Every year, governments pump large sums of money into national and regional ecosystems to stimulate innovation. It is generally accepted that the allocation of such funding should be done in a competitive way, with applications from different organizations being reviewed by a neutral committee of experts who make a robust selection of projects that deserve support. In addition, there is a strong belief that politicians should be kept at a distance as much as possible in these decision-making processes. Therefore, projects which are selected in a competitive way are assumed to perform better than projects which are selected in a political way. To date, however, empirical evidence for this hypothesis has been lacking.

In a recently published article in the leading innovation journal Research Policy, Holmer Kok, Dries Faems and Pedro de Faria empirically explore this hypothesis, examining the outcomes of 321 R&D projects funded by the U.S. Department of Energy’s Hydrogen Program. Up until 2011, projects in this program could not only receive funding by means of a competitive selection process, but also by being earmarked by a U.S. member of Congress.

The researchers find that earmarked projects receive considerably lower peer review evaluation scores than non-earmarked projects. However, they also report that earmarked projects do not consistently underperform in terms of publications and patents. In other words, whereas reviewers evaluate earmarked projects as being inferior, this inferiority is not clearly present in tangible outcomes that reflect the productivity and impact of these projects. Conducting additional analyses, the researchers provide indications that this misalignment between reviewer evaluations and tangible outcomes might be driven by the existence of a bias of peer reviewers toward earmarked projects.

Although the researchers do not find a consistent underperformance of earmarked projects in terms of tangible outcomes, they acknowledge that there might be other reasons why societies in general, and policy makers in particular, prefer to stay away from politically driven selection processes. However, the researchers claim that when actors want to defend a ban on earmarking, they need to rely on the appropriate argumentation and cannot simply assume that politically selected projects underperform competitively selected ones.