Evaluations Help Ensure Effectiveness of State Research and Development Incentives
Policymakers can use regular reviews to improve and strengthen policies
Research and development (R&D) is a key driver of the American economy. Policymakers at the state and federal levels benefit from evaluation processes that can assess whether spending on incentives to encourage more R&D is working as intended. And state officials can learn from the experiences of other states on this front as they introduce or update their incentive programs.
The Pew Charitable Trusts has explored the effectiveness of various tax incentives since 2012. That work sheds light on the range of approaches being used by state policymakers.
Overall, R&D work accounted for nearly $890 billion in estimated activity in 2022, driving innovation across the economy. And tax incentive programs have become a widely used policy tool because the market on its own doesn’t produce sufficient R&D. Without government intervention, firms might be reluctant to invest in R&D because it can be risky, and the benefits tend to flow more broadly than to a single company.
Several studies of the federal program have shown it to increase R&D activity nationwide, though the increase is generally low. Results have been more mixed at the state level. Evaluations have questioned whether state R&D credits can meaningfully raise a state’s overall R&D activity. For instance, a 2022 evaluation of Virginia’s three incentives found that, while spending and activity increased among recipients, the credits were too small to meaningfully affect state-level R&D. Other evaluations, such as Georgia’s and Iowa’s, have raised questions about whether state incentives spur new activity or shift existing activity across state lines.
Evaluations can also offer considerations for policymakers seeking to refine these programs and for other analysts studying their effects.
In terms of design, policymakers should always start by clarifying the program’s goals and making sure that the design supports them. For instance, most states’ credits reward R&D spending that exceeds historical levels to encourage new R&D. Among those, Georgia offers an income tax credit equal to 10% of year-over-year increases in qualified expenses.
Rhode Island uses another approach. The state’s expense credit applies to qualified spending that exceeds a baseline amount. State law features a two-tiered structure: a 25% credit rate for the first $25,000 and a 16.9% credit rate for amounts exceeding $25,000. Rhode Island’s evaluation recommended an alternative design—providing credits for firms’ last dollar of research expenditures rather than the first. Maryland, meanwhile, separates its credits into two categories: a base credit to sustain ongoing R&D activity and a growth credit as an incentive for new activity.
Policymakers should also consider whether their incentives are set up to protect state budgets from unexpected costs. Nebraska created an R&D incentive in 2005 but did not put limits on the program’s cost. Then in 2020, the program exceeded cost estimates by $5 million. Caps are one way to avoid these situations. Another option could be to issue credits on a competitive basis or first-come, first-served model. Several states that cap their incentives prorate their credit awards when nearing the statutory limit so that every qualified business receives some level of assistance.
Another decision that policymakers must make while balancing fiscal considerations is whether state credits will be limited to a firm’s tax liability or whether they will be refundable or transferable. For instance, some states allow credits to be refundable only for businesses of a certain size—Maryland’s credit restricts refundability to businesses with 50 employees or fewer. Others, such as Pennsylvania, allow credits to be transferable, enabling recipients to sell them. Policymakers should be aware of potential economic leakage in this exchange: The Pennsylvania evaluation found that sellers receive only 93 to 94 cents per dollar of credit value—as well as the potential fiscal risks of allowing for broad refundability.
State policymakers also should consider whether to tie their tax credit to the federal program. Doing so lowers compliance costs and administration for states and businesses. For instance, in Nebraska, companies qualifying for the federal R&D credit can receive a state credit equal to 15% of the federal amount simply by submitting a claim with their tax return. But piggybacking on the federal credit can limit a state’s flexibility to design the credit.
Another consideration should be how credit recipients support other state priorities. To support small businesses, Virginia offers an R&D expenses credit specifically for firms with annual expenditures of less than $5 million. To encourage firms to partner with colleges and universities, Nebraska provides a higher credit rate of 35% for research conducted on campus, compared with the standard 15%. Oklahoma, meanwhile, redesigned its program to accept only applicants that address needs within one of the four industries identified as state priorities. Such targeted approaches can help align incentive programs with broader state objectives.
In terms of studying the effects of R&D incentives, several state reports note that some standard methods used to analyze other incentive programs may not work well. For example, an Arkansas evaluation found that the R&D credit cost more than the benefits it created. But the report highlighted that the credit’s goal was long-term growth, suggesting that a short-term analysis may be inappropriate.
Meanwhile, Florida’s evaluation of the state’s Innovation Incentive Program similarly emphasizes that R&D tends to have a low immediate return on investment because the benefits for participating industries may need more time to become apparent than in short-term calculations. Florida therefore requires program recipients to demonstrate a break-even economic benefit to the state over a period of 20 years.
Evaluators may also need to account for specific underlying characteristics that influence a state’s R&D activity. In Iowa’s second evaluation of its R&D credit, the state Department of Revenue initially found evidence supporting the credit’s positive effects by estimating the R&D credit’s influence on research input and output. However, after accounting for variables such as industrial composition, demographics, and state fiscal policies, the analysis told a different story: The credit was not associated with meaningfully higher research inputs or outputs.
In addition to accounting for external influences when studying R&D credits, Iowa’s evaluation also recommends assessing firm-level data. This approach could involve comparing businesses that previously claimed the credit and no longer qualify with those that either consistently claimed the credit or never did.
R&D credits are a widely used policy tool. That’s why they warrant routine efforts to analyze their effectiveness. Evaluations highlight open questions regarding the credits’ success but point to opportunities to improve their impact. Policymakers should rely on this evidence to determine whether to offer such credits, as well as how to structure them and how to refine the programs to increase their effectiveness.
Elizabeth Gray is an associate and Alison Wakefield is an officer with The Pew Charitable Trusts’ state fiscal policy project.
People
