Selecting the TP method
Example 12 — CUP and limited procurement support
Draft (June 2026) · Annex I · Example 12 · ¶46–50
OECD discussion draft, June 2026 (not final)At a glance
Company S provides limited procurement support to Company A, while Company A keeps the substantive negotiation and risks. Company S benchmarks itself against full-service procurement agents. Are those agents really comparable?
The parties
Suppliers
Third-party manufacturerssupply
Company A
Parent · negotiates, ships, manages relationshipssupport
Company S
Procurement support · analytical onlyWhat happens
- Company A coordinates sourcing from third-party contract manufacturers in low-cost jurisdictions and holds the supplier relationships.
- Company A establishes Company S to provide procurement support — supplier-selection analytics and review of contract terms.
- Company S does not interact with suppliers, does not hold title to goods, and does not manage the supply chain.
- Company S benchmarks its fee under the CUP method against independent sourcing agents who do the full procurement role.
Analysis
Company S only does analysis and contract review. The independent agents it's being compared to negotiate, set prices, run the supply chain and bear those risks.
Are full-service agents the right CUP comparable for Company S?
OECD analysis · CUP unreliable
Given Company S's limited operations and the differences in functions and risks compared with independent agents, Company S is not comparable to those agents. The independent-party sourcing contracts cannot be used to reliably apply the CUP method.
Key takeaway
Match comparables to the actual functions, assets, and risks performed by the tested party. A label-level match is not enough — look at what the comparable enterprise actually does and bears.
Related: Chapter II (comparability and method selection) · Example 11 (CUP and geographic differences)