Part 1 — Identification and entitlement
A practical guide for tax professionals, considering the OECD Transfer Pricing Guidelines (January 2022). Paragraph references appear in parentheses, e.g. (6.42).
This is the first of three parts. Part 1 deals with the two questions that precede any pricing exercise: what counts as an intangible for transfer pricing purposes, and which entity in a multinational group is entitled to keep the profit an intangible generates. Part 2 turns to pricing — method selection, comparability, and the valuation of intangibles for which no reliable comparables exist. Part 3 addresses the documentation that supports the analysis, and closes with the considerations that matter most in practice.
Section 1 — The issue
A multinational group develops a drug, a brand, a software platform or a manufacturing technique. That asset generates profit in twenty countries. Every one of those twenty tax authorities would like a share of the profit, and each of them believes the share should be larger.
Transfer pricing is the body of rules that decides how the profit is split. Its governing test, the arm’s length principle, asks what independent businesses dealing with each other at commercial arm’s length would have agreed. Where physical goods are concerned that question is usually answerable: you can find out what steel or handsets sell for. Where the asset is a patent portfolio or a brand, there is frequently nothing to compare it to, because the whole commercial point of the asset is that nobody else has one.
That is the intangibles problem. It is now the dominant problem in the field, for the reason set out next.
Section 2 — Why intangibles moved to the centre
Corporate value has shifted decisively from tangible to intangible assets over the past half-century. Where a large listed company’s worth once sat mainly in factories, inventory and land, it now sits mainly in patents, processes, brands, data and know-how.
Two consequences follow.
First, the money is in assets that are hard to locate. A factory has an address. A brand does not — or rather, it has as many addresses as there are people participating in it. When most of the value sat in tangible assets, profit allocation broadly followed physical presence. It no longer does, and no equally simple rule has replaced it.
Second, the money is in assets that are hard to price. A factory has a market value. A unique patent, by definition, does not.
The OECD Transfer Pricing Guidelines (the Guidelines) address this directly. They require the functional analysis in an intangibles case to be grounded in an understanding of the MNE’s global business and how intangibles are used to add or create value across the entire supply chain (6.3) — a deliberately wide instruction, and a warning against analysing a licence agreement in isolation from the business it serves. Also, where an item or activity generates economic value, it should be taken into account in pricing whether or not it meets the definition of an intangible (6.2). So the contention that something is not technically an intangible, and therefore attracts no compensation, does not work. If it is worth something, it gets priced.
Section 3 — What counts as an intangible
The definition
Define an intangible too narrowly, and taxpayers or governments will argue that certain items fall outside the definition and can therefore be transferred or used without any payment — even though independent parties would have charged for them. Define it too broadly, and the opposite argument appears: that payment is due for items independent parties would have handed over for nothing (6.5).
The resolution is a definition that is functional rather than legal or accounting-based:
An intangible is something that is neither a physical nor a financial asset; that is capable of being owned or controlled for use in commercial activities; and that independent parties would have paid for, had they dealt with each other over it in comparable circumstances (6.6).
Note what the test turns on: would independents have paid for it? Not whether it is registered, not whether it is capitalised, not whether it has a name.
Some considerations
Several consequences follow from that definition.
The first is that accounting does not govern. Money spent on R&D and advertising is usually expensed rather than capitalised, so the resulting intangibles never appear on the balance sheet, yet they may generate very significant economic value and must still be considered for transfer pricing (6.7). The practical consequence is that a group whose accounts show almost no intangible assets may nonetheless have a large intangiblesexposure, and the accounts should not be allowed to set the scope of the review.
The second is that legal protection is not required. The availability of legal or contractual protection may affect an item’s value, but it is not a condition of the item being an intangible; neither is separate transferability, so something that can only move as part of a larger bundle still qualifies (6.8). Unregistered know-how, trade secrets, customer relationships and internal processes are therefore all squarely in scope — as is the negotiationknow-how of a procurement team.
The third is that market features are not intangibles at all. Local market characteristics such as household disposable income, market size and relative competitiveness are not capable of being owned or controlled, and so fall outside the definition entirely (6.9); they are dealt with through the comparability analysis instead (Chapter I, Section D.6).
The fourth is that not every intangible earns a premium. Not all intangibles deserve compensation separate from the payment for goods or services, and not all give rise to premium returns (6.10). Not all R&D expenditure produces an intangible, and not all marketing activity creates one (6.11). Identifying an intangible is therefore the beginning of the analysis rather than the end of it; the questions that follow are whether it is unique, and whether it is actually driving the profit.
The requirement that does the work: specificity
The functional analysis must identify the relevant intangibles with specificity — what they are, how they contribute to value in the transaction under review, the important functions performed and specific risks assumed in relation to them, and how they interact with other intangibles, with tangible assets and with the business (6.12).
One label is defined, and it matters because it drives method selection later. “Unique and valuable” intangibles are those that (i) are not comparable to intangibles used by or available to parties in potentially comparable transactions, and (ii) whose use is expected to yield greater future economic benefits than would be expected without them (6.17). Both must be satisfied and it is a conclusion to be reached on evidence.
The catalogue
An illustrative list:
| Category | Notes |
| Patents (6.19) | Exclusive right, limited time, specific geography. May cover a process, not just an object. Note that small R&D spend can yield very valuablepatents — cost is a poor guide to value. |
| Know-how and trade secrets (6.20) | Proprietary information assisting commercial activity, not registered, protected by confidentiality rather than by statute. Usually the largest category by value and the least documented. |
| Trademarks, trade namesand brands (6.21–6.23) | A “brand” may be a combination of trademarks, trade names, customer relationships, reputational characteristics and goodwill, which may be difficult or impossible to separate. |
| Government licences and concessions (6.24) | Spectrum licences, resource rights, permissions to conduct a business. But routine company-registration obligations are not intangibles. |
| Rights under contracts(6.25) | Supplier contracts, key customer contracts, agreements making employees available. |
| Limited rights (licences)(6.26) | A licence limited by field of use, term or geography is itself an intangible. |
Goodwill
The Guidelines do not define goodwill precisely (6.28). What they insist on is the practical consequence: a significant part of what independent parties pay when a business is transferred may represent goodwill or ongoing concern value, and labelling a contribution “goodwill” does not render it non-compensable (6.28). Where reputational value is transferred or shared alongside a trademark, it must be taken into account in setting the price (6.28).
Two further points. First, accounting or valuation measures of goodwill — typically the residual after allocating purchase price — do not, as a general rule, correspond to an arm’s length price (6.29). Second, attempts to separate a trademark artificially from the reputational value factually associated with it should be identified and critically analysed (6.96).
Section 4 — Who is entitled to the return
This is the core of the subject. The question is not “who owns the intangible?” but “who in the group is entitled to keep the profit it generates?” — and the Guidelines are emphatic that these are different questions.
The framework
Chapter VI sets out a six-step framework for assessment of the asset (6.34):
- Identify the intangibles used or transferred, with specificity, together with the specific, economically significant risks associated with their development, enhancement, maintenance, protection and exploitation.
- Identify the full contractual arrangements — registrations, licences, other contracts and other indicia of legal ownership — including who contractually assumes which risks.
- Identify who actually does what, by functional analysis: who performs functions (including the important functions), uses assets and manages risks; and in particular who controls any outsourced functions and who controls each economically significant risk.
- Confirm that contract and conduct match. Test whether the party contractually assuming a risk actually controls it and has the financial capacity to bear it.
- Delineate the actual transactions in light of legal ownership, contracts and conduct.
- Price them, consistently with each party’s contributions.
DEMPE, in plain terms
DEMPE stands for Development, Enhancement, Maintenance, Protection and Exploitation — the five activities that create and sustain intangible value. The concept exists to answer a simple question: value in a group is created by people doing things, so which people, in which entities, are doing the things that make this intangible worth what it is worth?
- Development — creating it.
- Enhancement — improving it.
- Maintenance — keeping it working and current.
- Protection — defending it against infringement, keeping registrations alive, enforcing confidentiality.
- Exploitation — putting it to commercial use, including licensing it out.
The point of DEMPE is that entitlement follows these activities rather than following title.
What legal ownership does and does not buy
Legal ownership is the starting point of the analysis and a necessary reference point for identifying the transactions. It is not the answer. In the Guidelines’ own words, legal ownership by itself does not confer any right ultimately to retain returns derived from exploiting the intangible, even though those returns may initially accrue to the legal owner (6.42). What the legal owner ultimately keeps depends on the functions it performs, the assets it uses, the risks it assumes, and on what the other group members contribute (6.42–6.43).
At the extreme, a legal owner that performs no relevant functions, uses no relevant assets and assumes no relevant risks — a pure title-holding entity — is entitled to nothing beyond arm’s length compensation for holding title (6.42; 6.54). Conversely, where the legal owner in substance performs and controls all DEMPE functions, provides all assets including funding, and assumes all DEMPE risks, it is entitled to all of the anticipated ex ante returns (6.71).
Between those extremes of the spectrum, entitlement follows contributions weighted by control, on a clear gradient (6.55). Not all functions weigh equally. Certain important functions carry special significance: design and control of research and marketing programmes; direction of and priority-setting for creative work, including determining the course of research; control over strategic decisions on development programmes; management and control of budgets; important decisions on defence and protection of the intangibles; and ongoing quality control over work performed by others where that work materially affects value (6.56).
Section 5 — Where this leads
Identification is an exercise in specificity, not in cataloguing. The definition at 6.6 is functional, and the requirement at 6.12 is that each relevant intangible be identified individually, with its contribution to value articulated. A review that begins and ends with the IP register will miss the unregistered know-how, the internal processes and the contract rights that often carry more value than the registered rights do — and the balance sheet will not correct the omission, because the intangibles that matter most are frequently the ones that were expensed rather than capitalised.
Entitlement is an exercise in evidence about people. Legal ownership identifies the transactions to be analysed; it does not determine who keeps the return. That is settled by who performs and controls the DEMPE functions, who assumes the risks, and whether conduct actually matches what the contracts say. The practical test is whether the important functions at 6.56 can be attributed to named individuals who genuinely make the decisions in question. If they cannot be named, the position is difficult to defend.
Neither exercise prices anything. What identification and entitlement establish is which intangibles are in play and which entities have a claim on the return; how large that claim is turns on method selection, comparability and — where no reliable comparables exist — valuation technique. That is the subject of Part 2. It is worth noting that step six is itself subject to a narrow exception: in exceptional cases, where an arrangement lacks the commercial rationality independent parties would display, the transaction as delineated may be disregarded and replaced rather than priced at all. That threshold is high, and the burden of establishing it rests with the tax administration.
Part 2 turns to the pricing itself — method selection, comparability, valuation technique and hard-to-value intangibles. Part 3 completes the series with documentation and the key practical considerations.
For more information: Paul.valdivieso@basefirma.com