We conceptualize downsizing as an attempt to reduce organizational slack. We suggest that the degree to which downsizing will improve firm performance will be contingent on conditions under which the downsizing occurs. We emphasize the level of organizational slack as an important contingency, and also examine two other contingencies: (1) whether the scope of the downsizing is narrow (restricted to personnel reductions) or broad (involves organizational redesign); and (2) if the downsizing is conducted proactively (when performance is stable or improving) or reactively (when performance is declining). By analyzing a panel dataset of downsizings conducted by the 100 largest American industrial firms from 1977 to 1993, we find broad support for our hypotheses that downsizings are more likely to lead to improved performance when firms have high slack, when their scope of the downsizing is broad, and when the downsizing is done proactively. We also explore and find evidence for interactions among these contingencies. We discuss the implications of our findings for the literatures on downsizing and organizational slack. In doing so, we bring together two literatures that have an obvious affinity but have been only loosely coupled in the past. Copyright © 2005 John Wiley & Sons, Ltd.