Many school districts have turned to summer school as a strategy for addressing persistent declines in student achievement after COVID-19. We examine the impact of 2023 summer programs on academic recovery across 400,000 students in eight large U.S. districts. Using value-added models, we find modest but significant improvements in student math achievement (average effect = 0.024 SD), and no significant improvements in reading. Effects were similar to 2022, despite small declines in participation and dosage. In four districts, we also examine differences in summer 2023 impacts for students who attended the districts' summer school the prior year and those who did not. Students attending for a second consecutive summer generally benefited as much or more than first-time attendees.