Industry effects and firm effects: No effect is an island

At the heart of business research is the identification and prioritization of the factors that create competitive advantage. Previous studies that decompose observed firm performance into industry and other effects have not considered the higher moments of the performance distribution (i.e., the excess kurtosis), nor estimated the correct null hypothesis level of those effects (i.e., of the industry effect). We address these issues to find that: (1) the industry effect impact on firm performance is likely overstated; and, (2) the interaction between the calculated component firm and industry effects is likely one that includes synergistic effects (i.e., the standard additive model of decomposition is likely mis-specified). We offer a new model, test it, and then discuss the resulting implications.