We report findings from a quasi-experimental evaluation of the recently implemented $5000 retention bonus program for effective teachers in Tennessee’s priority schools.  We estimate the impact of the program on teacher retention using a fuzzy regression discontinuity design. We exploit a discontinuity in the probability of treatment conditional on the composite teacher effectiveness rating that assigns bonus eligibility.  Although we find that estimates are sensitive to functional form and bandwidth selection, point estimates for the main effect of the bonuses are consistently positive across all specifications, and for teachers of tested subjects the program appears to have an effect that is generally both statistically and substantively significant. Implementation concerns, including the timing of application process and observed noncompliance in bonus distribution, present obstacles for both the program’s effectiveness and its evaluation.