5. Discussion and conclusion
The results of this study indicate that improved performance of a routine at time t0 leads to increased routine contraction at time t1. This implies that an organizational routine can shrink because of healthy happenings, such as, (1) selecting out organizational slack (Singh, 1986) from profitable routines and thereby achieving “more from less”; (2) using costly but fewer resources; (3) taking advantage of intelligent machinery or new technology that requires fewer resources or steps; (4) taking advantage of accumulated experiential learning or gains already made elsewhere in the firm, such as, using a common resource within the firm instead of buying a new resource; (5) making byproducts that are saleable; and (6) contracting the existing routine temporarily while searching for a better routine. The results also show that the efficiency of a routine negatively moderates the relationship between the performance of a routine at time t0 and its contraction at time t1. These empirical findings indicate that managers need not panic when organizational routines contract.