A common justification for HOPE-style merit-aid programs is to promote and reward academic achievement, thereby inducing greater investments in human capital. However, grade-based eligibility and retention rules encourage other behavioral responses. Using the longitudinal records of all undergraduates who enrolled at the University of Georgia (UGA) between 1989 and 1997, we estimate the effects of HOPE on course enrollment, withdrawal and completion, and the diversion of course taking from the academic year to the summer, treating non-residents as a control group. First, we find that HOPE decreased full-load enrollments and increased course withdrawals among resident freshmen. The combination of these responses results in an 11\% lower probability of full-load completion and an annual average reduction in credits completed of 1.0. The latter implies that between 1993 and 1997 Georgia-resident freshmen completed 15,710 fewer credit hours or 3,142 individual course enrollments than non-residents. Second, the scholarship's influence on course-taking behavior is concentrated on students with GPAs on or below the scholarship-retention margin. Third, the effect increased as the income cap was lifted and more students became eligible for the award. Fourth, these freshmen credit-hour reductions represent an intertemporal substitution, not a general slowdown in academic progress. Finally, residents diverted an average of 1.65 more credits from the regular academic year to the first summer term after their matriculation, which amounts to a 72\% rise in summer course taking.
