Doctoral programs in the humanities and related social sciences are characterized by high attrition and long time-to-degree. In response to these long-standing problems, the Andrew W. Mellon Foundation launched the Graduate Education Initiative (GEI) to improve the structure and organization of PhD programs, and in turn reduce attrition and shorten time-to-degree. Over a 10-year period starting in 1991, the Foundation provided a total of $80 million to 51 departments at 10 major research universities. This paper estimates the impact of the GEI on attrition rates and time-to-degree. Our analysis is based on a competing-risk duration model and student-level data spanning the start of the GEI, including data on students at a set of control departments. We estimate that, on average, the GEI had modest impacts on student outcomes in the expected directions: reducing attrition rates, reducing time-to-degree, and increasing completion rates. The overall impacts of the GEI appear to have been driven in part by reductions in cohort size, increases in financial aid, and increases in student quality.