by the legislature.
The increasing tendency to view treaty shopping as a taboo or an abuse, without adequate regard to legislative intent and historical policy context, therefore warrants a careful and pragmatic reassessment.
A useful contemporary parallel can be found in the development of the ‘Gujarat International Finance Tec’ (GIFT) City. GIFT City has been deliberately created as a financial enclave offering a comprehensive bouquet of tax incentives, regulatory relaxations, and operational ease, with the explicit objective of attracting global capital, fund managers, and financial institutions into India. These incentives are not regarded as abusive, suspect, or impermissible. On the contrary, they are actively marketed as a policy tool to position India as a competitive international financial hub.
Significantly, investments routed through GIFT City are not viewed through a presumption of avoidance, nor are they reflexively subjected to GAAR scrutiny merely because they avail legislatively sanctioned tax benefits. The policy narrative surrounding GIFT City recognises an important principle: tax incentives offered by the State as part of an economic strategy cannot simultaneously be treated as a misuse of law.
In substance, the Mauritius route and GIFT City serve a similar macroeconomic purpose. Both are instruments consciously crafted to attract foreign capital into India by offering tax certainty and competitive outcomes. If GIFT City is not considered a taboo or an abuse, and its incentives are not seen as inviting the ire of revenue authorities or GAAR, it becomes difficult to justify a fundamentally different perception for treaty-based investments made under a grandfathered DTAA regime.
This analogy highlights the need for a more balanced approach to GAAR and treaty interpretation, one that distinguishes legitimate policy driven tax structuring from artificial abuse, and that respects the State’s own role in designing capital inflow frameworks. Treaties and tax incentives are not loopholes. They are instruments of economic policy. Treating them otherwise risks eroding investor confidence and undermining the very objectives they were created to achieve.
Indirect Transfers: Treaty Threshold & Domestic Law
An additional dimension of the judgment emerges from the observations contained in paragraphs 16, 17 and 18, where it has been held that, for availing grandfathering protection under Article 13(4) of the India–Mauritius DTAA, the person claiming treaty protection must not only qualify as a resident of the other Contracting State, namely Mauritius, but must also establish that the movable property or shares forming the subject matter of the transaction are directly held by such resident entity. On this reasoning, an indirect transfer of shares would not, at the threshold, fall within the treaty protection contemplated under Article 13.
However, when examined closely, this reasoning presents an inherent conceptual complexity. If indirect ownership is held to be insufficient for threshold treaty eligibility, then, by the same logic, a transaction involving the sale of shares of a Singapore incorporated company by a Mauritian resident entity would itself stand outside the chargeability framework under the Indian Income Tax Act. It cannot logically be the case that an indirect transfer is disregarded for treaty entitlement purposes, yet simultaneously treated as taxable under domestic law. If such a construction were to be accepted, the very basis of taxability of the transaction in India would become questionable, and the authority of the Revenue to subject the transaction to tax under the Act would stand considerably weakened. Viewed thus, the reasoning adopted on this aspect appears to traverse a line of analysis that was neither urged nor canvassed by either side before the adjudicatory authorities, and yet assumes a determinative role in the ultimate outcome.
TRC, DTAA and the Risk of Over Extension
Concerns have been expressed that the Supreme Court’s observations on the non-conclusivity of the Tax Residency Certificate may spill over into routine foreign remittance transactions. Such apprehensions appear overstated. In cases involving royalties, fees for technical services, dividends or business income, the foreign recipient typically carries on real and substantive business operations in its country of residence. In such contexts, treaty relief continues to serve the legitimate objective of avoiding double taxation, a principle expressly acknowledged by the Supreme Court itself.
GAAR, Treaties and Constitutional Balance
The discussion on the interplay between section 90(2) and section 90(2A) raises deeper constitutional questions. Treaties are implemented under Article 253 of the Constitution, while the subsequent elevation of GAAR regime introduces tension between international commitments and domestic anti-avoidance policy. This delicate balance may yet warrant reconsideration by a larger Bench of the hon’ble Supreme Court of India.
Conclusion: Balancing Investor Confidence & Anti-Avoidance
The Tiger Global judgment invites reflection on several interconnected themes that go well beyond the facts of the case. It traces an unusual journey from a routine withholding tax application to a wide-ranging adjudication on treaty interpretation, grandfathering, GAAR, and JAAR principles, all in a context where GAAR was never formally invoked through the statutory process. The discussion on Rule 10U, the treatment of grandfathered investments, the distinction drawn between investments and arrangements, and the reliance on both GAAR and JAAR doctrines collectively invite careful reflection on procedural discipline, legislative intent, and the durability of policy commitments. Equally, the judgment prompts a reconsideration of how treaty-based capital inflow frameworks and incentive driven regimes are perceived, and whether availing such structures should, by itself, invite suspicion of abuse. As India positions itself as an investor friendly jurisdiction in a complex global environment, the enduring lesson lies in ensuring that anti avoidance tools remain calibrated and principled, that past assurances are honoured with consistency, and that the pursuit of substance is balanced with certainty, proportionality, and respect for the rule of law.
[This Article, authored by our Founder, Shri Mayank Mohanka, has also been published in Taxsutra. The pdf copy of the same is enclosed below.]
.jpg)