Leader–member exchange (LMX) theory originally has strong theoretical ties with social exchange theory (SET) (Liden, Sparrowe, & Wayne, 1997). Still, several scholars have argued that the connection to SET has been lacking in the empirical work on LMX (Gottfredson, Wright, & Heaphy, 2020; Scandura & Meuser, 2021). Through three different papers, I hope to contribute to constructive development of the LMX field by overall strengthening the connection between LMX and SET. The first article provides a literature review of research to date that has applied a twodimensional measure of the LMX relationship, namely social and economic LMX (Kuvaas et al., 2012). With this review, we aim to expose a measure that more coherently corresponds to SET, while also elucidating its implications for various employee outcomes. The results reveal that economic LMX provides unique explanatory variance on various employee outcomes beyond social LMX alone (Andersen, Buch, & Kuvaas, 2020), thus allowing for enhanced understanding of the LMX relationship. In the second paper, I utilize a novel statistical technique, ‘necessary condition analysis’ (Dul, 2016), and insights from SET, to extend the understanding of how both individual level LMX and dyadic level LMX relate to employee outcomes. The findings reveal that both individual level LMX, particularly leaders’ perception of the LMX relationship, and dyadic level LMX are necessary conditions for employee outcomes. Thus, these findings provide enhanced insights regarding the critical role of the LMX relationship in employee outcomes. Leaders differentiate among group members by providing some employees with more resources (e.g., support, attention, bonuses) than other employees (Graen & Scandura, 1987). Despite, resources’ central place in the exchange process, it has largely been a neglected area of scientific inquiries among LMX scholars (Law-Penrose, Wilson, & Taylor, 2015). By placing an explicit focus on resources in the third article, we investigate whether it matters what and how resources are allocated differently by the leader on interactional justice. The overall finding is that the association between LMXD and interactional justice is weaker when socioemotional resources are based on need compared to when based on equity.