Innovation Files has moved! For ITIF's quick takes, quips, and commentary on the latest in tech policy, go to

Critique of OECD R&D Tax Credit Report

IBM Super computer

R&D is fundamentally important to economies because it is a primary source for innovation and new technologies. But markets rarely provide enough incentives for innovation on their own—innovations are expensive to create but easy to copy.

For those reasons many countries provide R&D tax incentives to companies that spend money on basic or applied research. The best way to think of this policy is as actually as a fix—R&D has positive benefits for the economy as a whole, but because individual companies have trouble capturing all the benefits of R&D they are unlikely to invest the socially optimal amount.

Tax breaks for businesses are fraught with controversy because they “distort” the market and according to conventional neoclassical economics thinking distortions are by definition bad, even if they are pro-growth. To be sure certain tax incentives outlive their usefulness, as they have in the fossil fuel industry, and some tax incentives are only on the books because they serve special interests, not the public interest.

But the R&D tax incentive doesn’t suffer from any of these problems. In fact, there is solid evidence that R&D incentives spur additional R&D spending and that this funding increases innovation and productivity.

In that light, the new brief from the OECD, Maximizing the benefits of R&D tax incentives for innovation, presents a curiously skeptical view of R&D tax incentive policies. Although the paper presents itself as helpful criticism generally in support of R&D tax credits, the critiques it makes are based on a number of incorrect or misleading ideas and the overall message is one arguing for less, not more, generous R&D incentives.

The paper makes a number of cautionary arguments: R&D credits may not reach smaller, domestic corporations who do not have the profitability; these smaller corporations are particularly important because they create jobs; R&D credits may have “unintended consequences” for national balance sheets when program costs run too high; R&D credits may enable “double non-taxation” through multinational corporations shifting their spending internationally; and that policymakers should balance R&D tax credits with direct support for R&D. Let’s take a quick look at each of these claims in turn.

R&D in smaller, domestic firms: The OECD report says small firms do not benefit from the credit as much as large firms in part because they have less international accounting sophistication to take advantage of it, but this is mostly a natural consequence of the way research is done and not a reason to limit R&D tax credits.

One reason why small firms get less support from the R&D credit is that they do relatively less early stage, exploratory research–exactly the kind of research the credit is designed to stimulate. But it is true that even small firms that do this kind of research may not be able to fully utilize the credit if they are young and have not yet turned a profit. The U.S. credit does allow firms to carry forward credits for up to 7 years, which can help. And there is currently legislation, such as the Start Up Act proposed by Senators Coons (D-DE) and Moran (R-KS) that would let newer firms take the credit against payroll taxes.

But even if smaller firms cannot take the credit as easily, we should not throw the baby out with the bath water. The economy would still be more innovative if midsize and larger firms receive R&D incentives than if no firms get the credit. Besides, despite the hype, small firms are in fact less innovative than large ones.

Part of the OECD claim that the R&D credit is biased against small firms is based on the logic that multinational corporations are better able to shift profits and intellectual property to cheaper tax jurisdictions. The R&D tax credits are allocated according to the location of research and development. Domestic firms will receive R&D credits if they conduct R&D domestically just like multinational corporations will. It makes sense that the credit is agnostic about who is conducting R&D, because there is no reason to favor one type of firm over another.

The OECD paper claims that the R&D credit should be tailored for smaller firms because new firms create proportionately more jobs. While job creation is an important goal, and there is some evidence that new small firms account for an outsized proportion of hiring, encouraging small-firm R&D in order to create jobs misunderstands the link. Successful job growth requires a competitive tradable sector, which creates wealth that spills over into other sectors. But the small firms doing the hiring are not disproportionately R&D firms. Moreover, employment is not the only, or even the dominant reason for R&D tax incentives. Innovation, productivity and high wages are also important. Large firms in fact pay higher wages.

The argument that R&D tax credits have “unintended consequences for balance sheets” is misguided as well. Evidence points to tax credits being effective ways to spur R&D. If the foregone revenue (tax expenditure) from the credit is greater than anticipated, then this is due to greater R&D spending—and in virtually no countries are businesses are spending too much on R&D. R&D spending provides significant boosts to economies by increasing productivity, innovation and human capital. Besides—if the issue is balance sheets, why does the OECD not recommend increasing corporate or personal tax rates (or cutting spending)? In other words, R&D tax expenditures are a form of investment, and if they produce more growth than other kinds of policies, they are worth expanding. Moreover, if designed properly they can pay for themselves in terms of higher tax revenues.

The OECD argument that multinational firms achieve double non-taxation by strategically valuing assets and R&D credits in different countries is also unfounded. R&D tax credits are based on actual expenditures within the countries offering the credit, and firms are not able to move them. Local firms receive the same credit as multinationals, so it is unclear why a bias toward multinational corporations would exist.

Finally, the call for balance between R&D tax credits and direct government funding may be appropriate in some countries where R&D tax credits more than double direct government funds, such as France, Canada, and Australia, but in the United States and many other countries it is direct government funding that vastly outweighs tax credits. In fact, US government direct funding for R&D is more than four times higher than funding for the R&D credit. In the United States we should increase the R&D tax credit.

The OECD report also ignores an important aspect of R&D tax credit design: incrementalism. In the United States and some other countries the tax credits are designed to incentivize companies to increase their R&D expenditures with respect to the previous year. The U.S. rewards expenditures that are 50% of the base year expenditures, for example. While it can be difficult to design the correct balance between incentives to increase expenditures and incentives to maintain expenditures, incremental tax policies are nevertheless an important tool. Thus, it would have been better for the OECD to argue that all nations should adopt the kind of quasi-incremental system in the United States.

While the OECD report portends to present the R&D tax credit as an important tool for increasing growth and innovation, the cautionary note struck by the report will end up both misleading and discouraging policymakers. We cannot forget the important role that R&D tax incentives have come to play in our innovation ecosystem: increasing productivity, enhancing competitiveness and creating value and (indirectly) jobs for workers. Clearly, R&D tax credits must be well-designed, but there are many countries in which they can be improved and expanded. What is less clear is that the OECD’s advice will help assure that is the case.

Print Friendly, PDF & Email