China’s rapid urbanization has led to unprecedented growth of rural-to-urban migrant children. Although previous studies have shown that migrant children report more peer victimization and internalizing problems, limited studies have yet examined associations between peer victimization and internalizing problems, as well as the direction of these relationships. Therefore, the current longitudinal study aimed to explore characteristics of peer victimization and associations between peer victimization and internalizing problems over a two-year period. A sample of 426 rural-to-urban migrant children was recruited from one migrant children’s primary school (n = 185) and two public primary schools (n = 241) in Beijing, China. Of the 426 participants, 157 (36.9%) were girls and 269 (63.1%) were boys. The sample showed moderate attrition over a two-year period, with 281 (66.0%) participants completing all surveys during the four waves of assessment. Quantitative surveys were administered to participants in October of 2011(Wave 1), June of 2012 (Wave 2), October of 2012 (Wave 3), and June of 2013 (Wave 4). At each wave, participants completed a self-administered questionnaire that requested information on demographic variables, experiences of various types of peer victimization (i.e., physical, verbal, relational, property), and internalizing problems (i.e., social anxiety, depression, loneliness). Results indicated high prevalence of perceived peer victimization, with participants reporting the prevalence of the four types of peer victimization (i.e., physical, verbal, relational, property) to be 76.5%, 77.8%, 71.4%, and 63.2%, respectively. Boys reported experiencing more physical victimization than girls, and migrant children who moved more often reported experiencing more peer victimization and greater internalizing problems. Migrant children enrolled in public schools reported less peer victimization and fewer internalizing problems than their counterparts enrolled in the separate migrant school. Cross-lagged analysis showed a significant bidirectional relationship between peer victimization and internalizing problems when controlling for key demographic variables (i.e., gender, grade, socio-economic status, and frequency of moves). Specifically, peer victimization in Wave 1 significantly predicted depression and loneliness in Wave 2, and loneliness in Wave 1 also significantly predicted peer victimization in Wave 2. In addition, internalizing problems in Wave 2 significantly predicted peer victimization in Wave 3, which, in turn, increased reported internalizing problems in Wave 4. Multi-group cross-lagged analysis showed that the bidirectional relationship between peer victimization and depression, as well as between peer victimization and loneliness, did not differ by school group (i.e., migrant or public). However, the association between peer victimization and social anxiety was stronger for children enrolled in the migrant school than for children enrolled in public schools. These findings contribute to our understanding of the complex relationships between peer victimization and internalizing problems among rural-to-urban migrant children in China. Moreover, the findings have important implications for the development of prevention and intervention programs for this at-risk group. Programs that aim to improve the mental health of migrant children and to reduce incidents of peer victimization among this vulnerable group may be helpful in breaking the detrimental cycle between peer victimization and internalizing problems.