The field of leadership research today has limitations in that it has excessively focused on leader-subordinate dyadic relationships and the positive aspects of leader behavior. Under this critical awareness, the present study expanded the leadership perspective to a triadic relationship of "higher-level leader–leader–subordinate" and directed attention to leaders' counterproductive behaviors that undermine organizational effectiveness. The leader's leader (higher-level supervisor) plays an active role in various aspects of leadership enactment, thereby determining the success or failure of leadership. In particular, the higher-level supervisor's abusive behavior toward the leader diminishes the leader's authority and reduces commitment, negatively affecting leadership and consequently lowering subordinates' effectiveness. To investigate this, the present study introduced social exchange theory, the norm of reciprocity, and displaced aggression theory to construct hypotheses positing that the higher-level leader's abusive supervision toward the leader negatively affects subordinates' trust in the leader and leader-member exchange (LMX) through the leader's abusive supervision toward subordinates and transformational leadership as mediating variables. Data for this study were collected from military officers (231 supervisor-subordinate pairs), and the analysis results are as follows. The higher-level leader's abusive supervision toward the leader had a positive effect on the leader's abusive supervision toward subordinates and a negative effect on transformational leadership. Subsequently, the leader's abusive supervision toward subordinates showed a negative relationship with subordinates' trust in the leader and LMX, while the leader's transformational leadership showed a positive relationship with subordinates' trust in the leader and LMX, thereby supporting the mediation effects. Additionally, a negative direct effect was found between the higher-level leader's abusive supervision toward the leader and subordinates' trust in the leader. The analysis results support all of the researchers' hypotheses and provide various theoretical and practical implications for leadership research, abusive supervision, and the trust domain. Discussion of these findings along with limitations and directions for future research were presented.