Earlier this week, the OECD released the Secretariat Proposal for a "Unified Approach" for the new tax nexus and profit allocation rules to address the tax challenges of the increasingly digitalized economy. The proposal covers highly digitalized business models, but is increased in scope to include consumer-facing businesses. The Unified Approach creates 1) a new nexus rule, not dependent on physical presence and instead largely based on sales, and 2) a new profit allocation rule using a formulaic approach to determine a share of residual, or non-routine, profit allocated to market countries. In addition, if a taxpayer has a traditional nexus in the market country, an additional amount of profit consisting of a fixed return for certain baseline marketing and distribution functions that take place in the market country may further be allocated to the market country. In exchange for this new taxing right of market countries, the countries should agree to a binding and effective dispute prevention and resolution mechanism even if there might be cases where there are more functions in the market countries—that is, more allocable profits to market countries—than the baseline marketing and distribution functions.
This proposal offers a possible consensus-based solution to be agreed to by the end of 2020. While the G20/OECD has worked on a globally coordinated solution, a number of market countries, such as the UK and France, have adopted or plan to adopt the Digital Services Tax ("DST") in order to collect more revenues from Multinational Enterprises (“MNE”). The UK and France mentioned that the DST is an interim measure or stated that the DST will be reconsidered once there is an international deal on digital taxation. But will we reach a consensus by the end of 2020? And will market countries drop the DST already in place despite the domestic political pressure for additional revenue? We have to wait and see what will happen, but it is possible that DST will still remain after an international deal.
In the current situation, Dan Shaviro's new article, Digital Services Taxes and the Broader Shift from Determining the Source of Income to Taxing Location-Specific Rents, offers a silver lining of DSTs. Shaviro provides support that a properly designed DST can address reasonable concerns about tax avoidance and locally generated rents, and thus may improve global efficiency and distribution.
The Article starts by recognizing the rise of globally undertaxed rents of MNEs, many of which are technology companies. To address the tax-elusiveness of MNE rents and quasi-rents, a new notion of value creation emerged. The discussion of value creation, however, has a production-based flavor, because it implies that the location of creating intangibles may matter more than the location of markets or consumers. This notion of value creation might help to reach a short-term consensus with respect to addressing the so-called “stateless income” of MNEs. However, it is not good for a long-term solution because it revives longstanding disputes between production and market countries on what the geographical "source of income" means.
Another strand of tax policy proposals to reduce the efficacy of MNE tax planning assign more taxing rights for MNE profits to market countries. The suggested mechanics are to adopt either sales-based formulary apportionment for MNE profits or sales-based residual profit allocation between the affiliates of MNE groups. However, these sales-based proposals are limited in their capacity to achieve comparable effectiveness across MNE business models. For example, Starbucks might face higher taxation in market countries than Google and Facebook under the sales-based proposals because it offers tangible products using franchises and third-party intermediaries. Also, many digital companies with two-sided business models price below marginal cost on one side of consumers and make up the difference through increased prices on the other side of consumers. Thus, the sales-based approach cannot properly capture and allocate revenues among heterogeneous market countries.
The above problems of value creation and the sales-based approaches result in the need of adopting a third type of approach that is more ad hoc and consists of targeted taxes, such as DSTs, aimed at MNE rents. Shaviro acknowledges the criticism of the DST in being a gross receipts or turnover tax. However, he argues that using properly designed DSTs and similar instruments can be reasonable, especially because, as Wei Cui notes (reviewed by me here), digital platform companies subject to DSTs tend to have low (or even zero) marginal costs. Thus, a gross receipts tax might be better than a highly manipulatable net income tax in identifying the true net revenue increase of MNEs in a particular country. Furthermore, as suggested by Bankman, Kane and Sykes (reviewed by Elkins here), market countries may use excise taxes or optimal tariffs to capture MNE rents in imperfectly competitive markets. In this regard, Shaviro notes that DSTs can function as optimal tariffs.
I welcome Shaviro's article as an addition to a body of literature sympathetic to the DST including the papers by Cui and myself. However, such literature has common caveats in its proposals—that is, all of the literature supports a properly designed DST. Current DSTs are criticized, for example, for their limited scope, ambiguous method to capture the revenues generated in a market country, and potential discrimination against U.S. MNEs. On the final point of criticism, Shaviro advises that it would be wise for market countries to avoid unduly fostering the perception that they are targeting U.S. MNEs by demonstrating a reasonable degree of good faith. He understands that the United States may prefer that other countries not tax the rents of American MNEs, but he suggests to not view the DSTs as sufficiently offensive to call for retaliation. I couldn't agree more, because someday when many successful MNEs are not US-centric, Americans might benefit from a functioning international regime where inbound rents can be effectively taxed.