Print this page

General government

Opportunities exist to improve the targeting of the $6 billion research tax credit and reduce forgone revenue

Why Area Is Important


Since 1981, the research tax credit has provided significant subsidies (an estimated $6 billion for fiscal year 2011) to encourage business to invest in research and development. The credit, which has been a temporary provision since its inception, was most recently renewed at the end of 2010 and is scheduled to expire after December 31, 2011. The Department of the Treasury and the Internal Revenue Service play key roles in issuing guidance to clarify what types of spending qualify and ensuring that taxpayers adequately support their credit claims.

Two factors—the definition of research expenses that qualify for the credit and the credit's design—are important in targeting the subsidy in a manner that increases the social benefits stimulated per dollar of tax revenue foregone. (This ratio of benefits to forgone revenue is a key measure of credit's cost-effectiveness.) The research credit has always been an incremental subsidy, meaning that taxpayers earn the credit only for qualified spending that exceeds a defined threshold, known as the base spending amount. The credit's design is most cost-effective when the base spending amount accurately reflects the amount of spending that a taxpayer would have done anyway (in the absence of the credit).

The figure below compares the effects of a hypothetical incremental credit with a perfectly accurate base to a flat credit, which has no base spending amount. The flat credit gives the taxpayer 20 cents for every research dollar spent, while the incremental credit gives 20 percent for only the amount of spending above what the taxpayer would have done anyway.


A Comparison of an Incremental Credit with a Flat Credit

A comparison of an incremental credit with a flat credit

Both types of tax credit provide the same 20 percent reward for each additional dollar of qualified spending (referred to as "marginal incentive"). In each case that incentive encourages the taxpayer to increase spending for research by $100. However, the flat credit is less cost-effective for the government because it also gives the taxpayer a $200 windfall for conducting research that would have been done anyway.

The difficulty in designing an incremental credit to be as cost-effective as the one in the figure is to develop rules for computing the base spending amount so that the base accurately represents what the taxpayer would have spent anyway. GAO testified as early as 1995 that the computation method in place at that time had grown inaccurate and should be updated. An alternative approach for computing base spending (the alternative simplified credit) has been added but, the older computation option—commonly known as the regular credit—still has not been updated.

What GAO Found


The research tax credit, as currently designed, distributes incentives unevenly across taxpayers and provides many recipients with windfall benefits, earned for spending that they would have done anyway. The disparities in incentives can lead to an inefficient allocation of investment resources across businesses and the windfall benefits represent foregone tax revenue that does not contribute to the credit's objective.

In November 2009 GAO estimated that, due to shortcomings in the computation of base spending, the research tax credit has provided some taxpayers with more than a 10 percent reduction in the cost of additional research, while providing other research-performing taxpayers with a disincentive to increase their research in the current year. Moreover, some taxpayers earned credits on as much as 50 percent of their total research spending, even though the most favorable empirical estimates of the credit's stimulative effects suggest that less than 15 percent of that spending was actually new spending that they would not have done in the absence of the credit.

An important cause of these problems is that, as GAO has previously reported, the base spending amount for the regular version of the credit is extrapolated from the amount of research spending that taxpayers did as long ago as the early 1980s. That base is a poor measure of the spending that a taxpayer would be doing now in the absence of the credit. The alternative credit option uses a more current spending history for computing the incremental credit, but it provides lower incentives for new research, even as some taxpayers can receive larger windfalls than they would get from the regular credit.[1]

[1]As the figure comparing basic hypothetical credit designs above illustrates, the rate of incentive and the amount of windfall a credit provides are independent of each other.

Actions Needed

Based on analyses of numerous design alternatives in its 2009 study, GAO found that the targeting of the research tax credit could be improved by eliminating the regular credit and adding a minimum base amount (equal to 50 percent of a taxpayer's current spending) to the method for computing the alternative credit. GAO found that an alternative simplified credit with this modification could provide the same average incentive to taxpayers as the current version of that credit, but at a lower revenue cost by reducing windfalls. Cost reductions exceeded 3 percent under most of the alternative assumptions GAO used in its 2009 analyses and exceeded 1.4 percent under all assumptions that GAO considered likely.

The elimination of the regular credit not only would improve targeting, it would also significantly reduce compliance and administrative costs by eliminating the need for taxpayers to keep (and for IRS to review) records dating back to the 1980s.

Framework for Analysis

The information contained in this analysis is based on the related products under the "Related GAO Products" tab.

Area Contact

For additional information about this area, contact James White at (202) 512-9110 or