Laboratory federalism hypothesizes that decentralization fosters innovation and learning in multijurisdictional systems, conducing to good policies. We discuss this using evolutionary stability, focusing in this paper on public funds sharing. In a model of rich-to-poor redistribution with labor mobility, where policy choices based on relative success lead to drastic welfare erosion, suitably designed sharing schemes can restore efficiency. Their necessary properties are the same as in a Nash equilibrium, showing they are a powerful corrective device for different modes of government interaction. Two drawbacks appear: reduced accuracy due to large multiplicity and a tendency to overly reward inefficiently high redistribution.

中文翻译：

---

实验室联邦制与公共资金共享

实验室联邦主义假设权力下放促进了多辖区系统的创新和学习，有助于制定良好的政策。我们使用进化稳定性来讨论这个问题，在本文中重点讨论公共资金共享。在劳动力流动的贫富再分配模型中，基于相对成功的政策选择导致福利急剧下降，适当设计的共享计划可以恢复效率。它们的必要属性与纳什均衡中的相同，表明它们是不同政府互动模式的强大矫正工具。出现了两个缺点：由于大量的多样性而降低了准确性，以及倾向于过度奖励低效的高再分配。