Loading content...
Loading...
Loading...
Tested Party — Tested Party is the entity in a controlled transaction whose financial results are compared against independent comparables to determine arm's length pricing.
Tested Party is the entity in a controlled transaction whose financial results are compared against independent comparables to determine arm's length pricing. Under the Transactional Net Margin Method (TNMM) and Comparable Profits Method (CPM), you select one side of the transaction—typically the less complex entity—and benchmark its profitability against similar independent companies.
The tested party is always a participant in the controlled transaction, not an external comparable. The comparable companies form the benchmark set against which the tested party's results are evaluated.
The OECD Transfer Pricing Guidelines (2022) address tested party selection in Chapter II and Chapter III. The Guidelines indicate at that when applying the transactional net margin method, the tested party should be the entity for which reliable data can most readily be obtained—generally the one that is least complex and does not possess unique intangible assets.
US Treasury Regulations §1.482-5(b)(2) similarly require identification of the tested party as the participant in the controlled transaction whose operating profit can be verified using the most reliable data.
Both authorities emphasize selecting the party for which the most reliable comparables can be found—typically the entity with simpler, more routine functions.
Tested party selection is a critical decision that shapes the entire benchmarking analysis. The fundamental principle: select the less complex entity that doesn't own unique intangibles.
Typical Tested Parties:
| Entity Type | Why Commonly Tested | Benchmark Against |
|---|---|---|
| Limited-risk distributor | Routine functions, no unique IP | Independent distributors |
| Contract manufacturer | Produces to spec, doesn't own IP | Independent toll/contract manufacturers |
| Service provider | Performs defined services | Independent service companies |
| Licensee | Uses licensed IP, no development | Independent licensees (if available) |
Entities Rarely Selected as Tested Party:
| Entity Type | Why Not Tested |
|---|---|
| IP owner/developer | Unique intangibles, no comparables |
| Entrepreneur entity | Bears residual risk, complex functions |
| Principal | Makes strategic decisions, owns brand |
Selection Principle: Ask "For which party can I find reliable independent comparables?" The tested party should have functions that exist independently in the market—distribution, manufacturing, services. Entities with unique assets or strategic functions have no market comparables.
Transaction: US parent licenses technology to its German subsidiary, which manufactures and sells products in Europe.
| Entity | Functions | Assets | Risks |
|---|---|---|---|
| US Parent | R&D, product design, brand ownership | Patents, trademarks, know-how | Development risk, market risk |
| German Sub | Manufacturing, quality control, logistics | Production equipment (routine) | Production risk (limited) |
Tested Party Selection:
The German subsidiary is the tested party because:
The US Parent cannot be the tested party because its unique technology and brand have no independent comparables.
Analysis: Benchmark the German subsidiary's profitability (e.g., Net Cost Plus) against independent contract manufacturers. If the German sub's results fall within the arm's length range, the royalty/pricing is appropriate.
When both parties in a transaction could potentially be tested, apply these criteria:
| Criterion | Prefer as Tested Party |
|---|---|
| Functional complexity | Less complex party |
| Unique intangibles | Party without unique IP |
| Data availability | Party with better comparable data |
| Financial segmentation | Party with clean segmented data |
| Location | Party in jurisdiction with database coverage |
Common Mistake: Don't select the tested party based on which produces a more favorable result. Selection must be based on functional analysis and data reliability, not outcome. Tax authorities will challenge opportunistic tested party selection.
Occasionally, both parties have unique functions or intangibles—for example, a cost-sharing arrangement between two R&D entities. In these cases:
The less complex party performs routine functions that exist independently in the market—distribution, contract manufacturing, support services. These functions have independent comparables. The more complex party typically owns unique intangibles (patents, brands, know-how) or performs entrepreneurial functions that don't exist in arm's length transactions. You can't benchmark something that has no market equivalent.
Generally no. If an entity owns unique intangibles, you likely can't find comparable independent companies because those intangibles create value that isn't replicable. However, if the intangibles are routine (standard off-the-shelf software, non-unique processes), the entity might still be benchmarkable. The key question: are there independent companies performing similar functions with similar assets?
Consider alternative approaches: (1) use Profit Split Method which doesn't require external comparables, (2) use CUP method if comparable transactions exist, (3) apply adjustments to imperfect comparables with thorough documentation. If truly no comparables exist, it may indicate the transaction structure is unusual and warrants reconsideration.
Not necessarily. You benchmark the less complex party regardless of location. If the US is auditing and the tested party is the German subsidiary, you still test the German entity. However, some tax authorities (notably India) prefer the local entity as tested party when possible. Know your jurisdiction's preferences.
Only with valid business reasons (restructuring, function changes). Switching tested parties opportunistically to achieve favorable results will attract scrutiny. Consistency demonstrates the robustness of your analysis. If you must change, document the business rationale thoroughly.
The tested party's functions, assets, and risks determine the appropriate PLI. A tested distributor typically uses Operating Margin (revenue-driven). A tested manufacturer might use Net Cost Plus (cost-driven) or ROA (asset-driven). The PLI should capture the tested party's primary value driver. See PLI Selection Guide for detailed guidance.
Segment the tested party's financials to isolate the controlled transaction under review. If the German subsidiary both manufactures for the US parent and sells third-party products, segment manufacturing results from distribution results. Only benchmark the portion relevant to the intercompany transaction. Reliable segmentation is essential—without it, your analysis mixes controlled and uncontrolled activities.