Recent Research: What Kinds of Publicly Funded R&D Projects Fail?

SBIR projects are less likely to fail if research teams are smaller, have more experience and include women investigators, according to a new working paper by Albert N. Link and Mike Wright. The authors also found that larger SBIR awards lower the chances that a project will be discontinued before completion. While the study focuses on projects supported through federal SBIR programs, the findings could have implications for other kinds of public R&D support.

Link and Wright reviewed data from 1,878 Phase II projects funded through the U.S. Small Business Innovation Research (SBIR) programs at the Departments of Defense and Energy, the National Institutes of Health, the National Science Foundation and NASA. Of those projects, about a third failed, meaning they were not completed and were discontinued by the research team. A variety of reasons were given for these failure. The most common (24 percent of cases) was that the potential market for the new technology was too small. Another 15 percent reported that insufficient funding was available to complete the project, and 14 percent said that technical problems stood in the way.

An analysis of demographic data revealed a number of factors that contribute to the failure of research projects. First, teams with less experience are less likely to successfully complete their project. Specifically, teams without SBIR experience in R&D projects directly related to the technology being funded are more likely to fail. Second, teams receiving smaller awards are more likely to fail. This could be due to financing difficulties or the fact that riskier projects were given less support. Third, larger firms were more likely to fail, which Link and Wright attribute to a lack of flexibility in handling the unexpected turns of a long-term R&D project. Finally, projects with a female principal investigator (PI) were less likely to fail. The authors abstain from offering a theory about the success of women PIs.

Link and Wright acknowledge that more research is needed, and additional data on the backgrounds of PIs lead to more useful research. However, they suggest that since experience appears to play a key role in the success of research projects, mentoring programs could have a big impact on future projects. Experienced SBIR researchers could be paired with current teams in their knowledge area to pass on valuable information about project planning, budgeting and market assessment.

If mentoring is an effective practice in lowering the likelihood of failure, state and regional SBIR programs could play a key role in supporting research teams by connecting them to the right people. This could also prove useful in non-SBIR research projects funded by state programs.

Download On the Failure of R&D Projects at: https://ideas.repec.org/p/ris/uncgec/2015_003.html.