- Hitech City, Hyderabad
- Office Hours: 10:30 am - 7:30 pm
India Union Budget 2026-27
Home > How Do You Defend a Management Fee in an Indian Transfer Pricing Audit?
Written by Sudheer Polana, Chartered Accountant · August 2026 · Statutory references current to the Income-tax Rules 2026.
A service agreement and an invoice are not evidence that a service was rendered. To survive a challenge, you need to establish three separate things: a real service existed, an independent enterprise would have rationally paid for it or performed it itself, and the amount charged was computed on an arm’s-length basis. A Transfer Pricing Officer cannot price the fee at nil merely because profit didn’t visibly rise – but a generic agreement and a year-end invoice bundle won’t survive scrutiny either.
Management fees remain among the most frequently challenged cross-border payments in Indian transfer pricing, and the dispute usually follows the same pattern. The taxpayer produces a service agreement, invoices and a broad description of support received. The Transfer Pricing Officer asks for proof of actual rendition, questions the benefit, flags possible duplication or shareholder activity, and determines the arm’s-length price at nil. Both sides can overreach in this exchange, and the way through it is to keep three separate legal questions from collapsing into one.
What exactly do you need to prove to defend a management fee?
Three distinct things, not one blended argument: that the service existed, that an independent enterprise would rationally have paid for it or performed it itself, and that the amount charged was arm’s length. Evidence is not a procedural afterthought here – it defines the transaction that actually gets benchmarked, so a vaguely described service cannot be evaluated for an arm’s-length charge no matter how strong the pricing analysis behind it is.
Can a TPO price a management fee at nil just because there was no visible profit increase?
No. In CIT v. EKL Appliances Ltd., the Delhi High Court held that tax authorities should ordinarily examine the transaction as actually undertaken and should not substitute their own view of how the business ought to have been run. Losses, the absence of immediate profit, or the availability of internal employees cannot, by themselves, establish that the arm’s-length price is nil.
Does that mean an agreement and an invoice are always enough?
No – that’s the opposite failure mode. CIT v. Cushman & Wakefield India Pvt. Ltd. draws the line the other way: the TPO determines the arm’s-length price under the prescribed transfer pricing methods, while whether an expenditure is deductible for business purposes is a separate question for the assessing authority. Method selection, proof of receipt and deductibility should not collapse into a single subjective “benefit test” – but the taxpayer still carries the practical burden of putting credible material on record.
What does a defensible evidence file actually contain?
Five layers, each establishing something different.
| Evidence layer | What it should establish |
|---|---|
|
Service architecture |
Agreement, service catalogue, responsible teams, request process, deliverables, pricing clause and termination rights |
|
Actual rendition |
Dated emails, tickets, reports, presentations, meeting records, system logs, advice notes and identifiable work products
|
|
Indian benefit
|
The local decision, process, risk or capability supported; why the activity was useful when performed, not merely its eventual outcome |
|
Cost integrity |
Provider cost centres, employee roles, cost-pool bridge, exclusions, allocation keys, recipient universe and reconciliation to books |
|
Arm’s-length price |
Method selection, internal or external comparables, treatment of pass-through costs, mark-up support and tested-party logic
|
What’s the strongest kind of evidence to rely on?
Whatever was created in the ordinary course of business, not assembled after a notice arrives. A monthly cyber-risk report used by the Indian IT head, a tax position discussed with the local finance team, an ERP ticket resolved for the Indian entity, or a recruitment framework adopted locally is stronger than a generic slide deck put together after the fact. Volume doesn’t substitute for quality either – hundreds of emails that merely copy the Indian team prove little. Each service category needs a short evidence narrative: what was requested or provided, by whom, when, what local function it supported, and how the charge reached India. A representative sample can be adequate, but only when it’s linked to a complete service register with an explained selection basis.
Is every head-office cost chargeable to the Indian subsidiary?
No. Costs incurred solely because the parent owns an investment – parent-company shareholder meetings, consolidation done only for the parent’s own reporting, investor relations, or acquisition of the parent’s own interest – should ordinarily stay with the shareholder rather than being charged down to India.
Does having a local finance, HR or IT team automatically mean group support duplicates it?
No. The enquiry is functional, not structural: does the overseas team perform the same activity for the same purpose, or does it provide specialist capability, global coordination, systems access or surge capacity that the Indian team doesn’t have? The file needs to explain that distinction rather than simply assert that no duplication exists.
Is “belonging to a reputable group” itself a chargeable benefit?
No. Incidental benefit – an Indian subsidiary benefiting from group reputation or from a policy designed for the group as a whole – is not a chargeable service. A charge becomes defensible when a specific activity is performed for the Indian recipient and its business position is expected to improve, or a relevant risk or cost is addressed.
Where do most management fee cases actually fail?
In the cost pool, even after rendition is established. The provider needs to identify the personnel and third-party costs included, exclude shareholder and duplicate activities, and separate pass-through items. Allocation keys have to reflect the actual benefit driver – headcount for HR support, user count for software, transaction volume for processing, revenue for certain commercial support – and a single revenue key applied across every service category is easy to administer but hard to defend. The mark-up should also attach only to value-adding costs: third-party licences or agency-type expenses may need cost-to-cost treatment if the service provider performs no meaningful function or takes no risk in relation to them, and a routine 5% mark-up is not automatically arm’s length for every management service.
Is there a safe harbour for low-value management services?
Yes, but it’s tightly conditioned. The Income-tax Rules, 2026 provide a specific safe harbour for an Indian taxpayer receiving low value-adding intra-group services, where the aggregate amount – including a mark-up not exceeding 5% – does not exceed ₹10 crore. An accountant must certify the cost-pooling method, exclusion of shareholder and duplicate costs, and the reasonableness of the allocation keys used.
The definition is deliberately narrow: the services must be supportive, sit outside the group’s core and economically significant activities, must not use or create unique and valuable intangibles, and must not involve significant risk. R&D, manufacturing, software development, KPO, BPO and other specified activities are excluded – calling a charge “management fee” or “low value” does not by itself secure eligibility. The certification requirement effectively converts the cost pool and allocation keys from supporting schedules into the centrepiece of the safe-harbour file.
How do you build a year-round control framework instead of a year-end scramble?
Seven habits, maintained continuously rather than reconstructed at audit time:
- Maintain a service catalogue with local owners and expected deliverables for each service stream.
- Create a quarterly evidence pack while records and business context are still fresh.
- Map provider personnel and cost centres to services; remove shareholder, duplicate and non-beneficial activities before allocation.
- Use service-specific allocation keys and document why each key approximates expected benefit.
- Separate pass-through costs and benchmark the mark-up only on the value-adding service element.
- Reconcile the allocation schedule to the provider’s accounts, the Indian ledger, invoices, withholding tax records and the transfer pricing report.
- Evaluate the ₹10 crore low-value services safe harbour before filing, but don’t force excluded or high-value services into it.
Who should review your management fee structure?
Someone who can build the evidence narrative before a notice arrives, not after. Management-fee litigation is rarely won by citing a single judgment or presenting a longer agreement – it’s won by telling a coherent, evidenced story that connects the service provider’s work to an Indian business need, and connects the charge to a reliable cost and pricing mechanism. Ask the transfer pricing team to pressure-test your current evidence file against these five layers before the next audit cycle.
Frequently Asked Questions
Can a TPO price a management fee at nil just because the Indian entity had local staff doing similar work? Not automatically. The test is functional, not structural – whether the overseas activity performs the same function for the same purpose, or supplies specialist capability, global coordination or capacity the local team doesn’t have. The presence of a local team is a starting point for the enquiry, not a conclusion.
Does the low-value services safe harbour cover R&D or software development fees? No. The definition specifically excludes R&D, manufacturing, software development, KPO, BPO and other specified activities, regardless of how the charge is labelled. Only genuinely supportive services outside the group’s core and economically significant activities qualify.
What happens if the cost pool includes shareholder activities? Shareholder costs – parent-company shareholder meetings, consolidation for the parent’s own reporting, investor relations and similar items – should be excluded from the chargeable cost pool. Including them weakens both the arm’s-length pricing analysis and, where relevant, eligibility for the low-value services safe harbour.
Is a 5% mark-up always considered arm’s length for management services? No. A routine 5% mark-up is not automatically arm’s length for every management service outside the specific low-value services safe harbour. The nature of the activity, available comparables and the method selected still matter.
Can pass-through costs like third-party licences carry a mark-up? Generally not where the service provider performs no meaningful function or assumes no risk in relation to those costs – they may require cost-to-cost treatment instead. The mark-up should attach only to the value-adding element of the service.
What’s the difference between proving a service existed and proving its price was arm’s length? They’re separate questions decided on separate evidence. Rendition and benefit are established through service architecture and actual-rendition records; arm’s-length pricing is established separately through method selection, comparables and cost-pool integrity. A strong case on one doesn’t substitute for the other.
Sources and legal references
- Income-tax Rules, 2026 – definition and safe harbour for low value-adding intra-group services
- CIT v. EKL Appliances Ltd., (2012) 345 ITR 241 (Delhi High Court)
- CIT v. Cushman & Wakefield India Pvt. Ltd., (2014) 367 ITR 730 (Delhi High Court)
- OECD Transfer Pricing Guidelines for Multinational Enterprises and Tax Administrations 2022