Transfer Pricing Methods: A Practical Guide

The choice of transfer pricing method is the first substantive decision in any Indian benchmarking study. It determines which comparables are relevant, which financial indicators you compute, and how defensible your Accept-Reject matrix is under Rule 10D.

This guide compares the five OECD-recognised methods with a view to practical use in Indian documentation (Form 3CEB and master file).

The five methods at a glance

Method	Type	Best used when	Key indicator
CUP	Traditional	Comparable uncontrolled transactions exist	Price per unit
RPM	Traditional	Tested party is a distributor	Gross margin
Cost Plus	Traditional	Tested party is a manufacturer / service provider	Gross cost plus mark-up
TNMM	Transactional	Comparables can be found at company level	Net profit indicator (PLI)
Profit Split	Transactional	Highly integrated, unique intangibles	Relative profits

Choosing a method in practice

Indian Revenue Authorities generally prefer the method that produces the most reliable result given the facts. In practice:

• CUP is rarely usable in benchmarking because third-party comparable transactions at the same price terms are difficult to identify.
• TNMM is the workhorse of Indian benchmarking studies — it tolerates minor functional differences between the tested party and independent comparables, because it measures net profitability rather than a specific margin.
• Profit Split applies where both parties to an international transaction contribute unique intangibles or where the transaction is so integrated that separate pricing cannot be determined reliably.

The PLI decision under TNMM

Under TNMM you select a profit level indicator (PLI) aligned with the tested party’s functions:

• Operating margin on sales — for trading and distribution activities.
• Return on operating costs — for contract service providers and low-asset-intensity functions.
• Return on assets / capital employed — for capital-intensive or asset-heavy functions.

The PLI must isolate the tested party’s routine contribution and be measurable with reliable data from comparable financial statements (Prowess, Capitaline).

Building the comparability matrix

Once the method and PLI are fixed, the workflow becomes:

1. Map the tested party to the correct NIC code(s).
2. Pull a broad database pool using turnover, financial-year, and related-party screens.
3. Apply quantitative filters (turnover range, persistent losses, related-party revenue share).
4. Apply qualitative screens using the FAR profile and trade descriptions.
5. Run the arm’s-length range on the surviving comparables.

Each rejection — quantitative or qualitative — must carry a documented reason. That is exactly the discipline TP Analytics automates with AI-generated, Rule 10D-compliant justifications for every excluded comparable.