Despite the recurrence of scandals such as the ‘diesel-gate’ in bringing the focus into employee social responsibility (ESR) time and again, serious research into ESR has started only recently. Since there are considerable divergences in its conceptualization, we first attempt to bring clarity to the construct by laying its theoretical foundations by extrapolating Kohlbergian and neo-Kohlbergian moral theories to the social responsibility domain. Having defined the construct, through empirical research we then proceed to accomplish three broad objectives. First, in causal research, we investigate how business goal difficulty (BGD) impacts intention for ESR, mediated through a cognitive bias called focalism. Across two experimental studies, we establish that BGD predicts focalism, which in turn mediates the negative relationship between BGD and intent ion for ESR. Second, we investigate how contextual and dispositional moderators, specifically, ethical leadership (EL) of supervisors and reflective moral attentiveness (RMA) of employees, influence the aforesaid mediated relationships. The study reveals that both EL and RMA did not moderate the BGD-focalism relationship, whereas RMA moderately weakened focalism-intention for ESR relationship. These findings demonstrate an implicit mechanism by which difficult goals can cause ESR violations. Further, our research also indicates that focalism is a powerful bias, not easily attenuated by contextual and dispositional moderators. In the second part of our empirical research, we meet the third objective of our study by developing an ESR scale. The ESR scale is developed as a superordinate multidimensional measure incorporating the dimensions of concern orientation, norms adherence orientation, sociocentric orientation, and perseverance. Exploratory and confirmatory factor analyses on mult iple employee samples across several organisations and two countries corroborate the hypothesized factor structure. The second-order multidimensional ESR scale demonstrated good reliability and validity. The ESR scale exhibited configural and metric invariance between English (India) and German versions. The predictive validity of the ESR scale is also established by testing the hypothesized relationships between paternalistic leadership dimensions and ESR through structural equation modelling. The ESR scale contributes to both theory and practice.