This study examines two independent but related questions on the effectiveness of the free semester policy. In order to confirm the validity of public concerns that the free semester may stimulate the demand for private tutoring, this study analyzed the causal effect of the free semester on the participation in and expenditure on private tutoring. Data for the analysis was drawn from the Private Tutoring Expenditure Survey, which includes information on 178,213 middle school students surveyed during 2009-16 by Statistics Korea. Exploiting the cross-region and over-time variation in proportion to middle schools implementing the free semester, the analysis revealed that, on average, the free semester affected neither the participation in nor expenditure on private tutoring. However, a clear distinction in terms of the policy effect was found between income groups. That is, the policy induced high-income families to spend more on private tutoring while the opposite was true for middle-income households. The contrasting effect by income level is mainly because the demand for private tutoring is more complementary to school education for the middle-income group than for their high-income counterpart. Indeed, these results suggest that the free semester policy which accompanies the reduction in hours of academic courses in school may widen the gap in private tutoring investment between high-income and middle-income families. The impact of the free semester on students’ creativity, sociality and happiness was also analyzed, taking into account that an important goal of the policy is to improve non-cognitive skills. Specifically, a comparison was made of the changes in creativity, sociality and happiness during the 2016 fall semester (late August to late December) between 365 first-year students in four middle schools in Daegu city who had free semesters and 350 second-year students at the same schools who had regular semesters. The within-school and cross-grade comparison shows that the free semester improved students’ sociality, measured by the number of close friends; but not their creativity, measured by the remote associate test and; happiness, measured by the UNICEF’s life satisfaction index. It was also found that free semesters increased the frequency of classroom discussions, group projects, and extra-curricular activities. These results jointly suggest that the free semester policy may promote sociality in students by encouraging teachers to adopt teaching practices that emphasize student-to-student interaction. However, since the evidence found in this study is based on a small sample of a specific region, and because it is difficult to rigorously test the validity of the key assumption of the research design (comparability between grades), care should be taken in interpreting the results as a general effect of the free semester policy.