Taxing digital services: The devil's in the details

How to tax the digital economy, i.e., commercial transactions conducted electronically on the internet, has been a thorny issue for governments and business for years. In March the European Commission unveiled a proposal for two new directives to stem what the EC considers to be revenue losses caused by loopholes in the global corporate tax system. Officials estimate that digital businesses in the EU pay an average effective tax rate of 9.5%, while traditional businesses pay 23.3%.

A two-track proposal

The EC’s initiative is important not just because of the size of Europe’s digital economy, but because other jurisdictions may follow its lead in looking for new ways avenues to tax e-commerce. The complex proposal includes a directive, called the Digital PE, that creates a new definition of a company's “permanent establishment” for tax purposes. It would allow EU member states to tax digital profits generated in their territory, provided the company has either:

• Annual revenues in the member state exceeding €7 million;
• More than 100,000 users; or
• More than 3,000 business contracts for digital services in the state in a taxable year.

To make the new rule effective, however, the European Union would need to renegotiate the many bilateral treaties on corporate taxation that the EU and its member states have with other jurisdictions—a process that could take years.

The second directive—the EU Digital Services Tax (DST)--is billed as an interim measure until Digital PE can be implemented. It's also far more controversial. The DST would impose a 3% turnover tax on three specific types of digital commerce revenues: 


• Those created from selling online advertising space;
• Those created from intermediary activities that allow users to interact with other users to facilitate the sale of goods and services; and
• Those from the sale of data generated from user-provided information.

Since turnover or gross revenue taxes are not covered by treaties, the EC may impose the tax as soon as the EU member states agree on the details. In its current form, the DST would only apply to companies with total annual worldwide revenues of €750 million and EU revenues of €50 million, from which the EC estimates that member states would generate €5 billion in additional revenues a year.

“Ringfencing” the digital giants

Critics say the DST violates an understanding, going back to the OECD's 1998 Ottawa Taxation Framework, that taxation of e-commerce should not discriminate against companies from another country. That's because the DST appears to “ringfence” the big U.S. companies, e.g. Apple and Amazon.com, that dominate digital commerce in the three categories the directive singles out. Excluded are, among others, financial services, payment processing, online marketplaces and Cloud software—digital businesses in which other players are more competitive.

The OECD already had a process under way to find consensus across its 113-country membership on a definition of digital commerce that can form the basis for a global convention on taxation. But in June of last year, France, Italy, Spain and Germany wrote to the EC, maintaining that the OECD process was too slow and urging the commission to take action. Under pressure from its biggest economies, the EC is rushing to put a final version of its new directives in place by January 1, 2020: a very ambitious goal.

A perspective from KPMG

Tax authorities have expressed a wide range of views on the fundamental issues for taxation posed by the digital economy.  While some see no need for change, others believe that a broad rethinking of the international tax system is warranted given the increasing digitalization of the economy. Yet others believe that businesses providing virtual goods and services are digital businesses with significantly different characteristics from “traditional” businesses, thus requiring special rules for taxation. 

This debate highlights the critical importance of understanding digital business models. One reason governments believe change is needed (even if they disagree on the scope of that change) is that digitalization of businesses is driving an increasing misalignment between value creation and taxation. The reason often cited for this misalignment is that user participation creates value.  Terms such as user participation, user-generated content, user data, user network effect, and user contribution to brand are used to describe value drivers for a business in user locations that existing rules ignore in determining the location of taxation.

Equally important, but less widely considered, are the impacts of digitization of the business model itself.  As companies embark on digitally transforming their business models, technology will, in many instances, change the way value should be attributed along the value chain.  Whether it’s through the aggregation and monetization of data, utilization of IoT enabled devices, or investments in technologies such as blockchain to improve the transparency and efficiency of the supply chain, the development of new sources of intellectual property in the digitized business model are increasing rapidly.

Given the emphasis on value creation whether by users or otherwise, a deep understanding of the value chain, which will vary by industry and, within an industry, by business, is required. Thus, any policy that applies a tax uniformly across businesses based solely on the total volume of the business, such as the DST, ignores these differences. A policy that doesn’t ignore these differences, such as the proposed Digital PE, highlights the critical importance of understanding value creation in the specific business. One of the biggest challenges to attributing profit to a Digital PE will lie in determining the relative contributions of the digital technology platforms and the user contributions, not least because of their interrelation.    

Such a move could also be risky, says Robert Kovacev, partner at Steptoe and Johnson LLC, because the EU's relationship with the U.S. regarding trade and taxation is not especially friendly following passage of the Tax Cuts and Jobs Act, which includes tax provisions aimed at foreign companies.

The process also worries critics, because the current draft of the DST contains no hard-and-fast sunset provisions. The DST would in principle be superseded by the Digital PE, but only if it was included in the tax treaty with outside countries such as the U.S. and China. Yet if the EC were to enact the DST, and if the member states then become accustomed to the additional revenue from the turnover tax, they may not want to give it up, even if other jurisdictions agree to the Digital PE. What was intended as an interim levy could become permanent.

A blunt instrument

While the DST targets certain specific businesses, its impact on others is uncertain. “Digitalization is creeping into all aspects of the economy,” notes Jan Bart Schober, a member of the international tax practice at Loyens & Loeff in London. This means that companies in traditional industries are collecting and monetizing customer data—activities that might subject them to the DST.

Just collecting the data needed to calculate their digital tax obligations will be difficult. To do so, “companies will need to track what turnover is allocable to what users,” says Schober. “If some activities are covered by DST and some are not, the company has to calculate the appropriate split.” That could be a problem for companies that bundle advice into the digital services they sell, for example.

The Digital PE proposal will be not be easy to implement in this respect as well, says Kovacev, because it requires companies to track how many users they have in each market. Yet there is no fully reliable way to determine this: “They can use IP addresses, but that's not perfect.”

All of this comes at a time when digital companies are under attack for their collection and use of data on individuals—particularly in the EU, where the General Data Protection Regulation (GDPR) has just gone into effect. “The commission will need to set up appropriate rules,” says Schober.

While details such as these could be sorted out, critics have expressed a deeper concern about DST: that it's economically distortionary. A 3% tax on gross revenues that hits companies with high levels of revenue but low profit margins, which includes many in the digital economy, could send some of them “down the drain” in a bad year,” Schober says: “If your company has that business model, it might say, 'We don't want to do business in that country anymore.'”

For American digital giants operating in the EU, some of the impact might be reduced if they could deduct it against their corporate income tax in the U.S. But the Treasury Department has not yet weighed in on this approach, and even if it’s allowed, companies probably would not be able to recoup the entire amount, says Schober.

The taxable future

The big players in the digital economy are unlikely to curtail their activities in Europe as a result, says Stan Veager, resident scholar at the American Enterprise Institute (AEI). “The EU is a big enough market that nobody will move away,” he says. “To remain in business in China, they are subject to intellectual property sharing, strict censorship—and that's not a breaking point.”

The best outcome, Kovacev says, would be for the EC to jettison the DST and concentrate its efforts on the Digital PE. It would take longer to implement, but by taxing profits rather than gross revenue, it would bring the EU more closely in line with the OECD's efforts to develop a global consensus on taxation of the digital economy. The answer, however, won't be known for many months, says Kovacev; the EC's proposal “is just the opening bid.”