Congress should make the R&D tax credit permanent

Enacting legislation is seldom pretty. However, Congress has an opportunity now to permanently extend the research and development (R&D) tax credit. While neither the timing nor the substance of the proposed legislation is perfect, Congress and the president should enact the legislation this year. Although it would be preferable to treat this business provision on its own, political realities dictate that Congress should take the best deal that it can and make the credit permanent.

Congress first enacted the research and development tax credit in 1981. However, Congress made it temporary in order to reduce the immediate budget costs. When it was first passed, the tax credit was the most generous in the world. But as other countries have realized the importance of attracting innovation and investment to their shores, its relative impact has slipped. Today, the United States ranks just 27th out of 42 countries in the value of the credit to business.

Since original passage, the credit has been extended 14 times and has actually expired eight times before being renewed. The last extension expired at the end of 2013. In order to provide the credit to R&D expenditures made in 2014, Congress would have to make any extension retroactive. And because we are near the end of 2014, any delay in extending it retroactively will significantly complicate the tax returns for companies that want to claim the credit.

This temporary status has had two main effects. The first is that companies face a tremendous amount of uncertainty in calculating the after-tax cost of R&D investments. They must often commit to investments long before they know for sure whether the credit will be extended and what form it will take. The second and probably most important effect is that the tax credit has gotten lumped in with a wide variety of other tax provisions, including an important provision on Medicare reimbursement, that also expire on a regular basis. Most of these "tax extenders" are relatively minor pieces of legislation that have very little effect on economic activity. They merely transfer money from one group to another, often for reasons that make little policy sense. Good examples include employer payments for mass transit and parking, expensing for film and television production, and faster depreciation of race horses.

The annual exercise to extend the several dozen provisions for one or two more years and find ways to pay for them has occupied an inordinate amount of Congress's time each year, even as it struggles to negotiate a final budget for the current fiscal year. And the inclusion of the R&D tax credit in this group has increased the chance it will end up lapsing.

This situation is particularly concerning given the great deal of economic research which shows that the credit has a significant effect on the amount of research that private companies conduct in the United States. In 2012, the Obama administration concluded that each dollar of foregone tax revenue through the credit causes firms to invest at least a dollar in R&D.

It would be nice if Congress paid for the cost of any new tax cuts or spending increases as it went along. The United States does face a serious long-term debt problem that may already be having a negative effect on people's willingness to invest for the long term. But there are several reasons why this should not preclude a permanent extension of the R&D tax credit now, even if it is not paid for. The first is that the debt problem is not the most important threat that the economy faces. Sluggish growth and low investment levels are. Efforts to restore fiscal balance should focus on the debt-to-gross domestic product ratio, not the nominal value of debt. We cannot solve our fiscal problems solely by cutting spending and raising taxes. We also need to increase the amount of economic activity and the R&D tax credit helps us do that. Because traditional budget estimates do not measure dynamic effects, the official cost of extending the tax credit is much greater than its actual cost to the Treasury. Indeed, the Information Technology and Innovation Foundation (ITIF) has estimated that the credit pays for itself after about 15 years.

The fetish with paying for tax extensions is also largely the result of an anomaly in the budget laws, which treat spending and taxes differently. Many spending programs also expire on a regular basis. Yet budget laws do not require Congress to pay for the cost of legislation that extends them. In fact, if an extension reduces the size of the program, Congress is actually credited with net savings. But the same laws assume that tax provisions automatically expire as scheduled. Thus legislation that merely extends existing tax law is treated as if it is actually an important policy change. The net effect of this is to favor spending increases over tax cuts.

Lastly, by making the R&D tax credit permanent, Congress would make it easier for Congress to pass comprehensive tax reform, because the tax committees would not be burdened with the cost of permanently extending what is actually an important element of current policy. Although some worry that passage might make it more difficult to balance the budget in 10 years, the arguments above indicate why this may not be a significant problem.

The tentative deal that the administration is reported to have opposed last week would have apparently cost $450 billion over 10 years on a static basis. Yet $160 billion of this cost was due to the R&D tax credit. Economic growth caused by increased research and development benefits all Americans. Congress should make this the last year in which companies have to guess whether the R&D tax credit will be extended and in what form.

Atkinson is president of the Information Technology and Innovation Foundation (ITIF) and Kennedy is a senior fellow with ITIF and former chief economist for the U.S. Department of Commerce.