In the world of intellectual property (IP), licensing and transfer pricing, royalty benchmarking plays a crucial role. It helps determine arm’s length royalty rates for the use of intangible assets such as trademarks, patents, copyrights and know-how. However, while the concept sounds straightforward, implementing an effective royalty benchmarking study presents several challenges. This article explores what royalty benchmarking entails and the key issues that can arise in practice.
Royalty benchmarking is the process of identifying and analyzing comparable licensing agreements to determine an appropriate royalty rate for an intangible asset. This is commonly used for:
Benchmarking involves analyzing databases of licensing agreements (e.g., RoyaltyStat, RoyaltySource, ktMINE) to identify “comparables” agreements that are similar in terms of industry, geography, assets and terms and then applying a reasoned analysis to arrive at an arm’s length range or a specific rate.
Let take an example of a U.S. based parent company develops proprietary AI-based recommendation software. It licenses the software to its subsidiaries in Europe and Asia for use in localized e-commerce platforms. To comply with OECD transfer pricing guidelines and local tax laws, the company must determine an arm’s length royalty rate for this software license.
Despite a structured approach, several challenges make royalty benchmarking difficult (especially in the tech industry):
Scarcity of Truly Comparable Agreements
Tech licensing deals are often unique. For example:
Data Quality and Confidentiality
Many tech licensing agreements are:
Intangible Asset Complexity
Tech assets often bundle multiple intangibles. A single license may include:
Valuing the individual components becomes subjective and can lead to transfer pricing disputes.
Rapid Innovation and Obsolescence
The pace of change in tech means:
Geographic and Regulatory Variance
Some jurisdictions (e.g., India, Brazil) impose specific rules on royalty payments or deductibility. In cross-border tech licensing, companies must navigate:
Table: Benchmarking illustrations
Company | IP Licensed | Example Royalty Range | Notes |
U.S. AI firm to EU subsidiary | AI-based recommendation engine | 4%–6% of net sales | Adjusted for exclusivity and technical support included |
Cloud services provider | SaaS platform | 2%–5% of user-based revenue | Freemium model excluded from base |
Consumer tech brand | Trademark and branding | 1%–3% of sales | Applied to branded hardware only |
Note: These are illustrative examples; actual rates depend on facts and circumstances.
To mitigate these issues, companies and advisors should consider:
Some important cases and judicial decisions (in India and abroad) related to royalty benchmarking / reasonable royalty / arm’s‑length royalty
Johnson Controls (India) P. Ltd. vs DCIT (Mumbai, 2020)
Praxair India Pvt Ltd (Bangalore ITAT, 2021)
Royalty benchmarking is not just a regulatory requirement, it’s a vital component of strategic IP management, especially for technology-driven businesses. However, identifying the right comparables and applying consistent, defensible methodologies is challenging in a sector characterized by innovation, disruption and global reach.
For tech companies, the stakes are high. Mispricing can result in audits, penalties, reputational risk and even double taxation. Getting the benchmarking right with detailed documentation and sound economic analysis can offer protection and clarity in a complex regulatory landscape.