Performance effect of multiple control forms in a Lean organization: A quantitative case study in a systems fit approach

Over the last decades Lean has developed into a prominent management philosophy reaching beyond shop floor tools. However, substantial support of performance effects from Lean is still scarce and at best with mixed results. Recently, research has turned its focus towards perceiving Lean as a control package. In this paper we present statistical support for enhanced performance coming from Lean. Furthermore, our results strongly support the perception of Lean as a set of multiple control forms (output, behavioral, and social controls) that complement each other. Therefore, performance is increased if the average level of control forms is increased, and performance is further increased if the control forms are balanced at the same level representing a complementary effect between them. Our data are archival data spanning multiple years in a strong Lean organization. The dependent performance constructs are segmented, cost-allocation free, actual accounting numbers – thus the reliability is high. Moreover, we provide a general refinement to the statistical approach in testing systems fit models by supplementing the Euclidian distance with the city-block distance. Thereby, it is possible to test whether the control forms are complementing each other or whether the performance effect is simply an additive effect. This statistical approach could be applied as a test technique in other studies of systems fit models.