Introduction

Every cross-border sale between related entities within a multinational group carries a singular price tag that they would have agreed upon. Still, that price tag is required to withstand two distinct legal tests and interrogations. For example, when a German parent company sells a product to its Indian subsidiary, India's customs authorities ask whether the price is high enough to reflect what an unrelated buyer would have paid, since a deflated or artificially lowered price erodes customs revenue. On the other hand, India's income tax authorities, examining the same transaction months later, would ask the opposite question: whether the price is low enough that it does not artificially raise the subsidiary's costs and shift taxable profit abroad (where it can, perhaps, be taxed at a lower rate than in India)? Thus, the same transaction is put under scrutiny in two lights by two distinct regulatory regimes that, on one hand, share a common antecedent with respect to the arm's-length principle but fundamentally differ in method, timing, and administrative control. This article examines this divide, the insubstantial confluence attempted at the level of the WTO, WCO, and OECD, and whether India has harmonised or just a coordinated framework.

Two Regimes Based on the Same Fiction

Customs valuation in India is set out in Section 14 of the Customs Act, 1962, and the Customs Valuation (Determination of Value of Imported Goods) Rules, 2007 (“CVR 2007”), which were created to give domestic implementation of the WTO's Agreement on Implementation of Article VII of GATT 1994. Rule 3(3) of CVR 2007 prescribes and permits customs to accept the transaction value between two related parties only on the condition that an examination of the “circumstances of the sale” reflects that the relationship between the parties did not influence the stated price. This is tested by comparing the stated price to prices of identical or similar goods sold to unrelated buyers in India around the same time. It is a narrow approach that is shipment-specific and a price-to-price comparison. Further, it is administered by CBIC through Special Valuation Branch proceedings.

Contrastingly, transfer pricing functions under a distinctly different statute. This was envisioned in Chapter X of the Income Tax Act, 1961 (primarily Sections 92 to 92F) until 31^st March, 2026. Thereafter, from 1 April 2026, with the establishment of the Income Tax Act, 2025, the operative provisions remain Chapter X, Sections 161 to 174, which reproduce the same substantive content but under new numbering. Section 161 prescribes the arm's-length principle, which was previously highlighted in Section 92. Section 162 prescribes “associated enterprise”, which was previously highlighted in Section 92A. Furthermore, Section 165 states the methods of the arm's length principle, which were previously in Section 92C. These methods include the Comparable Uncontrolled Price method, Resale Price Method, Cost Plus Method, Profit Split Method, and lastly the Transactional Net Margin Method (TNMM). In practice and in reality, the Transactional Net Margin Method is most often used, and it doesn’t compare individual shipment prices; rather, it compares the combined operating margin of the Indian entity for a full financial year with the margins earned by a selected group of comparable independent companies. It is to be noted that the numbering in this piece is that of the 2025 Act, with the section number of the 1961 Act included where it helps with continuity in existing literature.

Thus, the two separate systems test the same transaction, but they use different parameters of analysis. One is transactional and instantaneous, triggered at the point that goods cross the border, while the other is at the entity-level and retrospective in nature, triggered at the close of an accounting period, i.e., year-end. They also have different statutory definitions of what constitutes “related” parties in the first place. In customs, under Rule 2(2) of CVR 2007, the threshold is much lower and control-based, whereas in tax, under Section 162 of the Income Tax Act, 2025, the threshold is higher and more on the grounds of ownership or dependency. So, a relationship that comes under scrutiny under one regime may not automatically trigger it under the other.

Where the Friction Gets Real

The most immediate flashpoint is the year-end "true-up." Transfer pricing practice often involves retrospective price adjustments, which are basically credit notes issued after the end of the financial year, and it adjusts the transfer price so that the margin of the Indian entity is within the arm’s length range provided by its TP study. But by the time the adjustment is made, the goods to which they relate have already been imported, assessed, and cleared through customs, usually many months prior. Thus, the legal question that arises is whether a post-importation transfer pricing adjustment with a retrospective effect alters the customs value, and whether it entitles the importer to a refund of duty paid (or exposes it to a demand for additional duty)? There is no settled answer to this in Indian law.

The Court of Justice of the European Union dealt with exactly the above question in Hamamatsu Photonics Deutschland GmbH v Hauptzollamt München in 2017, holding that a taxpayer could not simply apply a year-end transfer pricing adjustment to reduce customs value and claim a refund, because the adjustment as structured did not correspond to any method recognised under the EU's customs valuation framework, itself that was modelled on the same WTO Valuation Agreement that India has implemented through the CVR. The decision is of value, essentially because it shows that even in a jurisdiction with a much more sophisticated customs-tax system than India, the two values are legally uncoupled until the adjustment is built, from the outset, in a form that customs law would recognise.

This decoupling is not only a timing inconvenience. It means a company can pay customs duty based on a number its own tax filings would disavow as not being the true arm's-length price, which will be an outcome in which the state collects duty on a number it does not, for tax purposes, believe is real.

The Limited Role of the WTO, WCO and OECD

None of the three main international institutions in this arena has attempted to properly align and unify the tests. The WTO's Valuation Agreement is silent on transfer pricing altogether. It was negotiated decades before the issue of profit-shifting through intra-group pricing became a central international tax issue, and despite time and concerns, it hasn't been sufficiently amended to address the interaction.

Rather, the World Customs Organization and the OECD have pursued soft convergence through guidance rather than binding rules. The WCO’s 2015 publication, Guide to Customs Valuation and Transfer Pricing (revised in 2018 and developed with input from the OECD and ICC), encourages customs administrations to view a taxpayer’s transfer pricing documentation as one relevant piece of evidence under the “circumstances of sale” test in Article 1.2(a) of the Valuation Agreement, without automatically treating a tax-compliant transfer price as determinative for customs purposes. The WCO's Technical Committee on Customs Valuation had already recognised, in Commentary 23.1 (2010), that a transfer pricing study may be used to examine the 'circumstances of the sale,' and has since elaborated this through Case Studies 14.1 (2016) and 14.2 (2017), illustrating how TP documentation such as functional analyses and economic studies might inform, but not dictate, a customs valuation decision.

Separately, the OECD's Base Erosion and Profit Shifting project, notably Actions 8–10 on aligning transfer pricing outcomes with value creation, tightened the substance-based analysis expected in transfer pricing but had no corresponding customs valuation component, and indeed complicated one specific customs question that is the dutiability of royalties and licence fees bundled into an import price under Rule 10 of CVR 2007 by pushing multinational groups to more elaborate intangible-ownership and risk-allocation structures that customs authorities must now unpack.

The picture is of parallel institutional tracks that do sometimes acknowledge each other's existence but have not produced binding cross-recognition. The guidance is nonetheless not binding at the national level, and no WTO dispute panel has yet been asked to rule on the interaction directly,

The Accuracy-Administerability Tradeoff

It’s worth stopping to ask why the two tests are built this crudely in the first place, because the answer is directly based on what “fixing” the divide should actually mean. Customs processes thousands of shipments a day at the border. It cannot conduct a searching economic inquiry into whether the price of a particular shipment was affected by, say, a currency swing or a market-entry discounting strategy, so it deliberately employs a fast, mechanical price-to-price comparison. It doesn't take into account other factors. TNMM itself is a blunt proxy; however, transfer pricing audits take more time. An Indian subsidiary’s annual margin can be dragged down by one-off marketing spend, warehouse set-up costs, or a price war with a competitor. These aforementioned factors have nothing to do with whether the import price was fair. “Comparable” companies selected for benchmarking are never a perfect match in scale, product mix, or accounting policy. So, a real arm’s-length price may, for that reason, not pass one test, or either, for reasons unconnected with any manipulation at all.

This is not an Indian peculiarity, as it is a live, unresolved tension in both fields independently. The OECD’s own move under BEPS Actions 8–10 to align transfer pricing outcomes with value creation rather than with pure aggregate-margin comparison is itself a partial response to criticism that TNMM is too blunt an instrument. [MJ2] In customs valuation practice, the mirror criticism is that a strict, shipment-level comparable test cannot accommodate genuine commercial reasons for price variation. Any reform agenda for India shouldn’t end with procedural bridging between CBIC and CBDT; rather, it should also be in line with this larger, still ongoing international move toward more substance and factor-based testing on each side, while openly admitting that neither the WCO nor the OECD has completely worked out the trade-off between administrability and accuracy after umpteen years of trying. Co-ordination between the two Indian authorities may reduce duplicated conflict, but cannot in itself cure the underlying bluntness of either test.

Should India Harmonise?

India’s two regulators work in near-total isolation. CBIC’s Special Valuation Branch audits related-party import pricing notwithstanding the taxpayer’s transfer pricing documentation (accountant’s report, erstwhile Form 3CEB and now revamped under Section 172 of the Income Tax Act, 2025) or its Advance Pricing Agreement under Section 168 (earlier Section 92CC). On the other hand, transfer pricing officers of CBDT carry out margin-based audits without reference to SVB findings on the same transaction. A taxpayer may have a customs order accepting its declared value and a TP audit adjustment restating the arm’s length price for the same transaction, with neither authority bound by or even required to consider the other’s conclusion.

Full substantive harmonisation, wherein one legal test applied by one authority is highly unlikely to be good policy because the two regimes are not measuring the same thing even in theory. Customs protects the revenue base by comparing one price at one moment to real market comparables. On the contrary, transfer pricing protects the tax base by comparing an entity’s overall profitability to comparable business outcomes over a period of time. To combine these into a single test would likely distort one objective in the service of the other.

More defensible reform is procedural, not substantive. It begins with statutory recognition of transfer pricing documentation as rebuttable, but not conclusive, evidence in SVB proceedings under Rule 3(3) of CVR 2007. This would reflect the WCO-OECD guidance that India has yet to implement domestically. Secondly, a collaborative CBIC-CBDT advance ruling mechanism will enhance the existing APA framework under Section 168 of the Income Tax Act, 2025. This would allow formally seeking and recording a customs valuation perspective in conjunction with it and not just a singular merged ruling. This would be a linked process akin to the dedicated advance-ruling channels operated by Australia’s Border Force and Singapore Customs, run alongside the Advance Pricing Arrangement programmes maintained separately by the ATO and IRAS. [MJ3] Such a bridge would not ask either regulator to abandon its own legal test, but rather would stop the current position wherein honest, correctly-priced transactions are forced to litigate their legitimacy twice, before two different arms of the same government, using two different, non-communicating standards of proof.

Conclusion

The tension between the customs valuation and transfer pricing cannot be boiled down to a drafting oversight that a single amendment can fix. In reality, it reflects two regimes designed for genuinely different purposes, each defensible in isolation, and each deliberately blunt for administrability reasons that neither the WCO nor the OECD has fully resolved. What cannot be justified is the total lack of institutional communication between them in India when the WCO and OECD have already mapped out, albeit tentatively, how such communication might work. A single harmonised test is neither necessary nor desirable. A coordinated procedural structure linking CBIC and CBDT in a pre-dispute setting rather than in a post-dispute setting is necessary and, on the comparative evidence, feasible even if it leaves the deeper, still-unresolved question of how much sharper either test can get without losing its administrative ability for another day.

*Authored by- Anisha Giri Goswami, a 4th year B.A.LL.B (Hons.) student at National Law Institute University, Bhopal. Views expressed are personal.