Chronic stress has been consistently documented to engender persistent changes in the hypothalamic-pituitary-adrenal (HPA) axis reactivity to stress. As a chronic social stressor, bullying victimization is associated with dysregulation of the HPA axis that is typically assessed by salivary cortisol levels. However, the relationship between victimization and cortisol responses is still inconsistent in existing literatures, which have demonstrated that school bullying victimization can both induce persistent elevation of cortisol concentration as hypercortisolism hypothesis suggested or be related to the attenuated cortisol responses as suggested by hypocortisolism hypothesis. Except for the mixed relation between victimization and cortisol responses, little is known about the cortisol responses to stress in the sample of younger children who may experience shorter times periods of being bullied compared with adolescents and adults. As Miller and his colleagues (2007) suggested, the HPA axis tends to exhibit first hyper- and later hypoactivity in response to chronic stressors during the lasting stress process. The present study hypothesized that children who experienced shorter term of bullying victimization would exhibit an elevated cortisol response to the stressor compared to nonbullied children. Specifically, this experiment explored whether cortisol concentration and subjective stress response to the Trier Social Stress Test (TSST) differed between bullied and nonbullied children. A total of 52 children (M = 10.83 years, SD = 0.60 years, 31 bullied children) were selected from 573 students after parent and children consent were obtained. Two phases were included in the experiment. In the first phase, 573 children completed the self-reported questionnaire of victimization. In the second phase, the selected children completed the modified TSST in which they prepared and delivered a free speech followed by a mental arithmetic. During the TSST, six saliva samples and information on subjective stress for each child were collected. The present study conducted repeated-measures analyses of variance (ANOVAs) mainly on their data using SPSS 22.0. The results indicated that the modified Trier Social Stress Test elicited a subjective stressful response and a curve cortisol response in both bullied and nonbullied children. The cortisol levels over the course of TSST were higher in bullied children relative to nonbullied, especially after the stress task removal. However, there were no significant differences in subjective stress response to the TSST between bullied and nonbullied children. In conclusion, the present study suggested that bullying victimization can promote dysregulation of the HPA axis by exaggerated cortisol responses (hypercortisolism) to the standardized laboratory TSST. The results among Chinese children supported the hypothesis that the activity of HPA axis when coped with the chronic stress may undergo a transition from hyper- to hypoactivity in response to stressors.