DEMPE rules, do they reflect market reality?

(Published in IDC, Asesor Fiscal, Jurídico y Laboral magazine on September 15, 2023)

Jesús Aldrin Rojas M.

José Augusto Chamorro Gómez

Introduction

In October 2015, the Organization for Economic Cooperation and Development (OECD) published an action plan to combat tax avoidance and profit shifting (the BEPS plan). The BEPS plan contained 15 actions aimed at “realigning taxation with economic substance and value creation while preventing double taxation.” Action No. 8 (intangibles) set out the following objectives:

“Develop rules to prevent base erosion and profit shifting arising from the transfer of intangibles between enterprises belonging to business groups. This will involve: (i) adopting a broad and clearly delineated definition of intangible assets, (ii) ensuring that profits related to the transfer and use of intangibles correspond (rather than being detached) to their value creation, (iii) developing transfer pricing rules or special rules for cases where hard-to-value intangible assets are transferred, and (iv) updating the guidelines on cost contribution arrangements (regarding the development and exploitation of intangible assets).”

The new transfer pricing rules on intangibles have had a significant impact, reshaping the transfer pricing policies of business groups worldwide. And rightly so—the new OECD standard for confirming the arm’s length condition of intercompany transactions requires the identification of:

• The intangible subject to the transaction (if multiple intangibles are transacted, it is necessary to assess whether they should be disaggregated for individual analysis);
• The entity that holds the legal ownership of the intangible;
• Which entities within the multinational group have contributed to the development, enhancement, maintenance, and protection of the intellectual property; and
• The nature of the intercompany transactions involving intangibles, including how these transactions contribute to value creation.

In this document, we analyze the limitations of the OECD’s proposed rules in their theoretical conception and economic foundation to gain perspective on their application.

Who Owns the Rights to Exploit Intangible Assets from a Transfer Pricing Perspective?

As previously discussed, Action 8 of the BEPS plan redefines the application of transfer pricing rules on intangibles by introducing a new concept: “The term intangible is intended to refer to something that is not a physical or financial asset, that can be owned or controlled for commercial use, and whose use or transfer would be compensated if it were conducted in a transaction between independent parties under comparable circumstances.”

In other words, under the transfer pricing regime, it is not enough for an asset to be intangible or to have the ability to control future economic benefits. Instead, it must be demonstrated that, in analogous transactions, the asset would be subject to economic compensation (which is only possible given the economic benefit that the transfer of the asset represents for the parties). This clarification is important because certain intangibles may lack economic value in a specific context of exploitation and, therefore, may not qualify as assets for transfer pricing purposes.

Once the condition of the asset as an intangible is confirmed, the OECD’s analytical framework emphasizes the need to identify the legal owner, as recorded in its respective patent and trademark office (e.g., the United States Patent and Trademark Office or the Mexican Institute for the Protection of Industrial Property – IMPI). However, mere legal ownership is not sufficient to claim exploitation rights over an intangible asset.

Transfer pricing guidelines stress the necessity of identifying, through a functional, asset, and risk analysis, who has contributed to the intangible’s value creation. This entails identifying the development, enhancement, maintenance, protection, or exploitation activities related to the intangible (the DEMPE functions).

The execution of DEMPE functions by related parties other than the legal owner should be compensated as would have been agreed with an independent third party. Otherwise, for transfer pricing purposes, “economic ownership” may be established, leading to various tax implications, such as the loss of royalty expense deductions (why pay a royalty for an intangible asset of which one is a co-owner?) or even the redistribution of income derived from the exploitation of intangible assets (if an entity has contributed to the creation of an intangible asset’s value, should it be entitled to claim a share of the resulting benefits?).

Needless to say, the execution of DEMPE functions may be classified as a “unique and valuable contribution,” which can even influence the selection of specific transfer pricing methods (e.g., profit split methods).

How Are Independent Third Parties Compensated for Executing DEMPE Activities?

For the above reasons, it is crucial to correctly identify who, when, and how contributions to the development of intangible assets have been made. Practical considerations are also relevant. For example, the development of patents in the pharmaceutical industry may take decades and be executed in various stages by different members of a business group through diverse research, development, and clinical trial activities. Similarly, trademarks managed by marketing departments often reach peak value after years or even decades of advertising campaigns in mass media. These distinctions are important because the functional, asset, and risk analysis associated with transfer pricing for intangibles should trace back to the inception of the intangible and document any value contributions (identified through DEMPE functions) throughout the asset’s lifecycle.

Evidence of Third-Party Behavior Concerning DEMPE Activities

Does the execution of DEMPE functions between independent third parties imply co-ownership or even an adjustment to the compensation agreed for the exploitation of the intangible? This question is critical because reading the transfer pricing guidelines might suggest this conclusion, and, in fact, tax authorities often systematically challenge the deduction of expenses related to DEMPE functions when the taxpayer also pays royalties. However, is this stance correct? Is there evidence of independent third-party behavior supporting this approach?

To address this question, we conducted a quantitative economic analysis using data from the specialized database Data Insight (formerly Intangible Spring), yielding 12,345 records of contracts between independent third parties across 51 industries.

Since contract records account for the number of functions assigned to the licensor and the licensee, the analysis was conducted based on the following type of variables:

F= F _licensor-F _licensee

Where:

• F licensor = Number of functions assigned to the licensor.
• F licensee = Number of functions assigned to the licensee.
• F = Difference between the functions performed by the licensor and the licensee.

The variable Fi would capture the impact of the distribution of functions between the two contracting parties to analyze its effect on the pricing of compensation.

The regression model would be as follows:

T_i=α+β(F_i)

Where:

• F licensor = Number of functions assigned to the licensor.
• F licensee = Number of functions assigned to the licensee.
• F = Difference between the functions performed by the licensor and the licensee.

The variable Fi would capture the impact of the distribution of functions between the two contracting parties to analyze its effect on the pricing of compensation.

The regression model would be as follows:

T_i=α+B_MKi(F_MKi)+B_PTi(F_PTi)+B_QCi(F_QCi)+B_RDi(F_RDi)

Where:

FRDi = Difference between the research and development functions performed by the licensor and the licensee in the i-th industry.

FMKi = Difference between the marketing functions performed by the licensor and the licensee in the i-th industry.

FPTi = Difference between the protection functions performed by the licensor and the licensee in the i-th industry.

FQCi = Difference between the quality control functions performed by the licensor and the licensee in the i-th industry.

• En la perspectiva por industria se exigieron tener un mínimo de 100 observaciones para considerarla.
• Por otro lado, se estableció el criterio de considerar el impacto de las variables como significativo si llegan a un nivel de confianza del 95%.

Analysis Criteria

• From an industry perspective, a minimum of 100 observations was required to consider it in the analysis.
• A variable’s impact was considered significant only if it reached a 95% confidence level.

Results

Without differentiating by function type, we found a significant impact of DEMPE functions in only 4 industries out of the 40 that met the minimum requirement of 100 observations. (For confidentiality reasons, industry names are withheld to prevent misuse of the results.)

In the extended version of the model, specific DEMPE-related functions were identified as relevant, particularly Research & Development, Marketing, Quality Control, and Protection.

• Research & Development had a significant impact on compensation in 9 industries.
• Marketing had a significant impact in 6 industries.
• Quality Control had a significant impact in 4 industries.
• Protection had a significant impact in 5 industries out of the 40 industries analyzed.

Conclusions

Based on this information, we can preliminarily conclude (subject to case-by-case review of specific taxpayer situations) that, in general, there is no sufficient evidence to prove that performing DEMPE functions inherently leads to an adjustment in the agreed-upon compensation between independent parties.

This finding is relevant because if tax authorities attempt to disallow royalty deductions based on DEMPE functions, or conversely, reject the deduction of expenses related to DEMPE functions by arguing economic ownership, the situation should be re-evaluated based on the actual conduct of independent third parties, as required by the arm’s length principle.

^{_{1 OECD (2014), Guidance on Transfer Pricing Aspects of Intangibles, OECD/G20 Base Erosion and Profit Shifting Project, OECD Publishing. Develop rules to prevent BEPS by moving intangibles among group members. This will involve (i) adopting a broad and clearly delineated definition of intangibles; (ii) ensuring that profits associated with the transfer and use of intangibles are appropriately allocated in accordance with (rather than divorced from) value creation; (iii) developing transfer pricing rules or special measures for transfers of hard to value intangibles; and (iv) updating the guidance on cost contribution arrangements.}}

 _{^{2 OECD Transfer Pricing Guidelines, 2017, 6.6}}

^{_{3 El criterio no vinculativo 39/ISR/NV provee la siguiente definición de contribuciones únicas y valiosas: “aquellas condiciones o atributos del negocio que generan valor de manera significativa y que implican la expectativa de generar mayores beneficios económicos futuros de los que se esperaría en su ausencia, tales como intangibles creados o utilizados, o factores de comparabilidad que definen alguna ventaja competitiva del negocio, incluyendo las actividades de desarrollo, mejora, mantenimiento, proyección y/o explotación de intangibles”.}}

_{^{4 Incluso, hay que considerar el juicio de atracción 15378/16-17-09-2/1484/18-S2-08-04 en donde las autoridades fiscales niegan la deducción de gastos de propaganda y publicidad a un contribuyente que simultáneamente realizó el pago de regalías (previa reestructura corporativa) en una aparente interpretación de la acción 8 del plan BEPS.}}

_{^{5 Data utilized in this research was sourced from the comprehensive Royalty Rates database provided by Data Insight AI (www.datainsightai.com) in July 2023. As an information services firm specializing in transfer pricing analysis, Data Insight AI offers a suite of databases and an extensive Contracts Search Engine. The firm’s unique approach employs a combination of internal large language models, Natural Language Processing (NLP), and manual processing. This results in enriched, contextual data, such as detailed Development, Enhancement, Maintenance, Protection, and Exploitation (DEMPE) information for each transaction, and developmental stage information for pharmaceuticals and medical devices. Data Insight AI’s enriched data aids professionals in identifying potential comparables with greater precision and ensuring a full understanding of the licensed intangible assets. This combination of depth and breadth in data contextualization not only improves accuracy but also reduces the risk of false negatives and positives, significantly enhancing the transfer pricing analysis process. // Los datos utilizados en esta investigación se obtuvieron de la base de datos integral de tasas de regalías proporcionada por Data Insight AI (www.datainsightai.com) en julio de 2023. Como empresa de servicios de información especializada en análisis de precios de transferencia, Data Insight AI ofrece un conjunto de bases de datos y un Amplio Buscador de Contratos. El enfoque único de la empresa emplea una combinación de modelos internos de lenguaje extenso, procesamiento de lenguaje natural (NLP) y procesamiento manual. Esto da como resultado datos contextuales enriquecidos, como información detallada de Desarrollo, Mejora, Mantenimiento, Protección y Explotación (DEMPE) para cada transacción e información sobre la etapa de desarrollo de productos farmacéuticos y dispositivos médicos. Los datos enriquecidos de Data Insight AI ayudan a los profesionales a identificar comparables potenciales con mayor precisión y garantizan una comprensión completa de los activos intangibles con licencia. Esta combinación de profundidad y amplitud en la contextualización de datos no solo mejora la precisión, sino que también reduce el riesgo de falsos negativos y positivos, mejorando significativamente el proceso de análisis de precios de transferencia.}}