Exploring Applicability of Transfer Pricing Provisions on Data-Sharing Transactions
- CCL NLUO
- Mar 31
- 6 min read
Updated: Apr 3
Authors: Tanmay Doneria
Second year student at Rajiv Gandhi National University of Law, Punjab

I. Introduction
Enterprises functioning in the digital realm primarily generate revenue by offering targeted advertising services. Approximately 77% of Google’s revenue is from advertising services. Similarly, Meta generated USD 46 Billion from advertising services in 2024. Such a method of advertising involves personalising ads based on user preferences and delivering them in a ‘targeted’ manner, ensuring higher engagement. The success of an enterprise utilising this business model depends on accumulating user data (their choices and preferences). This economic segment is called the attention economy. It entails treating human attention as a finite commodity, that should be captured to gather information about their preferences which is subsequently utilized to deliver targeted advertisements. Considering the share of the targeted advertising business in the total revenue tech giants, it is evident that it is crucial for these enterprises to capture the attention of the users and gather relevant user data for generating profits.
To gather this user data, enterprises utilize multi-sided business models to collect a vast number of data points, that help them profile individual users and gain insight about their preferences. For example, on the consumer side Meta (Instagram and Facebook) and Google (search engine, YouTube, YouTube Music, YouTube Shorts, maps etc.) offer free-to-use services which facilitate the accumulation of vast amounts of organic (first-hand) user data, which is monetized on the commercial side by offering targeted advertising services to businesses. The more users and preference data an enterprise has the better its ability to tailor its services to capture more attention and subsequently, offer better targeted advertisements. Now to further enhance their data pool enterprises these companies usually enter into data-sharing agreements (“DSA”). DSAs entail a mutual non-exclusive sharing of data without any monetary consideration between two or more entities. Evidently, it is more profitable and competitively prudent for enterprises to not share their data with third-party enterprises i.e., enterprises outside their ecosystem. Therefore, generally, these DSAs are entered between various enterprises within the same group for instance- various services within the Meta group including, Instagram and Facebook, share their data.
As discussed DSAs involve the mutual sharing of data pools between two enterprises, thereby from an economic standpoint these are essentially barter transactions. From a taxation perspective, per se, there is no tax being levied on data or DSAs. However, as the DSAs are being executed between associated enterprises (as they are related to the same parent/holding company) it gives way to concerns regarding the applicability of transfer pricing provisions.
This article will demonstrate the applicability of transfer pricing provisions on barter transactions and consequently, its applicability on DSAs making them subject to taxation. Furthermore, it will also highlight the implications of taxing DSAs on businesses and its impact on potentially addressing competitive concerns in data-driven markets.
II. Transfer Pricing and its Application Thereof on DSAs
The term “transfer price” refers to the price charged by an enterprise in a transaction with its related or associated enterprise. Given the variation in the rate of taxes in various jurisdictions, related enterprises set the transfer price in such a manner that minimizes their tax liability. The practice of shifting the profits of an enterprise from a country where it operates to a tax haven is called “transfer pricing”. In order to prevent related enterprises from engaging in tax evasion, tax authorities utilise the principle of arm’s length price (ALP). ALP refers to the price of the transaction between two unrelated parties, transfer pricing provisions mandate that related party transactions must be conducted at ALP and tax is levied on the same. If the parties fail to abide by the ALP principle, tax authorities utilise various methodologies to arrive at the ALP and adequately tax the transaction.
DSAs are usually in the nature of barter transactions as they involve a mutual exchange of data. Herein, it is important to note that the ALP principle is applicable even to barter transactions. A barter transaction can be said to conform to the ALP principle only if the market value of the commodities being exchanged is equal. Similarly, a DSA can only be an arm’s length barter transaction if the market value of data of both entities is equal. In such a situation, it is important to measure the value of the data being offered by parties to the DSA. There is no standardised valuation method for data but mechanisms such as market-based approaches, economic models and dimensional models can be used to determine the value of data. These methods can be coupled with the standard comparable uncontrolled price (CUP) method to determine the fair market value (FMV) of the data of each party. This FMV will allow us to determine the ALP of the transaction if there is a difference between the FMV of the data being shared by the parties to the DSA, the same can be appropriately accounted for and taxed accordingly.
For instance, two companies, Company A and Company B, enter into a data-sharing agreement where they agree to exchange user data. The FMV of data- for Company A is Rs. 10,000 and for Company B is Rs. 50,000. Herein, we can see two transactions, Company A will buy data worth Rs. 50,000 from Company B at a cheaper price of Rs. 10,000 and secondly, Company B will buy data worth Rs. 10,000 from Company A at a higher price of Rs. 50,000. The ALP for the first transaction is Rs. 50,000 and for the second transaction is Rs. 10,000. In such a situation for Company B according to the ALP principle, an expense of Rs. 40,000 will be disallowed and added back to its income and taxed accordingly. Buying goods at inflated prices from related entities located in a tax haven represents a classic case of transfer pricing and profit shifting.
III. Implications of Taxing DSAs
Presently, DSAs are not taxed at all, but the aforesaid proposed framework illustrates a mechanism to tax the DSAs. This presents an opportunity for the Government to tax these agreements and enhance revenue. The fact that DSAs are currently not being taxed is not solely dependent upon the existence of a mechanism to tax them, it may also involve policy considerations like promoting ease of doing business for digital enterprises and since targeted advertising leads to the sale of other goods and services as well which is beneficial for the economy.
For businesses, it is important to be aware of such taxation possibilities as it will allow them to be more prudent in their dealings. It will also lead to increased compliance costs as they’ll have to maintain appropriate transfer pricing documentation to justify their ALP to tax authorities. In the absence of any clear guidelines on valuation methods for data, there may be differences of opinion between the businesses and tax authorities leading to increased litigation. Furthermore, if multiple jurisdictions tax the same data-sharing agreement, companies could face double taxation unless there are appropriate provisions for the same multilateral instruments (MLI) and Double Taxation Avoidance Agreements (DTAAs).
Applying this taxation framework may also have an impact on competition concerns regarding data accessibility. From an anti-trust perspective data is considered a means of achieving market dominance in the digital economy and the lack of data availability leads to the creation of entry barriers in the market. Currently, DSAs between group entities do not attract any legal complexities, taxing these DSAs may result in the companies dealing with third parties as well and not just related parties. This will also the existing tech giants to generate more revenue by selling or sharing their data and will also open up closed ecosystems of the tech giants, allowing more players to enter the market and drive innovation and growth. However, on the other hand, taxing these agreements may also discourage enterprises from engaging in collaborative data-sharing initiatives that may lead to innovation and growth as they may be wary of increased compliance costs and legal uncertainty. Additionally, enterprises may still refuse to deal with third parties at all, in an effort to maintain their dominance.
IV. Conclusion
The reliance of the digital economy on DSAs underscores the need for a robust taxation framework. While DSAs are currently untaxed, applying ALP to these transactions is extremely important. By determining the FMV of shared data using market-based approaches, tax authorities can ensure effective taxation of DSAs between associated enterprises and also have spillover effects addressing other issues related to data monopolisation.
Taxing DSAs could enhance government revenue and promote competition by encouraging data-sharing with third parties, breaking down monopolistic ecosystems of tech giants. But it may also risk stifling innovation if enterprises become wary of compliance costs and legal uncertainty. For the purpose of ensuring a balanced environment for businesses, policymakers will need to introduce a whole series of measures covering all bases that ensure effective taxation while preventing unnecessary burdens on businesses. It is also important to understand that taxing DSAs will require businesses, legal and accountancy professionals to adapt to this new dimension.
In conclusion, while taxing DSAs presents an opportunity to unlock value in the digital economy, it requires a nuanced approach to maintain a competitive landscape and not negatively impact the ease of doing business in the country.