Selecting the Right Transfer Pricing Method — Why It Is Not a Default Choice
Article 7 of Saudi Arabia’s TP Bylaws approves five transfer pricing methods for testing the arm’s length nature of controlled transactions. The goal is always the same: find the method that produces the most reliable arm’s length result for the specific transaction. There is no hierarchy that mechanically applies to all cases — and there is certainly no default to TNMM because it is familiar or convenient.
In practice, method selection is one of the most consequential decisions in any TP analysis. Choose the wrong method — or choose the right method for the wrong reasons — and the economic analysis that follows may produce an unreliable or unjustifiable arm’s length range. ZATCA has the power to challenge both the method selection and the resulting pricing, particularly where the documentation does not adequately explain why the chosen method is the most appropriate for the facts.
This article explains each of the five approved methods, when they work, when they are inappropriate, and provides worked examples using Saudi business scenarios.
No method is correct or incorrect in the abstract. The descriptions below indicate where each method is typically reliable — but the correct method for any specific transaction requires a full functional and comparability analysis first. Do not select a method without having completed that analysis. The method flows from the facts; it does not substitute for them.
The Selection Framework — Traditional Methods First
ZATCA’s Guidelines, consistent with OECD principles, hold that traditional transaction methods are generally the most direct means of establishing whether the conditions in controlled transactions are arm’s length. The reason is intuitive: a direct price or gross margin comparison to an independent transaction gets closer to the actual conditions of the deal than a net margin comparison that may aggregate many different cost and revenue elements.
Traditional transaction methods (CUP, RPM, Cost Plus) are therefore preferred over transactional profit methods (TNMM, PSM) — but only where both can be applied with equal reliability. In practice, traditional methods often cannot be applied reliably because comparable gross margin or price data is unavailable or too different from the controlled transaction to produce meaningful results. That is when profit methods become appropriate.
The practical reality in Saudi Arabia — as in most jurisdictions — is that TNMM is the most frequently applied method, largely because net margin data from commercial databases is more readily available than transaction-level price or gross margin data. This is not a problem if the TNMM is properly applied. It becomes a problem when it is applied as a shortcut, without a genuine assessment of whether traditional methods could have been used.
The Five Approved Methods — Explained with Saudi Examples
Method 1: Comparable Uncontrolled Price (CUP)
The CUP method compares the price charged in a controlled transaction to the price charged in a comparable uncontrolled transaction. It is the most direct application of the arm’s length principle — when it can be applied, it provides the most reliable result.
Two types of CUP exist. An internal CUP applies where the taxpayer itself transacts with independent parties in comparable conditions — the entity sells the same product to both related and unrelated customers. An external CUP uses price data from comparable transactions between independent third parties.
When CUP Works
CUP is reliable where the controlled transaction involves identical or near-identical products or services, comparable contract terms, and comparable economic circumstances. It is most commonly applied to commodity transactions, intercompany loans (where the interest rate is compared to market rates for loans with comparable terms, credit risk, and currency), and situations where the same product is sold to both related and unrelated parties.
Al-Farid Chemicals Co. is a Dammam-based manufacturer that produces a standard industrial solvent. It sells 80,000 litres per year to its French parent at SAR 42 per litre (controlled transaction). It also sells the same solvent to two independent customers — a Saudi petrochemical company and a Bahraini distributor — at prices of SAR 44 and SAR 43 per litre respectively under comparable volume and payment terms.
The internal CUP range is SAR 43–44 per litre. The controlled transaction price of SAR 42 is below the arm’s length range. Al-Farid’s TP documentation should address whether contractual differences between the related-party transaction (e.g., volume commitment or warranty terms) justify the lower price, or whether a pricing adjustment is needed to bring the controlled transaction within the range.
Method 2: Resale Price Method (RPM)
The RPM starts from the price at which a product purchased from a related party is resold to an independent customer. It deducts an appropriate gross margin — benchmarked to comparable independent distributors — to arrive at the arm’s length purchase price from the related party.
The RPM focuses on the gross margin earned by the reseller (the tested party). It works best when the reseller adds limited value before resale — when it distributes without significantly processing or transforming the product. The less the reseller’s functions differ from those of comparable independent distributors, the more reliable the RPM result.
Al-Manar Consumer Goods Co. is a Riyadh-based exclusive distributor of imported consumer electronics, 100% owned by a South Korean parent. Al-Manar purchases finished products from the Korean parent and resells them to Saudi retailers. Al-Manar performs standard distribution functions — warehousing, sales, marketing — and bears typical distributor risks. It holds no proprietary intangibles.
Al-Manar’s revenues are SAR 60 million. COGS (purchases from the Korean parent) are SAR 48 million. Gross margin is 20%.
A benchmark search identifies comparable independent distributors of consumer electronics in comparable markets. The interquartile range of gross margins for comparables is 18%–24%, with a median of 21%. Al-Manar’s gross margin of 20% falls within the arm’s length range — the transfer price (SAR 48 million in purchases) is supportable.
Method 3: Cost Plus Method (C+)
The Cost Plus method starts from the costs incurred by the supplier of goods or services in a controlled transaction, then adds an appropriate gross mark-up to arrive at the arm’s length price. It focuses on the gross margin earned by the supplier (the tested party).
C+ works best for contract manufacturing and intercompany service provision where the supplier performs routine functions, does not own unique intangibles, and where the cost base is reliable and consistently defined. The key challenge is ensuring that “cost” is defined consistently between the controlled and comparable transactions — different cost accounting policies can make gross mark-up comparisons unreliable without adjustment.
Al-Baraka Precision Parts Co. is a Jubail-based contract manufacturer, wholly owned by a German MNE. It produces components to the parent’s specifications, using parent-supplied tooling and designs. Al-Baraka bears no inventory risk beyond work in progress and does not own any significant intangibles. Its total cost of production for the year is SAR 35 million.
A benchmark search for comparable contract manufacturers identifies an arm’s length gross mark-up range of 8%–14%, with a median of 10%. Al-Baraka charges the German parent SAR 38.5 million — a gross mark-up of 10% on total costs. This falls at the median of the arm’s length range and is supportable under Cost Plus.
Method 4: Transactional Net Margin Method (TNMM)
The TNMM examines the net profit margin — relative to an appropriate base such as costs, sales, or assets — that a taxpayer earns from a controlled transaction. It compares that net margin to the net margins earned by comparable independent parties in comparable transactions.
TNMM is applied to one side of the transaction — the tested party. The tested party is generally the less complex entity: the one that performs more routine functions, owns fewer unique assets, and bears fewer risks. In Saudi Arabia, this is frequently the Saudi entity itself — particularly Saudi distributors or manufacturers operating within an MNE group where the parent owns the key intangibles.
The profit level indicator (PLI) — the metric against which net margin is measured — must be selected carefully. Return on Sales (operating margin) is common for distributors; Full Cost Mark-Up is used for service providers; Return on Assets or Return on Capital Employed may be appropriate for asset-intensive manufacturers.
Al-Nour Trading Co. is a Riyadh-based distributor, 60% owned by a UAE parent. It distributes imported construction materials purchased from the UAE parent to Saudi contractors. Al-Nour performs standard distribution functions, holds no proprietary intangibles, and bears routine commercial risks. The UAE parent owns all brand rights and bears market development risk.
Al-Nour’s revenues are SAR 45 million. Operating costs (COGS + operating expenses) are SAR 43.5 million. Operating profit is SAR 1.5 million — a return on sales (ROS) of 3.3%.
A benchmark search identifies comparable independent distributors of construction materials. The interquartile range of operating margins is 2.5%–5.5%, with a median of 3.8%. Al-Nour’s ROS of 3.3% is within the arm’s length range and is defensible.
Method 5: Profit Split Method (PSM)
The PSM identifies the combined profit from a controlled transaction and then splits that profit between the related parties on an economically valid basis — one that approximates how independent parties would have divided it given their respective contributions. It is the only method that analyses both sides of the transaction rather than testing one party against external benchmarks.
PSM is typically the most complex method to apply. It is most appropriate where both parties make unique, valuable contributions to the transaction — where neither party can be cleanly characterised as the “routine” tested party. Integrated global value chains, joint development of unique intangibles, and highly interdependent business models are the most common scenarios where PSM is appropriate.
Two approaches exist. Contribution analysis splits combined profits based on the relative value of each party’s contribution, measured by functions, assets, and risks. Residual analysis first allocates routine returns to each party using other TP methods, then splits the residual profit — the return attributable to unique, valuable contributions — based on the relative value of those contributions.
A Saudi joint venture — 60% owned by a German technology company, 40% Saudi-owned — jointly develops and exploits a unique industrial process that both parties contributed to creating. The Saudi entity contributed proprietary knowledge of regional industrial conditions and regulatory relationships. The German entity contributed advanced engineering technology and global customer access. Neither party can be characterised as purely routine.
Combined operating profit from the venture: SAR 80 million. Step 1 — routine returns: benchmark analysis establishes that routine returns for comparable functions (absent the unique contributions) would be approximately SAR 12 million for the Saudi entity and SAR 18 million for the German entity. Step 2 — residual profit: SAR 80M − SAR 30M = SAR 50 million residual. The residual is split based on the relative value of each party’s unique contribution — determined by relative R&D expenditure, headcount, and agreed commercial terms. The PSM documentation must fully explain and justify the splitting factors used.
Method Comparison — Quick Reference
| Method | Category | Tested Metric | Best Applied To | Key Limitation |
|---|---|---|---|---|
| CUP | Traditional | Transaction price | Commodities, financial transactions, identical products with internal comparables | Requires high degree of comparability; adjustments for minor differences are difficult |
| RPM | Traditional | Gross margin of reseller | Routine distributors adding limited value before resale | Gross margin comparisons are sensitive to accounting policy differences; unreliable if reseller significantly processes or transforms product |
| Cost Plus | Traditional | Gross mark-up on cost | Contract manufacturing, routine service provision | Cost definition must be consistent; inappropriate where supplier owns significant intangibles |
| TNMM | Profit | Net margin (various PLIs) | Routine distributors, manufacturers, service providers — one-sided analysis | Reliable only where tested party is the less complex party; net margin affected by factors unrelated to transfer pricing |
| PSM | Profit | Combined profit split | Integrated transactions with unique contributions from both parties; no reliable one-sided comparables | Most data-intensive; splitting factors require careful justification; limited external data available |
Frequently Asked Questions
Can I apply more than one transfer pricing method to the same transaction?
Yes. ZATCA’s Guidelines permit the use of more than one method as a cross-check, though the primary analysis should identify one method as the most appropriate. Using a second method — particularly TNMM as a corroboration for a traditional method result — can strengthen the documentation and demonstrate robustness. The documentation must explain which method is primary and why, and how the cross-check was conducted.
Is TNMM always acceptable in Saudi Arabia?
TNMM is acceptable where it is the most appropriate method based on the functional profile of the tested party and the availability of comparable data. It is not automatically acceptable as a default. If a traditional method could have been applied reliably but was not, ZATCA can challenge the method selection. The documentation must explain why TNMM was selected over traditional methods, not simply state that TNMM was applied.
What is the “tested party” in TNMM?
The tested party is the entity whose net margin is compared against the benchmark. ZATCA’s Guidelines indicate it should be the less complex party — the one that performs more routine functions, owns fewer unique assets, and bears fewer risks. This is usually the entity for which reliable comparable data can most readily be obtained. In most Saudi TP analyses, the Saudi entity is the tested party, though this depends entirely on the functional profile of each party.
When is the Profit Split Method required?
PSM is most appropriate — and may be required — where both parties to a controlled transaction make unique, valuable contributions and neither can be cleanly characterised as the routine tested party. OECD guidance (reflected in ZATCA’s approach) particularly supports PSM for highly integrated transactions, joint intangible development, and situations where the TNMM cannot be reliably applied because both parties own unique assets. PSM is not selected by preference; it is selected when the facts require it.
- Five methods are approved under Article 7 of the TP Bylaws: CUP, RPM, Cost Plus, TNMM, and PSM. The goal is always to select the most appropriate method for the specific transaction — not the most familiar or convenient one.
- Traditional transaction methods (CUP, RPM, Cost Plus) are generally preferred over profit methods. They provide a more direct comparison to actual market conditions. Use them where they can be reliably applied.
- TNMM is the most widely applied method in Saudi Arabia, but it must be justified — not assumed. The documentation must explain why TNMM is the most appropriate method given the functional profile of the tested party and the available data.
- PSM is reserved for transactions where both parties make unique, valuable contributions and no reliable one-sided benchmark exists. It is the most data-intensive method and requires careful documentation of the splitting factors.
- Method selection must flow from the functional analysis. The functional profile — functions, assets, risks — determines which method produces the most reliable arm’s length result. Never select a method before completing the functional analysis.
Leave a Reply