BONUS CONTENT: Harmful Tax Practices

This is a bonus post for paying subscribers. Sign up here for an Emperor Subscription to get full access, including past interviews and extra content.

One of the key principles of the international tax system is that everyone is supposed to tend their own garden. Basic transfer pricing gives nations the tools to protect their tax base, ensuring that their own taxpayers pay the right amount of taxation based on the laws that the legislature has set. What other countries do with their own tax bases may affect the transactions that multinational corporations make between the two—but so long as there are agreed-upon dispute resolution methods and arm’s-length pricing, it’s not one country’s business how the other levies taxes.

In theory. In practice, of course, countries do care what other countries do. They often will engage in retaliatory measures and draw up blacklists of tax havens based on their conduct. Whether or not that’s in keeping with the spirit of the arm’s-length standard, it’s going to happen regardless. And the Organization for Economic Cooperation and Development, the global tax norm-setting body and caretaker of the arm’s-length standard, long ago realized it would be better for these retaliatory actions to conform to general standards as well.

That’s the basic idea behind the OECD’s Forum on Harmful Tax Practices, created first in 1998 and eventually incorporated into the 2015 Action Plan for Base Erosion and Profit Shifting. Probably the most important result of this work is the evaluation of patent boxes and other regimes that favor intellectual property and intangibles-based income, by the “modified nexus approach.” This requires that such regimes include requirements for research & development activity in the jurisdiction—in essence, a rough safeguard against patent box-driven income-shifting through artificial means.

While the OECD monitors patent boxes to ensure compliance, they can't enforce the standard themselves. But many other countries have enacted anti-abuse and retaliatory measures based on the OECD's rulings.

This more or less ended the reign of the pure patent box, which countries such as Ireland had used to attract billions of dollars in income. They are now mostly replaced with OECD-compliant regimes that tend to be used less often by corporate taxpayers.

But these standards are also important to understand as they continue to influence global tax reform efforts moving forward—even though they aren’t yet a major influence on the Two-Pillar plan. In fact, because they aren’t yet a major influence on the Two-Pillar plan. Understanding the differences between the two philosophies for dealing with profit-shifting with intangible income can help understand some of the obstacles they’ve run into. And how they may borrow from each other to try to overcome them.

The nexus standard is based, somewhat, on the idea of a credit for research-and-development. Those rise and fall proportionate to the dollar amount of qualifying research activities. (Generally, experts view these as the economically optimal type of tax incentive anyways, compared to an income-based one. But, countries are going to pursue either, or both, regardless.)

A modified nexus approach is largely based on the same kinds of expenditures that an R&D credit would be, but the amount of the tax benefit can still be based on the amount of income the taxpayer earns. The key is that the benefit is now limited proportionate to the amount of R&D activity that was related to the IP earning the income.

“Expenditures therefore act as a proxy for substantial activities,” the OECD report states. “It is not the amount of expenditures that acts as a direct proxy for the amount of activities. It is instead the proportion of expenditures directly related to development activities that demonstrates real value added by the taxpayer and acts as a proxy for how much substantial activity the taxpayer undertook.”

The type of expenditures that qualify are generally the same as what qualifies for an R&D credit–according to the OECD, that can include “salary and wages, direct costs, overhead costs directly associated with R&D facilities, and cost of supplies so long as all of these costs arise out of activities undertaken to advance the understanding of scientific relations or technologies, address known scientific or technological obstacles, or otherwise increase knowledge or develop new applications.”

They can’t include the cost from acquiring IP from another company. (Or acquiring a company that has a lot of IP.) That’s an interesting restriction, because in theory it might introduce an arbitrary bias towards companies with “home-grown” patents. One way the rules counteract for this is by allowing an “uplift” of up to 30% to the amount of qualifying expenditures going into the formula. The amount of the uplift can’t exceed the total amount of the company’s expenditures. This is where the “modified” part of the “modified nexus” model comes from, and according to the OECD it “acknowledges that taxpayers that acquired IP or outsourced a portion of the R&D to a related party may themselves still be responsible for much of the value creation that contributed to IP income.”

But the tricky part isn’t identifying the expenditures, it’s tying those expenditures to IP income. To receive the tax benefit, a taxpayer must show that the expenditures helped develop intangible assets that are generating revenue. They can do this by either tracking the asset or the products that use it. Given the complexity of IP-based business models–as I always love to point out, a smartphone can have more than 200 patents in it–this is likely quite a challenge.

This complexity may be part of the reason why this approach hasn’t been used in the OECD’s Two-Pillar project. Pillar Two, the 15% global minimum tax, has been harshly criticized, especially in the U.S., for potentially negating the value of tax credits, including the R&D credit. Anything which lowers a company’s effective tax rate could put a company below the 15% level, increasing its tax costs under the Pillar Two rules.

Pillar Two does include a carveout for economic substance, under similar logic as the modified nexus standard. Taxable income should be tied to real activities that generate the income. But Pillar Two doesn’t look for any linkage between the activities and IP, or IP-generated income. It’s simply based on a percentage of tangible assets and payroll costs. It's a much rougher way to estimate intangible value.

And at 5% for both, it’s also apparently not big enough to cover many profitable, research-intensive companies that are expecting to see new tax bills under the Pillar Two regime.

I’ve wondered why the OECD hasn’t looked at its harmful tax practice work as a potential way forward on this issue, which is generating a lot of tension between the United States and the other OECD members. Part of the thinking behind simpler substance rules like the Pillar Two substance-based carveout is that defining and tracking intangible assets is just too hard. In a giant conglomerate, is it possible to understand which scientist in which lab contributed which percentage to a new life-saving drug? Or which computer engineer creating the algorithm behind the next Internet sensation? Maybe the company itself can track that, but will tax authorities be able to understand it and challenge it when necessary?

Nevertheless, the harmful tax practice rules are still in place and used throughout the world. It will be interesting to see how these differing approaches intersect in the years ahead.

Contact the author at amparkerdc@gmail.com.