Abstract
We estimate the impact of the NZ Marsden Fund on research success, by comparing the subsequent performance of funded and unfunded researchers. We control for selection bias using the Marsden panel rankings. We find that funding is associated with a 6%-15% increase in publications and an 11%-22% increase in citation-weighted papers for research teams. We attempt to pin down the dynamic structure of this effect by looking at the research trajectories of individual researchers, but this model yields inconsistent results. We find no systematic evidence that Marsden panel rankings are predictive of subsequent success.