Home › Is the Profit Split Method Becoming the New Transactional Net Margin Method? Evolution of Transfer Pricing in the Post-BEPS Era

Is the Profit Split Method Becoming the New Transactional Net Margin Method? Evolution of Transfer Pricing in the Post-BEPS Era

In the Indian Transfer Pricing (“TP”) environment, the Transactional Net Margin Method (“TNMM”) is the most frequently used method, both by taxpayers and by the tax authorities. The Profit Split Method (“PSM”), on the other hand, has generally been reserved for transactions involving unique intangibles, or for cases where each party brings unique and valuable contributions. In the Indian transfer pricing practice, TNMM has been widely applied, particularly for benchmarking routine services, manufacturing and distribution transactions. However, selection of the most appropriate method (MAM) depends on the facts and circumstances of each transaction.

The OECD / G20 project on Base Erosion and Profit Shifting (“BEPS”) has changed the focus. The question now is where value is created, and TP outcomes ought to be aligned with that. Under Action Plans 8-10, entities performing Development, Enhancement, Maintenance, Protection and Exploitation (“DEMPE”) functions in relation to intangibles must be remunerated for those functions. PSM may be appropriate where the parties make unique and valuable contributions or meet other relevant conditions for applying the method. Hard-to-value intangibles (“HTVI”) should also be analysed with rigorous TP analysis.1

The OECD’s June 2026 consultation draft proposing revisions to Chapter VII of its Transfer Pricing Guidelines, dealing with intra-group services, further emphasises the importance of selecting the most appropriate transfer pricing method based on the facts and circumstances of each transaction. The draft clarifies that cost-based methods, such as the Cost Plus Method or cost-based TNMM, should not automatically be preferred for intra-group services. Instead, the selection should consider the economically relevant characteristics of the services, including the use of unique and valuable intangibles and the risk profiles of the parties. However, the proposed revisions do not establish any preference for PSM or change the underlying principles of method selection. They reinforce the need to evaluate each transaction on its merits.2

This article looks at that question. The author’s view is that TNMM is not going anywhere, and it will remain the most widely used method for routine transactions that can be reliably benchmarked. What is changing is that PSM is receiving greater attention in cases involving complex and highly integrated transactions, including circumstances where a residual profit split may be considered after determining routine returns. Indian Tax Tribunals (“ITAT”) have gone both ways on this. They have accepted a PSM for integrated operations (Global One India Private Limited3, Infogain India Private Limited4) and rejected it for a licensed manufacturer performing routine functions (Toyota Boshoku Automotive India Private Limited5).

Introduction

TNMM has been selected as the MAM in the majority of cases for a long time. It is easy to apply, as comparable companies are selected from commercial benchmarking databases after quantitative filters and qualitative analysis, and their net operating margins are compared with those of the tested party. Taxpayers and tax officers have regarded this as the MAM for testing simpler functions. In India, captive service providers, contract manufacturers and limited-risk distributors are predominantly large in number, and almost all of them have been benchmarked using TNMM.

The BEPS Action Plans 8-10 recommendations placed greater emphasis on accurately delineating controlled transactions by examining contractual arrangements alongside the parties’ actual conduct, including the functions performed, assets used and risks assumed and controlled.6

In today’s era, businesses are different from the ones TNMM was designed to apply. A few years ago, a typical Multinational Group made its key decisions at headquarters, owned its intellectual property in one place and used its overseas subsidiaries mainly to carry out instructions. This model is fading. It is now seen that research may happen in one country, product development in another, data analytics in a third and key commercial decisions somewhere else. Functions, assets and risks are now spread across the Group, and the entities often depend on each other to earn a profit. Where different entities make unique and valuable contributions, or where their activities and risks are so closely integrated that their contributions cannot be reliably evaluated separately, a one-sided method such as TNMM may not provide the most reliable result. In such circumstances, PSM should be evaluated alongside other potentially appropriate methods.

Why TNMM Became the Preferred Method

TNMM became the preferred method because, in most cases, it gives a more reliable result than the other methods, mainly because reliable comparable data is available. Close comparables to traditional methods are often hard to find, as they require stringent comparability. TNMM tests net margins, which are less affected by small differences in products and functions, so it can be applied with reasonable confidence across a wide range of transactions. That reliability, together with its practicality, is why taxpayers and tax authorities have relied on it. However, TNMM works best when the tested party performs routine functions and the other party contributes most of the value. As businesses become more integrated and more entities contribute to the same profit, TNMM may not capture the full picture.7

Let’s take a simple example. An Indian captive software entity provides coding and testing services to its foreign associated enterprise. It owns no intellectual property, bears little risk and is remunerated at a cost-plus 15% mark-up. If its operating costs are INR 100 crore, it earns INR 15 crore as its operating profit. If the uncontrolled comparable companies providing software services selected from the commercial databases earn mark-ups in the range of 12% to 18%, the tested party’s mark-up of 15% can be considered to meet the arm’s length test as per the Indian TP regulations. The analysis is simple, given the availability of data on commercial databases.

The OECD TP Guidelines of 1995 in paragraph 3.49 gave preference to the traditional transaction methods over profit-based methods.8 The 2010 revision did away with the preference for traditional transaction methods and replaced the hierarchy with a “MAM”. In practice, this allowed taxpayers and tax administrations to settle on whichever method was simplest to apply and suited the facts of the case. In India, the TP cases so far have largely been about captive Information Technology (“IT”) and IT-enabled service providers, BPO entities, contract research and contract manufacturing entities and limited-risk distributors. Benchmarking these entities has typically involved running a search on Indian databases for companies with a similar functional profile. In my experience, this is why TNMM has ended up being applied in the large majority of Indian TP cases.

Where TNMM Falls Short

That is only one side of the analysis. TNMM is considered the MAM where the entity performs simpler functions. It is not considered the preferred method where both parties to the international transaction contribute to the intangibles. Consider a Pharmaceutical Group. The parent entity develops new molecules and is the owner of the intellectual property (which includes the patents). On the other hand, the Indian subsidiary undertakes clinical trials, deputes its scientists to global projects, obtains key regulatory approvals and builds knowledge of the local market. Both the parent and the Indian subsidiary contribute to the profit that the Pharmaceutical Group earns. If the Indian subsidiary is still treated as a routine service provider and remunerated on a cost-plus arm’s length mark-up basis, its remuneration will not be aligned with the actual conduct of the Indian subsidiary. This is the kind of case where PSM should be considered the MAM.

How DEMPE Changed the Analysis

The DEMPE framework has changed how intangibles are analysed. In the pre-BEPS era, the legal owner of an intangible generally retained the residual profits/losses. After the introduction of Action Plans 8-10, this position has changed. The key question now is which entity performs and controls the important functions in the value chain. Take a Software Group where the parent entity is the legal owner of the code. A team of engineers employed in the Indian subsidiary improves the code, a team of researchers in the US subsidiary builds new features, a team of data scientists in Singapore refines the algorithms, and strategic decisions are taken by the European entity. On the facts of this case, should all the profit go to the legal owner? Post-BEPS, the answer is likely no. Where several entities perform DEMPE functions for developing intangibles, it becomes hard to argue that any one of the entities is performing the simpler functions and can be considered the tested party. It may be reiterated that DEMPE performance, control of economically significant risks and funding must be evaluated together.

Integrated Business Models

Integrated Business Models are common in Multinational Enterprises, for example in pharmaceutical R&D, platform businesses, semiconductors, global networks and supply chain hubs. In these businesses, profit is rarely earned by one entity and then passed on to the others. More often, the entities collaborate to generate the return.

Transaction-by-transaction analysis: Aggregating several transactions under a single TNMM analysis does not stop the tax officer from looking at any one of them separately. In Tupperware India Private Limited9, the taxpayer had aggregated its dealings with Group companies and applied TNMM as the MAM. The tax officer accepted the other international transactions but picked out a payment of INR 90 lakh made to a US Group entity for intra-group services and determined its arm’s length price at nil on the ground that there was not enough evidence to satisfy the need-benefit test. The ITAT observed that an agreement and invoices do not, by themselves, prove that services were actually rendered, and sent the matter back to the tax officer to verify the need-benefit test documentation. The case does not involve a PSM, but it does show a limitation of an aggregated, one-sided TNMM approach.

This brings us to a basic question. Can a one-sided TNMM analysis test capture value creation when several entities jointly contribute to the profit? In many audits, taxpayers continue to defend routine returns under TNMM even though the tested party analysis does not deal with the main value drivers. That gap is what has driven the growing interest in PSM.

OECD’s Growing Acceptance of PSM

For a long time, PSM was treated as a last resort, used only when traditional methods were not considered appropriate.10 This changed in 2010, when the OECD dropped the idea of a strict order of methods. It now asks taxpayers to pick the method that best suits the facts of each case.11 The BEPS project went further, saying profits should follow the place where value is created, and not just as per the agreements between Group companies.12 Then in June 2018, the OECD issued revised guidance on the PSM. It says a PSM is likely to be suitable where each party makes unique and valuable contributions, or where the business is so closely integrated that the parties’ contributions can’t be assessed separately.13 Further, the shared assumption of economically significant risks, or separate assumption of closely related risks, are important factors when considering PSM. The guidance also says the method shouldn’t normally be used where one party performs only simpler functions. So, the OECD isn’t saying PSM is the new default. It’s saying the method has become a normal, accepted method when the facts call for it.

Industries Where PSM Is Gaining Ground

The application of PSM is not uniform across industries. It is applied in the following sectors:

Pharmaceuticals: Drug development is among the most integrated activities in any business. Research, clinical trials, regulatory approvals and commercial launch are often spread across several countries. Where each Group entity contributes to a joint research and development effort, a PSM is considered to be a more appropriate method.

Technology: Global technology businesses typically derive their value from algorithms, data, customer networks and intellectual property developed in several countries at the same time. Allocating routine returns to individual entities under TNMM cannot be considered a fair allocation of profits.

The same applies to platform businesses, where network effects are created by users, engineers and marketers in different locations. A captive software centre that grows from back-end coding into end-to-end delivery, as in Infogain India Private Limited, sits on this boundary.

Telecom and media: Global telecom networks and broadcasting Groups earn revenue only because many entities across many countries each perform a part of the work. The Global One India Private Limited and Star Group14 rulings both come from these sectors.

What Are Courts Saying?

The ITAT has decided a fair number of cases on whether PSM is the MAM, as tabulated below:

RulingWho relied on a splitDecision
Global One India P Ltd (Supra)Company (residual split)PSM upheld over TPO’s TNMM
Infogain India P Ltd (Supra)Company (40:60 split)PSM upheld; sent back, 40:60 ratio not to be disturbed on same facts
IndiaNIC Infotech Ltd (ITA No. 1639/Ahd/2024)CompanyPSM upheld by acknowledging that the AE and the Company contributed to generating revenue
Fractal Analytics Pvt Ltd (ITA No. 764/MUM/2026)Company as well as Tax OfficeRPSM is accepted as the MAM for the integrated IT-enabled/data analytics services transaction

The pattern across these rulings is fairly consistent. The ITAT raises a fundamental question as to whether each party makes a unique contribution and whether the operations are integrated. Further, in Global One, the Tribunal distinguished between benchmarking routine returns and allocating residual profits where reliable external comparable information was difficult to obtain. In the case of Infogain, the Tribunal accepted PSM because the Indian and US entities performed interlinked functions and contributed significantly to the value chain. Acceptance of PSM in preceding and succeeding years was also relevant. In cases where the operations are highly integrated, a PSM may be accepted, and in cases where they are not, TNMM is generally considered the MAM.

Is PSM Really the New TNMM?

This is where the debate needs the most care, and there are good arguments on both sides. The answer is not a simple yes or no.

The arguments for PSM are strong. TNMM asks whether one entity earns an arm’s length net profit margin, whereas PSM asks how the combined profits should be divided between the parties based on their respective contributions to value creation. This makes PSM relevant for highly integrated businesses, unique intangibles and cases where reliable external comparables are unavailable. Global One India Private Limited and Infogain India Private Limited show that PSM can succeed in India where a TNMM-based approach falls short.

TNMM will continue to dominate routine services, contract manufacturing and limited-risk distribution. But where a business is highly integrated, relies on unique intangibles or creates significant value jointly, a PSM is likely to be found more appropriate in a growing number of cases.

PSM has its limitations, though. It needs reliable combined financial data and a sound basis for allocating the profit. Sony Pictures Films India Private Limited15 shows that an attribution of percentages without proper support cannot lead to an appropriate conclusion, while Infogain India Private Limited shows that assigning weightages based on evidence and applied consistently over the years can survive scrutiny. Toyota Boshoku Automotive India Private Limited is a reminder that PSM cannot take the place of TNMM for routine transactions where one party performs limited functions and operations are not integrated.

So, is PSM the new TNMM? Not across the board. TNMM will continue to dominate routine services, contract manufacturing and limited-risk distribution. But where a business is highly integrated, relies on unique intangibles or creates significant value jointly, a PSM is likely to be found more appropriate in a growing number of cases.

India’s TP Landscape: The Next Decade

In principle, India’s TP regulations leave room for PSM to play a bigger role. Section 165 of the Income-tax Act, 2025 (“the Act, 2025”), corresponding to Section 92C of the Income-tax Act, 1961 (“the Act, 1961”), does not rank the methods, and Rule 80 of the Income-tax Rules, 2026 (“the Rules, 2026”), corresponding to Rule 10C of the Income-tax Rules, 1962 (“the Rules, 1962”), requires the MAM to be selected having regard to the nature of the transaction, the functions performed, assets used and risks assumed, and the availability of reliable data. The difficulty lies in Rule 79(1)(d) of the Rules, 2026, corresponding to Rule 10B(1)(d) of the Rules, 1962, which refers to reliable external market data for evaluating the relative contribution of each enterprise. This can make PSM hard to apply where there is no external benchmark. Even so, Global One India Private Limited and Infogain India Private Limited show that the ITAT will accept a contribution-based split where the taxpayer lays down a solid factual and analytical foundation.

The wider environment is also changing. Amount B under Pillar One provides a simplified, formula-based return for baseline marketing and distribution activities, which could shift more attention to non-routine profits. India has not adopted Amount B so far, but the direction is clear. Routine returns are increasingly being standardised, and disputes are likely to centre on intangibles, DEMPE and residual value.

Over the next decade, taxpayers should expect closer scrutiny of DEMPE functions, marketing intangibles, location-specific advantages and integrated value chains. The ITAT’s verdict in Tupperware also shows that an individual transaction can be examined separately, even where the taxpayer has applied TNMM on an aggregated basis. Bilateral and multilateral Advance Pricing Agreements (“APAs”) and the Mutual Agreement Procedure (“MAP”) are likely to become more important for complex arrangements, where both tax administrations need to agree on the profit pool and the allocation key.

For taxpayers, the practical takeaway is: do not assume that the Indian entity is either purely routine or automatically a co-owner of the value. The method must follow the factual analysis. Where PSM is appropriate, the analysis should be supported by functional interviews, clearly identified value drivers, well-reasoned weightages and consistency from year to year. For high-value or highly integrated arrangements, it is worth considering an APA to obtain certainty.

Conclusion

The real question is not whether PSM will replace TNMM. It is whether a one-sided analysis is still sufficient when businesses are as closely integrated as they are today.

For routine transactions undertaken by limited-risk entities, TNMM is likely to remain the dominant method and, in many cases, the best one. It brought practicality and consistency to a difficult area of tax, which is why it became the default. However, the business environment in which it earned that position has changed. Multinational Groups now create value through collaborative innovation, data, technology and intangibles developed across borders. In that context, a routine profit for one entity is often considered a one-sided analysis.

In my view, the future of TP will not belong to either method on its own. What will matter is choosing the analysis that best reflects how value is actually created within the Group.

——◆◆——
Author may be reached at eboard@icai.in

Footnotes

  1. OECD/G20 BEPS (2015), Aligning TP Outcomes with Value Creation, Actions 8-10 – 2015 Final Reports, Chapter VI
  2. Paragraph 7.52 of the OECD’s June 2026 consultation document
  3. ITA No. 5571/Del/2011 and ITA No. 5896/Del/2012
  4. ITA No. 6134/Del/2012
  5. ITA No. 1646/Bang/2017 and 2586/Bang/2019
  6. OECD/G20 BEPS (2015), Aligning Transfer Pricing Outcomes with Value Creation, Actions 8-10 – 2015 Final Reports, Chapter VI
  7. Rule 79(1)(e) of the Income-tax Rules, 2026, corresponding to Rule 10B(1)(e) Income-tax Rules, 1962
  8. OECD TP Guidelines for Multinational Enterprises and Tax Administrations (1995), Chapter III, para. 3.49, p. III-16.
  9. ITA No. 2333/Del/2019
  10. OECD, TP Guidelines for Multinational Enterprises and Tax Administrations (1995), Chapter III
  11. OECD, TP Guidelines for Multinational Enterprises and Tax Administrations (2010), Chapter II, Part I.
  12. OECD, Aligning TP Outcomes with Value Creation, Actions 8–10 – 2015 Final Reports, OECD/G20 BEPS Project.
  13. OECD, Revised Guidance on the Application of the Transactional PSM: Inclusive Framework on BEPS – Action 10 (June 2018). It now forms part of Chapter II, Part III, Section C of the OECD Transfer Pricing Guidelines (2022).
  14. ITA No. 8683/Mum/2011
  15. ITA No. 554/MUM/2022

The Chartered Accountant — Theme November 2026 | www.icai.org