Leadership and Overcoming Coordination Failure

该奖项资助的研究旨在检验有关领导力如何减少协调失败的经济学理论。协调失败是影响许多组织的常见问题。例如，一个人的薄弱努力可能会拖累整个团队的表现。装配线的速度不会比生产线上最慢的人快，还是一份报告会快？直到最后一个人完成他们的部分，我才能完成。有潜在“薄弱环节”的组织容易陷入生产力陷阱，所有个人都付出很少的努力。领导力是管理和促进向更好结果过渡过程的天然工具。pi及其合作者对协调与领导之间的关系有着长期的兴趣。先前的研究表明，有能力的领导者可以增加成功协调的几率，但强加一个领导者并不是万灵药。成功的几率取决于领导者是谁，他们是如何被选择的，他们可用的工具，以及他们如何行使他们的领导力。我们提出了三个项目，这些项目着眼于协调和领导很重要的新环境，并研究了以前未经测试的机制，以使领导者更有效。(1)集中式管理与分散式管理：考虑一个有两个部门的公司。这些部门可以通过协调他们的活动来获利，但每个部门都有关于世界状况的私人信息，而且，关键的是，这两个部门的利益并不完全一致。是让一个中央管理者从各个部门收集信息，并将选择强加给各个部门，还是让各个部门共享信息并单独做出决策？集中管理在很大程度上消除了各部门面临的协调问题，但却激励各部门歪曲他们向中央经理提交的报告，从而导致次优结果。Alonso， Dessein和Matouschek（2008）表明，在广义条件下，权力下放使企业利润最大化，但这一结果依赖于可能不现实的行为假设。该奖项资助了简化Alonso等人的环境的实验，通过消除部门之间的信息不对称来对抗集中化。pi推测，即使在一个反对集权的环境中，与理论相反，使用集中管理的公司也会比采用分散管理方法的公司表现更好。(2)领导力和建议：想象一群投资者独立地选择是否把钱投入一个投资机会。投资的回报取决于投资者的数量和未知的世界状态。投资者从一位了解世界现状的顾问那里获得建议，但在不符合投资者最佳利益的情况下，他有动机提出投资建议。反复的互动迫使顾问在欺骗投资者的短期收益与维护信誉的长期关切之间进行协调。pi通过实验来研究顾问是否能够成功地利用他们的优势信息在有效均衡上产生协调。他们计划将这个实验的结果与经济政策制定者所面临的一些问题联系起来，因为他们试图使一个国家摆脱像目前影响美国的经济衰退。(3)学习领导：领导者可以帮助团队克服协调失败。尽管如此，许多有领导的团体未能就有效的结果进行协调。这不仅是因为运气不好，还因为领导者犯了系统性错误。因为特定的错误是可以被识别出来的，所以如果坏的领导者能够学习（或被教导）避免这些错误，他们就可以变成好的领导者，这似乎是合理的。第三个项目使用经济实验来检验提高领导者绩效的可能方法：直接经验、间接经验和几种不同类型的建议。智力价值：了解领导者如何在各种环境中发挥作用，对于提高我们对组织如何实现有效协调的认识至关重要。除了扩大我们对领导力的了解，这里提出的研究旨在改善机构设计和领导者培训，以减少生产力陷阱持续存在的可能性。预期的更广泛影响：协调问题在许多经济失败中起着重要作用。简而言之，经济可以是一个自我实现的预言。如果人们认为经济将表现不佳，它就会表现不佳。这里提出的工作旨在帮助领导者制定制度和战略，使其更容易摆脱生产力陷阱。根据提议的研究开发的软件和技术应该在教授各种各样的经济学和商业课程中有用。研究助理将接受宝贵的培训。

This award funds research that tests economic theories about how leadership can work to reduce coordination failures. Coordination failure is a common problem affecting many organizations. For example, a single individual's weak effort can drag down the performance of an entire group. An assembly line moves no faster than the slowest person in the line or a report doesn?t get finished until the last person completes their section. Organizations with a potential 'weak link' are prone to being caught in productivity traps where all individuals exert low effort. Leadership is a natural tool for managing and facilitating the process of transitioning to a better outcome. The PIs and their collaborators have a long standing interest in the relationship between coordination and leadership. Previous research demonstrates that able leaders can increase the odds of successful coordination, but imposing a leader is not a cure-all. The odds of success depend on who the leader is, how they are selected, the tools they have available, and how they exercise their leadership. We propose three projects which look at new environments in which coordination and leadership are important and examine previously untested mechanisms for making leaders more effective. (1) Centralized vs. Decentralized Management: Consider a firm with two divisions. The divisions can gain by coordinating their activities but each division has private information about the state of the world and, critically, the interests of the two divisions are not perfectly aligned. Is it better to have a central manager collect information from the divisions and impose choices on the divisions or to have the divisions share information and make decisions separately? Centralized management largely eliminates the coordination problem facing the divisions, but gives the divisions incentives to distort their reports to the central manager, leading to suboptimal outcomes. Alonso, Dessein, and Matouschek (2008) show that decentralization maximizes firm profits under broad conditions, but this result relies on behavioral assumptions that may be unrealistic. This award funds experiments that simplify the environment of Alonso et al, stacking the deck against centralization by eliminating information asymmetries between the divisions. The PIs conjecture that even in an environment biased against centralization, and contrary to theory, companies that use centralized management will outperform firms with a decentralized approach. (2) Leadership and Advice: Imagine a group of investors independently choosing whether to put their money into an investment opportunity. The return of the investment depends on the number of investors and an unknown state of the world. The investors receive advice from an adviser who knows the state of the world, but has incentives to suggest investment when it is not in the investors' best interest. Repeated interaction forces the adviser to reconcile short term gains from deceiving investors with a long-term concern for maintaining credibility. The PIs conduct experiments to study whether advisers can successfully use their superior information to generate coordination on an efficient equilibrium. They plan to relate the results of this experiment with some of the problems facing economic policy makers as they try to lift a country out of a recession like the one currently affecting the U.S.(3) Learning to Lead: Leaders can help groups overcome coordination failure. Nonetheless many groups with leaders fail to coordinate on an efficient outcome. This occurs not just because of bad luck, but also because leaders make systematic errors. Because specific errors can be identified, it seems plausible that bad leaders can be transformed into good ones if they can learn (or can be taught) to avoid these errors. The third project uses economic experiments to examine possible ways of improving leaders' performance: direct experience, indirect experience, and several different varieties of advice.Intellectual Merit: Understanding how leaders function in a variety of settings is central to improving our knowledge of how organizations can achieve efficient coordination. Beyond an expansion our knowledge of leadership, the research presented here is intend to improve the design of institutions and the training of leaders to make it less likely that a productivity trap will persist. Expected Broader Impact: Coordination problems play an important role in many economic failures. Put simply, the economy can be a self-fulfilling prophecy. If people believe the economy is going to do poorly, it will. The work proposed here is intended to help develop institutions and strategies for leaders that will make it easier to escape from productivity traps. Software and techniques developed with the proposed research should be useful in teaching a wide variety of economics and business courses. Research assistants will receive valuable training.