Citizen Scrutiny and Government Efforts to Fight Corruption

政府腐败是世界许多地区经济和社会发展缓慢的部分原因。这项研究将使用两个项目来调查如何最有效地减少政府合同中的腐败。各国政府监督政府活动各方面的能力有限，因此它们依赖公民自愿提供信息来打击腐败。然而，公民有自己的利益，可能不具备与政府工作人员相同的培训和能力。fiRST项目调查如何最好地将监督公共工程项目的任务授权给公民，同时考虑到他们的动机和执行复杂审计相关任务的能力的差异。只有政府才能根据公民监督员提供的信息调查和制裁公共或ffi情报机构。然而，各国政府分析公民监督员提供的大量信息的能力有限。因此，第二个项目调查如何利用技术来处理大量收到的公民报告并对其进行分类。该项目使用机器学习和人工智能(AI)从公民举报的大量腐败报告和其他类型的腐败信息中检测腐败模式。这一研究项目的结果不仅将为减少政府腐败的政策提供投入，还将确立美国在反腐败方面的全球领先地位。这项研究使用了两个项目，对发展经济学文献做出了重大贡献。首先，它有助于通过开发机器学习和人工智能(AI)工具并将其与举报人报告相结合来预测腐败的存在，并研究如何将反腐败ff作为目标，从而有助于日益增长的关于腐败检测和衡量的实证文献。其次，该项目有助于研究贸易组织ffS在自上而下和自下而上对政府的监督之间的关系，该文献基于对一个独特的混合计划的研究-由依赖公民参与的中央政府实施；该项目还调查了如何最好地设计这样的政策。第三，这项研究有助于在公共政策中使用机器学习的文献，应用这些工具来加强反腐败ff工作。最后，本研究通过研究稀缺的组织资源对fi反腐败的配置，对国家组织的文献做出了贡献。这一研究项目的结果不仅将为减少政府腐败的政策提供投入，还将确立美国作为反腐败斗争的全球领导者的地位。这一奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Corruption in government partly accounts for the slow economic and social progress in many parts of the world. This research will use two projects to investigate how best to efficiently decrease corruption in government contracting. Governments have limited capacity to monitor all aspects of government activity, hence they rely on citizens volunteering information to fight corruption. However, citizens have their own interests and may not possess the same training and capabilities as government workers. The first project investigates how best to delegate the monitoring of public works projects to citizens, taking into consideration differences in their motivation and their ability to perform complex audit-related tasks. Only governments can investigate and sanction public officials based on information provided by citizen monitors. However, governments have limited capacity to analyze large volumes of information provided by citizen monitors. The second project therefore investigates how technology can be leveraged to process large volumes of incoming citizen reports and triage them. This project uses machine learning and artificial intelligence (AI) to detect patterns of corruption from the large number of reports and other types of information about corruption that citizens report. The results of this research project will not only provide inputs into policies to reduce corruption in government; it will also establish the US as the global leader in the fight against corruption.This research uses two projects to make major contributions to the development economics literature. First, it contributes to the growing empirical literature on the detection and measurement of corruption by developing machine learning and artificial intelligence (AI) tools and combining them with whistleblower reports to predict the presence of corruption, and study how anti-corruption efforts can be targeted. Second, the project contributes to the literature on tradeoffs between top-down and bottom-up monitoring of governments based on the study of a unique hybrid program—implemented by a central government that rely on citizen’s participation; it also investigate how best to design such a policy. Third, the study contributes to the literature on the use of machine learning in public policy, applying these tools to enhance anti-corruption efforts. Finally, the research contributes to the literature on the organization of the state by studying the allocation of scarce organizational resources to the fight corruption. The results of this research project will not only provide inputs into policies to reduce corruption in government; it will also establish the US as the global leader in the fight against corruption.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.