Not all money earned in a tax haven such as Bermuda is necessarily from income shifting, two accounting professors said. (AP)
The authors of some of those previous estimates strongly dispute some of Blouin and Robinson's contentions, while also admitting that any method will have flaws and will require making assumptions.
The dispute is not just over number-crunching. Experts are debating over how best to define artificially shifted income, and how to distinguish it from money that flows from jurisdiction to jurisdiction for any number of different reasons.
Companies can use a variety of techniques to try to push as much of their income as possible to jurisdictions with low tax rates, or no corporate tax at all. Through transfer pricing — transferring assets from one subsidiary to another, using market rates to estimate a price — a taxpayer can move valuable intellectual property to wherever its returns will see the lightest taxation. Corporations can also use intercompany lending to strip earnings from a high-tax jurisdiction and assign them to whichever country they desire. Or, sometimes, "stateless" income ends up in no jurisdiction at all.
Tax officials from around the world have long agreed that tax avoidance is a problem. At the Organization for Economic Cooperation and Development, which sets global tax norms, nations are currently negotiating over new measures to address income shifting.
But despite this intense focus on income shifting, measuring the scale of the problem and estimating how much revenue countries are losing is harder than it might seem.
In a paper released this November, "Double Counting Accounting: How Much Profit of Multinational Enterprises Is Really in Tax Havens?" Blouin and Robinson argue that many of their peers have been double-counting income from the Bureau of Economic Analysis, or BEA, inflating their estimates. The issue comes down to income from subsidiaries of subsidiaries — which, due to the complexity of international tax structures, are where much of a company's income ends up.
Many accounting experts, including Kimberly Clausing of Reed College, have estimated corporate tax avoidance based on the variation of reported income in each jurisdiction. Calculating for other factors, they estimate an amount of income that would have been reported in different jurisdictions, such as the U.S., if not for pure tax considerations.
But without adjustments, the analysis would count some income twice — first in the entity reporting the income, and then in the foreign subsidiary that controls that entity.
This, in and of itself, isn't news. The problem has been known to researchers for years.
"It's just an artifact of the data," said Kyle Pomerleau, a fellow and tax expert at the American Enterprise Institute, a right-leaning think tank in Washington, D.C. "When you start to get into BEA data, there are lots of little trolls and artifacts that you just have to get familiar with."
Clausing addressed the issue in a 2016 paper, "The Effect of Profit Shifting on the Corporate Tax Base in the United States and Beyond." But Blouin and Robinson claim the adjustment she and other researchers used doesn't go far enough to eliminate the double-counted income. They claim that to fully account for the issue, all of the income passed up from one subsidiary to a second, known as equity income, should be subtracted.
By doing that, they arrive at vastly different estimates than what other researchers had found. For instance, Clausing estimated that the U.S. lost as much as $111 billion annually due to profit shifting. Blouin and Robinson adjust that figure to $32 billion.
Blouin and Robinson said this explained an apparent discrepancy in the data. Companies report large amounts of income in tax havens, implying a loss of revenue that can't be found in any of the other research.
But not every dime earned in an apparent tax haven such as Luxembourg or Bermuda is necessarily attributable to income shifting, they argue. Corporations may choose to locate a holding company in those jurisdictions for tax reasons, but that entity would have earned income anywhere. As a holding company, part of its function will be to collect earnings from around the world that have already been taxed by the countries where they were generated.
"Equity income's growth in tax havens stems from the use of tax havens as foreign holding companies," Blouin and Robinson said. "Although their existence is certainly attributable to tax planning incentives, these tax haven affiliates' largest assets are typically their ownership of other foreign affiliates that generate profits elsewhere."
Beyond that, the two academics said differences in how individual countries report earnings indicate potential misstatements throughout the academic research.
"Overall, we conclude that researchers must take care when making comparisons of profit shifting activity using samples of affiliates of [multinational entities] with different parent home countries," they wrote.
Blouin and Robinson would not comment for this article.
Clausing admits that Blouin and Robinson found one error in the BEA data that would reduce her estimate by a small amount. But otherwise, she claimed that their conclusion was wrong, not just as a matter of computation but in concept as well.
"In my opinion, they're overcorrecting," she told Law360. "Their correction to the double income problem actually gets rid of all of the foreign-to-foreign shifting."
When companies shift income from one foreign jurisdiction to another, it still affects U.S. revenue, she argued. Some of that income may have ended up in the U.S. if not for tax havens. Clausing claimed a method that ignores that activity arrives at a misleading conclusion.
"While it may be suitable for some research questions, it is not suitable for estimating the overall size of profit shifting, since foreign-to-foreign profit shifting is quantitatively important," she wrote in a soon-to-be-published paper as a response to Blouin and Robinson.
Aside from the calculations, this is also about defining shifted income.
"It is possible that some of the disagreement is due to terminology," she wrote. "Blouin and Robinson's method will show us where income is earned from an accounting perspective. However, the costs of profit shifting are generated by deviations between where income is truly earned and where it is booked for tax purposes."
To back her estimate, Clausing points to the aggregate data published by the Internal Revenue Service from taxpayers' country-by-country reports, now required by OECD standards. For instance, U.S. companies reported nearly $25 billion in 2016 before-tax profit in Bermuda, despite having very few workers employed there. These figures suggest income-shifting on a scale similar to her estimates, Clausing said.
But she also admits that any correction to the issue will involve some guesswork. Using her best estimate, Clausing attributes a portion of the equity income — the profits earned by subsidiaries of subsidiaries, which could represent foreign-to-foreign income-shifting — to the U.S., but ultimately there's no way to know for certain.
"For more than a decade, economists at the BEA have reported that there was no simple correction for the double-counting," she wrote in her response.
And while she tried to consider every factor for her calculation, Clausing admits that she still may catch variations in income due to issues other than tax rates. Ireland, for instance, has been accused of luring income through lax treatment of profit shifting, yet tech companies have also built up significant workforces there, generating real activity.
"Finding truth on that question is nearly impossible," she said. "We can't run the counterfactual where Ireland just has a normal tax rate."
Over the years, researchers have used a variety of methods to try to paint a picture of global profit shifting.
In September, the International Monetary Fund issued a report on "phantom investment," or income held in entities with "no real business activities," which the agency said rose to $15 trillion in 2017. But while tax avoidance is certainly one reason to use a shell company, it isn't the only one.
In his 2015 book "The Hidden Wealth of Nations," Gabriel Zucman, an economist at the University of California at Berkeley, estimated that the U.S. loses $130 billion annually due to profit shifting, a figure even higher than Clausing's. Zucman's estimate comes from past research that analyzed the portion of income held in low-tax jurisdictions, which he said could be attributed to profit shifting.
Blouin and Robinson also claimed that Zucman's estimate didn't correctly account for double-counting, which he agrees with — partially.
Zucman told Law360 that he revised his methodology in a 2018 paper as well as in his 2019 book, "The Triumph of Injustice." But he also said Blouin and Robinson erred by not including Puerto Rico, a U.S. territory that has its own tax system with incentives that are often attractive to multinational corporations.
"In my view, Blouin-Robinson does not add anything that was not known before," Zucman said.
Researchers have said that the wide variation in estimates is due to many factors: the difficulty in compiling data across jurisdictions, questions about how to identify shifted income, as well as the fact that so much of the data remains secret and unavailable.
And, while the numbers certainly matter for choosing the best solution, they're not necessarily the end of the discussion.
"Companies can use leverage, or transfer pricing, to reduce their profits in certain jurisdictions. That undermines people's trust in the tax system, regardless of what you think the magnitudes are," said Pomerleau at the American Enterprise Institute. "You can look at the Zucman or the Clausing estimates — those magnitudes are larger — but in the grand scheme of things they aren't all that big, either."
Lack of faith in the tax system can undermine credibility in all government institutions.
"I think it's ultimately a problem of tax fairness," Clausing said. "I think coming up with better tax systems that are really more immune to these tax competition pressures is really better for democracy, so everyone feels treated fairly."
--Editing by Robert Rudinger and Neil Cohen.
Editor's Note: "Double Counting Accounting: How Much Profit of Multinational Enterprises Is Really in Tax Havens?" was published by SSRN, which is owned by Elsevier Inc., a division of RELX Group, which owns Law360.
For a reprint of this article, please contact reprints@law360.com.
