Study explores tax tactic that lets multinationals shift costs back to the U.S.

It’s no secret that many U.S.-based global companies will park profits overseas to minimize taxes paid to Uncle Sam. Routing income through low-tax countries has been so common among multinationals that they can take on catchy names, like the “Double Irish” or “Dutch Sandwich.”

But shifting profits from high-tax to low-tax countries isn’t the only way that American multinationals lower their tax tabs. They also bring costs incurred in lower-tax countries back to the U.S. so they can claim them as deductions and credits on their tax returns. This reduces what they owe the federal government.

Research by Juan Carlos Suárez Serrato, a senior fellow at the Stanford Institute for Economic Policy Research (SIEPR), sheds new light on this cost-shifting tactic and its broader effects, including on how much companies invest in research and development, which is key to economic growth and global competitiveness. The study, published this month in the journal International Tax and Public Finance, comes as Congress mulls extending business tax cuts, including incentives for R&D spending, enacted during President Trump’s first term.

Not much is known about the economic impacts of these cost-shifting strategies, in part because companies don’t always have to disclose them. But like all tax-avoidance planning, they can have wide-ranging consequences beyond the hit to federal tax revenues that aren’t fully understood, says Suárez Serrato, who is also the Edward B. Rust Professor of Economics at Stanford Graduate School of Business (GSB).

“Too often, policymakers are flying blind when it comes to taxing multinationals, because they don’t have a good understanding of how companies shift their income and costs for tax-planning purposes and what their broader economic effects are,” says Suárez Serrato, whose extensive research has delved into U.S. corporate tax planning strategies.

An indirect tax subsidy

At the heart of Suárez Serrato’s study – co-authored with Lysle Boller, an economist at the Penn Wharton Budget Model, and Clare Doyle, a predoctoral research fellow at the GSB – is one lesser-known way that U.S. multinationals minimize their tax burden: cost-sharing agreements that allow them to split the costs of research and development with their foreign affiliates. These contracts are also attractive because American multinationals can use them to avoid paying U.S. taxes on profits earned abroad from the patents or other intellectual property that the deals generate.

The effects of these cost-sharing agreements are not well understood. But Suárez Serrato and his collaborators gained access to Internal Revenue Service data on corporate tax returns that allowed them to study them using a 2005 U.S. Tax Court ruling that changed how American multinationals accounted for the portion of R&D spent on employee stock options. The court held that U.S. multinationals did not have to split these costs with their foreign businesses under terms of their cost-sharing agreement; they could be treated as costs incurred solely by the parent company. This emboldened U.S. multinationals with cost-sharing agreements to begin deducting the full cost of their stock-option compensation, thereby reducing their U.S. tax liability.

In effect, the court had created a tax shield for these companies. The researchers find that some U.S.-based multinationals with cost-sharing agreements responded by investing significantly more in R&D compared to other multinationals. They also rejiggered their pay packages so that stock options comprised a bigger component not just of their R&D budgets but also their overall labor costs. And in the short run following the court decision, they benefited from an increase in their stock market values as investors recognized the ruling as good news for U.S. multinationals with large intellectual property portfolios.

In other words, the tax shield doubled as an inadvertent boost to R&D and company valuations.

Although a subsequent court case effectively ended this strategy in 2020, the R&D findings are relevant today, Suárez Serrato and Doyle said in an interview with SIEPR. Research has shown that American companies vastly underinvest in R&D because it’s hugely expensive with no guarantee of a return. A recent SIEPR Policy Brief details this problem and warns that American businesses risk losing their competitive advantage without more R&D tax subsidies.

On the one hand, it can seem like a good thing that some U.S. multinationals with cost-sharing agreements spent more on R&D following the 2005 court case that Suárez Serrato and his co-authors study. But the finding also raises important issues around fairness and efficiency in tax policymaking generally.

The 2005 ruling, for example, benefited only certain U.S. businesses – namely large multinationals that use intellectual property to generate profits. But other American companies, from multinationals without cost-sharing agreements to businesses that only operate in the U.S., were left out.

“Evaluating tax policy is not a question of whether it’s good or bad,” Doyle says. “It’s a question of trade-offs.” And, she said, it’s not clear that an obscure practice like expensing stock-option compensation in a cost-sharing agreement is the most efficient way to promote innovation and, ultimately, economic growth.

Still, Suárez Serrato says the study’s findings underscore the need for further research and better information about the many ways large companies use tax shields and their economy-wide impacts.

“It’s just hard to get corporate tax policies right when all of the costs and benefits have not been properly considered,” he says.