Does double-blind peer review effectively correct for gender disparities in research funding?

New mechanisms for mitigating biases in grant peer review have been proposed. One such mechanism is blinding reviewers to the identity of grant applicants, but evidence of its effectiveness remains scarce. We leverage detailed information on more than 2,000 applications to the Villum Foundation’s “Villum Experiment”, a double-blinded grant scheme, to evaluate how blinding can reduce gender bias in research funding. We show that small gender differences can persist despite an effective double-blinded evaluation of applications. These differences are likely caused by differences in gender compositions across disciplines, and a strong underrepresentation of highly experienced women among the applicants and in the population in general. Our analysis highlights how policies aimed at leveling demographic disparities in research funding rates may eliminate direct bias but fall short of confronting broader structural inequalities.