Those at risk from natural disasters are typically under-protected, possibly because they expect benefactors such as governments and donors to come to their aid. Yet when relief comes, it is often insufficient, delayed or misallocated. Benefactors may wish to commit to provide an efficient amount of fast well-targeted relief, and leave the rest up to recipients, but such commitments are difficult. This article analyses how transferring risk to third-parties such as private insurers may help resolve these commitment problems. Using a simple model of disaster risk finance is used to identify three distinct commitment problems and then show how various properties of risk transfer schemes can help to resolve these problems. The paper illustrates how these commitment problems play out using examples from around the world, and demonstrates where risk transfer schemes seem to have helped in practice. Overall, the findings show that the benefits of such schemes depend on the relative severity of the different commitment problems.