Sage Publications/RM Division Best Paper Award Nominee
Management researchers often rely on instrumental variables to address endogeneity bias. To correctly estimate the average causal effect of a predictor on an outcome, instrumental variable estimation hinges on two key assumptions: Relevance and exogeneity of the instrument. However, management scholars generally overlook a third critical assumption—effect homogeneity. This paper reviews this assumption, illustrating how its violation can bias instrumental variable estimates, exploring its plausibility in typical management research settings, and discussing possible solutions. Finally, it makes practical suggestions on how to estimate, interpret, and report instrumental variable analyses in contexts where homogeneity is violated, offering guidance for more rigorous and transparent causal inference in management studies.
