Technological innovation is the major factor driving economic and income growth in America today, and one of the key federal (and state¹) policy tools used to stimulate this innovation is the research and experimentation tax credit. Commonly known as the R&D credit, it provides tax incentives for firms to invest in research and development. But partly because of how the cost would have to be scored in the budget, the credit has never been made permanent. Instead, since its enactment in 1981, the credit has been extended nine times and the lack of permanence is only one of many factors that limit the credit's effectiveness.

The new realities of technological innovation require not only that the credit be made permanent, but also that it be updated to more effectively promote innovation in the New Economy. A number of bills from the 105th Congress, and now in the 106th, would address one or more aspects of the credit. But one bill, S 951, introduced this month by Senators Pete Domenici (R-NM), Jeff Bingaman (D-NM), Joe Lieberman (D-CT), Bill Frist (R-TN) and Olympia Snowe (R-ME) represents the kind of comprehensive modernization of the credit that is needed. In the House, a companion bill (H.R. 1682) was introduced by Representatives Heather Wilson (R-NM), Harold Ford (D-TN), Tom Udall (D-NM), and Joe Skeen (R-NM). Not only do the bills propose making the credit permanent, they also include a number of important reforms to make the R&D tax credit a more effective tool for stimulating research in the New Economy. These provisions include: modifying the credit's two-tier structure to give more firms access to the higher marginal rate; creating a collaborative R&D credit; and making it easier for new entrepreneurial firms to take advantage of the credit.

The Joint Committee on Taxation has estimated that the comprehensive improvements of the sort envisioned by the Domenici- Bingaman bill can be achieved for $38 billion over 10 years ($3.8 billion per year), compared to approximately $28 billion for just making the credit permanent and not reforming it.

Innovation is the Key to Faster Economic Growth in the New Economy

There is clear consensus among economists that investment in research fosters the lion's share of our economic and productivity growth. At least two-thirds of per capita economic growth stems directly from technological innovation.² A one percent increase in the nation's investment in research increases productivity 0.23 percent.³ This is important because slow growth in productivity in the last two decades has been the factor most responsible for slow growth in wages for American workers. Boosting technological innovation and adoption is the prerequisite to raising wages; if productivity had increased after 1973 the way it did in the 30 years before, half of all American households would now be earning at least $63,000 instead of the current $37,000. Boosting productivity even a half a percent more than baseline forecasts over the next 30 years can go a long way toward addressing the looming retirement crisis stemming from too few workers supporting too many retirees.

The R&D Credit Addresses a Clear Market Failure

Firms seldom capture all of the benefits of research, particularly riskier, early-stage research. Research results can often be used by more than one firm firms can often utilize the results of others' research to improve their own products and processes at a fraction of the cost of doing their own original research. As result, the benefits to society from a company's R&D investments are significantly higher than the company's own rate of return.

Numerous studies by economists, including the President's Council of Economic Advisers, have found that the public rates of return on R&D are at least twice the estimated private returns.⁴ The large difference between the private and public rates of return means that absent public action, industry under-invests in R&D, particularly earlier stage R&D.⁵ And, significantly heightened business competition in the New Economy is making this even more pronounced. Industry cut its investment in basic research 4.6 percent per year from 1991 to 1995.⁶ This kind of market failure means that government incentives to stimulate private research and reduce the difference between the private and public rates of return are needed to obtain the optimal levels of research in the economy.

This market failure is also why many other nations provide generous R&D tax credits. For example, the Canadian government provides a flat, non-incremental 20 percent R&D credit for large companies, and up to a 35 percent credit for small companies. Provincial government credits are added to these. For example, companies in Quebec receive an additional 20 percent credit for internal R&D expenses, and a 40 percent credit for collaborative R&D with universities or consortia.

The Credit Effectively Stimulates Economic Growth

There is general agreement that the R&D tax credit has been a cost- effective policy tool. For example, the former Congressional Office of Technology Assessment concluded that "for every dollar lost in tax revenue, the R&D tax credit produces a dollar increase in reported R&D spending, on the margin."⁷ Other studies have found even greater benefits, with the economic benefit to tax-cost ratio between 1.3 and 2.0. A recent study by Coopers & Lybrand estimated that if the R&D tax credit were made permanent (as opposed to being repealed), companies would spend $41 billion more on R&D over the next 12 years, boosting productivity and adding $13 billion (in 1998 dollars) in economic output per year by 2010.⁸ Because higher productivity results in lower relative prices, the benefit of the higher productivity triggered by the R&D tax credit would be equivalent to a $47 billion tax cut for families over 13 years. By 2010, Coopers & Lybrand estimates that the credit would generate net higher tax revenues that would offset 47 percent of the "static" revenue loss to the government of the credit. In the long run, they estimate that since the benefits continue to accrue, the credit would pay for itself through increased tax collections. All of these studies, however, are premised on economic analyses of the credit during the 1980s, when most firms could qualify for the incremental credit. It is not clear that, absent reforms in the credit as described below, the cost effectiveness of the credit will remain as positive.

Current Structure of the Credit

The 1999 Omnibus spending bill (P.L. 105-297) retroactively extended the R&D tax credit for one year from July 1, 1998, to June 30, 1999. Under current law, it allows firms to take one of two types of credit. Under the regular credit, a taxpayer's current-year "qualified research expenses," in excess of a specified base amount, are eligible for a 20 percent tax credit. The base amount is a ratio of R&D expenses relative to sales for the years 1984 to 1988. Because the base period is fixed, many firms do not qualify for the credit because their R&D-to-sales ratio is higher in the base period than it is now. This is the case for many firms that have merged, downsized, faced defense spending cuts, or otherwise restructured, or that for a particular reason had either low sales during the period or very high spikes in R&D spending. As a result, in order to stimulate R&D at these kinds of firms, Congress created the Alternative Incremental Credit in 1996. Firms that cannot qualify for the regular credit can take a credit of between 1.65 percent and 2.75 percent of qualified research expenses in excess of 1 percent of a company's average annual gross receipts.

Also, firms started after 1988 can use the regular credit. They just have to contend with the complex base phase-in rules which are unfavorable for a lot of small, high-tech startups.

Changes Needed

If the R&D credit is to play a renewed role in fostering innovation in the New Economy, a number of changes are needed, all of which would be accomplished under the Bingaman-Domenici bill. These include: