Selection at the Gate: Difficult Cases, Spillovers, and Organizational Learning
分析了英国生育诊所的纵向数据，发现接收更多疑难病例的诊所反而因更强的组织学习能力而长期绩效更优，挑战了“严格筛选输入能提升输出成功率”的直觉。
We analyze longitudinal data on British fertility clinics to examine the impact of “selection at the gate,” i.e., the attempts of organizations to improve the success rate of their output by selecting promising cases as input. In contrast to what might be expected, we argue that more stringent input selection is likely to lead to lower overt performance compared with those firms that admit difficult cases, because the latter develop steeper learning curves. That is, difficult cases enable greater learning from prior experience because they promote experimentation, communication among various actors, and the codification of new knowledge. Our results confirm this prediction and provide clear evidence that organizations with more difficult cases in their portfolios gradually begin to display performance figures that compare favorably with those of firms that do select at the gate.