Association for Public Policy Analysis & Management

We use administrative data to investigate the effect of the New Mexico Legislative Lottery Scholarship (NMLLS) on graduation rates at the University of New Mexico.  The NMLLS is available only to New Mexico residents.  Effects of the scholarship are estimated by comparing pre- and post-program outcomes for residents and nonresidents.  Although the program is not associated with any overall impact on completion rates, meaningful effects appear when the model is disaggregated by student ability.  Large gains (of 18-21 percent) are estimated for students with high school grade point averages in the top half of the grade distribution.  By contrast, program-eligible students in the bottom half of the grade distribution were either unaffected or graduated at lower rates compared to similar ineligible students.  To address observable differences between resident and nonresident students, we also perform difference-in-differences matching, where effect sizes, although smaller in magnitude, support the initial findings.  Completion effects may be driven by compositional changes in the resident student population.