The OECD has laid out a slew of options for how to tax the digital economy, but it is far from alone in trying to settle this question. The Oxford International Tax Group has drawn up a draft proposal for global tax reform to resolve the dilemmas of the digital economy through an RPA-I system.
Under the RPA-I regime, a company’s routine profits would be taxed where functions and activities (F&A) costs are undertaken, whereas the residual profits would be taxed on a destination basis. In other words, traditional transfer pricing would still apply to routine profits.
Many taxpayers would prefer to go for a division of residual returns and routine profits rather than go down the road to full-blown formulary apportionment. However, reallocating profits according to the source of income is tough in practice.
“It all depends how you understand your key value drivers,” said Katherine Amos, VP for global transfer pricing and tax disputes at Johnson & Johnson. “In our case, marketing intangibles are crucial to our residual profits.”
“We use residual profit splits all the time,” Amos said. “You might decide that 30% of marketing spend is value-added. You might decide for particular areas that this does not really involve marketing intangibles.”
“We have had a lot of audits on this area and it’s never about the residual profit,” she explained. “It’s about how we got the routine profit.”
Even though the reallocation of residual profits might be a ‘lesser evil’ to formulary apportionment. It would leave the arm’s-length principle (ALP) in place for routine profits and grant taxpayers much more leeway than a unitary tax system.
“The RPA is a hybrid system which would take the global profits of the MNE and divide the profits into the routine and residual returns, before benchmarking with third-party functions,” said Li Liu, economist at the International Monetary Fund (IMF).
“You allocate the routine profit by the ALP to the source country and the rest of the profit is the residual return and goes to the destination country,” Liu said.
“You could allocate the residual returns by formula. It could be user participation or marketing intangibles,” she added. “You could even use a more conventional formula, e.g. capital, employment and sales.”
Bill Sample, tax chair of the US Council for International Business (USCIB), and Amy Roberti, director of global tax policy at Procter & Gamble, have both stressed that the profit reallocation method has to be “modest” for these reforms to work. Sample favours a residual profit split method, whereas Roberti is open to other options proposed.
“A global minimum tax rate would be elegant and simplistic. It’s a kind of bottom-up approach rather than a top-down approach,” Roberti said. “But I think you’d need mandatory binding arbitration to make it work.”
A growing number of US companies are open to a global minimum rate, even though it would almost certainly require the end of the arm’s-length principle and a shift to a formulary system. This is because the US has already established minimum rates of its own.
The alternative of RPA-I would create its own level playing field, but it would lead to a massive redistribution of income. All countries in which a company generates profit would get the right to tax a part of those profits and any countries where there are no real F&A would be denied taxing rights.
“The residual profits would be taxed in the destination country whether the sale is made remotely or not,” said John Vella, associate professor of tax law at Oxford University. “This reduces location distortions as only routine profits would be taxed where F&A take place.”
“That being said, corporate tax under the RPA-I system would still distort investment and financing decisions,” Vella said. “We could have fixed these problems, but we tried to keep it as close as possible to the existing system.”
The RPA-I proposal has two things in common with the OECD’s pillar one proposals. Firstly, it makes the distinction between routine and residual profits, while it’s also destination-based. The same is true of the modified residual profit split and the destination-based approach to marketing intangibles.
There are several important differences though. The RPA-I is guided by where value is created, while the allocation of residual profits would go to the destination jurisdiction. This could help reduce the problems with profit shifting around intangibles.
It’s still unclear how the OECD’s approach to fractional apportionment would deal with intangibles. The formula includes traditional keys like sales, payroll and assets, with the addition of users. It is possible that intellectual property (IP) and other intangibles would be classified under ‘assets’ but this is yet to be clarified.
“We as academics have the luxury of sitting in ivory towers and thinking about these policy issues for a long time. The OECD doesn’t have this luxury,” said Michael Devereux, professor of tax law and chair of the Oxford International Tax Group.
Devereux and Vella founded the group in 2013 with fellow academics Alan Auerbach and Wolfgang Schön, as well as IMF Economist Michael Keen and Paul Oosterhuis, partner at Skadden, Arps, Slate, Meagher & Flom.
The Oxford International Tax Group is best known for its flagship policy of a destination-based cash flow tax (DBCFT) and the RPA-I system would be a step towards this ideal.
“When it comes to tax avoidance, the RPA-I system reduces the hardest problems of transfer pricing,” Vella said. “Namely debt-shifting and profit shifting by relocating IP in low-tax jurisdictions.”
“Though there are still some clear ways to shift profits against the RPA-I,” he continued. “You could route sales through a third party independent distributor in a low-tax jurisdiction. So we need to think about that.”
In many ways, the RPA-I proposal is halfway between the existing system and a destination-based cash flow model. The DBCFT was briefly on the agenda in the US, but the Republicans dropped the policy in favour of anti-abuse measures more in line with OECD recommendations.
Even still, the dynamics of international tax reform are more unpredictable than ever before. Radical ideas that were once marginal are now mainstream. This is one reason why reallocating residual profits by income source might not just be a part of future policy.
Business leaders are still playing the waiting game, while policymakers argue about the details. What companies fear is that the OECD will not find a digital tax solution rooted in sound principles. If this is the case, the organisation could be launching BEPS 3.0 in years to come.