The recent interest in learning from failure has led to repeated praise of failure as part of the innovation process. The challenges in detecting and terminating failing projects are, however, often underplayed. This study suggests that firms exhibit different propensities to terminate failing innovation projects, and this heterogeneity in part accounts for long-run performance differences between firms in industries with high rates of innovation. An empirical examination of investments in the venture capital (VC) industry shows that VC firms are heterogeneous with respect to their propensities to terminate, and provides evidence that firms with higher termination capabilities have higher performance. In contrast, management of successful projects does not explain performance differences. The findings suggest that, in innovation-intensive contexts where success is rare, the management of failing investments may be at least as important as picking and managing successes, if not more so. The study also investigates the organizational antecedents of a capability to terminate.

Note: Research papers posted on SSRN, including any findings, may differ from the final version chosen for publication in academic journals.