A budget-neutral way to encourage business investment in research

The signature accomplishment of President Trump’s first term in office — the Tax Cuts and Jobs Act of 2017 — was a mixed bag for America’s long-term growth. It made U.S. companies more competitive by lowering the corporate income tax rate and moving the country toward a so-called “territorial” tax system that calculates how much firms owe based only on their domestic earnings. But while a strategic rationale for the corporate tax cut was to boost the kind of business investment that drives productivity, it failed to increase the value of the research and development tax credit, and it eliminated firms’ ability to write off research expenses in the first year starting in 2022, meaning companies conducting R&D will pay more taxes, not less. The net effect will be a nearly $8 billion increase in the after-tax cost of R&D.

R&D tax incentives are justified by the fact that most of the value from private research goes to society overall, not just to the companies that pay for it. Numerous studies show that each dollar of tax revenue foregone due to the R&D tax credit leads to at least a dollar of extra R&D, which in turn boosts social welfare by two to three dollars. The country needs strong research incentives not only because economic evidence is clear that innovation powers improvements in living standards, but also because innovation will be critical for achieving many important national missions, such as staying ahead of China technologically and developing the technologies needed to address climate change.

Yet despite the overwhelming consensus of scholarly studies showing the tax credit’s efficiency, the relative value of the U.S. credit has fallen compared to other nations. Whereas the United States in 1994 boasted the world’s most generous R&D tax credit, today its value is tied for 30th place among OECD countries. 

To solve this problem and make the United States more technologically competitive, Congress should build on the 2017 tax bill by significantly increasing the Alternative Simplified Credit (ASC). Under the ASC, companies are credited 14 percent of the difference between their current research spending and half of their average research spending over the last three years. The idea is to provide an incentive for companies to keep ratcheting up their investments in research. To make the credit an even stronger engine for R&D investment here in America, Congress should at least double the rate (to 28 percent). Doing so would create jobs, innovation, and GDP growth. The Information Technology and Innovation Foundation estimated in 2010 that changing the credit to just 20 percent would create 162,000 jobs, generate 3,850 additional patents each year, while increasing productivity by 0.64 percent and GDP by $66 billion per year. In real, net-present-value terms, the taxes from this increased economic activity would balance the revenue loss within 15 years.

To be sure, this would cost the federal treasury money, at least in the short-term. In 2014, the latest year for which detailed data are available, companies claiming the ASC had performed $356.6 billion in research qualifying for the credit over the previous three years, for an average of $119 billion annually. Taking half of this annual base and increasing the credit on it from 14 to 28 percent would cost an additional $8.3 billion per year. The cost would be higher if the expanded credit caused companies to increase their research.

The common pushback to such proposals is, how do we pay for it? The answer is that there is an easy way for Congress to offset the cost while also spurring growth: tax dividends as normal income.

Qualified dividends (which includes most dividends) currently are subject to a maximum tax rate of 20 percent, whereas the top rate on regular income is 37 percent. In other words, stockholders — who are generally much wealthier than average Americans — pay a much lower rate on this income than they do on their regular income. But there is also evidence that reducing the rate on dividends in 2003 actually led companies to funnel more capital to dividends rather than invest it in things like research and development, new capital equipment, or worker training. So, eliminating this tax break that largely goes to the wealthy and using the revenue to expand the ASC would compound the growth-generating effect by providing both a push and a pull — allowing firms to retain more money to invest in R&D, and giving them a greater incentive to do so.

The Treasury Department estimates the tax break on qualified dividends costs the government $384 billion over 10 years — more than enough to pay for a large expansion in the ASC. If the United States wants to maintain its lead in new technology and innovative science, then it needs to lower the net cost of investing in research and product development. Swapping the dividend break for a more generous ASC would be the perfect solution. 

Joe Kennedy is a senior fellow at the Information Technology and Innovation Foundation (ITIF), a think tank for science and technology policy.