Doctoral Dissertation Research in Political Science: The Delegation of Bureaucratic Policymaking Authority in Latin America

拉丁美洲的政治家是如何在官僚机构臃肿低效的情况下取得成功的政策成果的？为什么有些政策，如巴西的有条件现金转移支付计划Bolsa Familia，取得了成功，而在同一地区的类似计划却步履蹒跚？政治学严谨地研究了法律的制定方式，但很少关注法律的实施，尤其是在发展中国家。这是学术界内外的一个突出话题。随着最近民主化的国家克服政治和经济不稳定，决策者和公民都越来越关注公共政策的质量。然而，如果不首先了解政策的预期执行方式与实际执行方式之间的差距，就不可能构建有效的政策。本项目试图通过使用一种新的理论、关于总统代表团和行政能力的原始数据以及对拉丁美洲各国政策成功的深入研究来解决这一空白。具体而言，该项目考察了18个拉丁美洲民主国家的政策授权决策和政策成功情况，以了解政府如何增加其所青睐政策的成功几率。报告认为，在能力低下的环境中，政治家经常在政治上重要的立法上使用“官僚规避”，绕过现有的政府机构，转而建立新的机构，将政策委托给军方，或外包。这些规避代理具有更高的能力或更容易管理，从而增加了政策成功实施的机会。该项目可分为三个部分：1)开发一种适用于广泛国家的官僚规避的正式模式；2)利用拉丁美洲50 000多条行政命令的原始数据库对规避的决定因素进行实证检验；3)对巴西、委内瑞拉和厄瓜多尔的12项政策进行对照比较，检验授权策略对政策成功的影响。研究结果应该为政治家如何改善跨政治体系的公共服务提供证据，即使面对表现不佳的官僚机构。除了本研究的理论和分析贡献之外，还有两个实际因素突出：1)法令数据库的规模和细节，以及2)四位本科生助理获得的宝贵研究经验。该数据库将包括超过50,000条拉丁美洲行政法令，包括每项政策的无数特征，将成为比较政治、公共管理和公共政策的其他学者的工具性资源。这些数据将在项目完成后通过Dataverse Network project公开提供。此外，该项目还为四名本科生研究助理提供了收集数据、阅读代码本并对这些数据进行编码、然后将汇总数据系统化和解释的机会。总而言之，本项目旨在为理解比较行政和官僚政治做出理论和实证贡献，同时为本科生提供学习机会，并为其他学者提供有价值的数据。

How do politicians in Latin America achieve successful policy outcomes with oftentimes bloated and inefficient bureaucracies? Why do some policies, like Brazil's conditional cash-transfer program Bolsa Familia, achieve success while similar programs in the same places falter? Political Science has rigorously studied the ways in which laws are made, but has focused far less on how they are carried out, especially in the developing world. This is a salient topic inside and outside of academia. As recently democratized countries overcome political and economic instability, policymakers and citizens alike are increasingly concerned with the quality of public policy. Constructing effective policy is impossible, however, without first understanding the gap between how policy is supposedly implemented, and how it is actually implemented. This project seeks to address this lacuna through use of a novel theory, original data on presidential delegation and administrative capacity, and an in-depth examination of policy success in various Latin American countries.Specifically, the project examines policy delegation decisions and policy success across 18 Latin American democracies to understand how governments increase the odds of success for their favored policies. It argues that politicians in low capacity environments often employ "bureaucratic circumvention" for politically important legislation, bypassing existing government agencies in favor of creating new agencies, delegating policies to the military, or outsourcing. These circumventing agents possess higher capacity or can be more easily managed, and thus increase the chances of successful policy implementation. The project can be divided into three parts: 1) development of a formal model of bureaucratic circumvention applicable to a wide range of countries; 2) empirical tests of the determinants of circumvention with an original database of over 50,000 executive decrees from Latin America; and 3) controlled comparisons of twelve policies in Brazil, Venezuela, and Ecuador to test the effect of delegation strategy on policy success.The results should offer evidence as to how politicians can improve the provision of public services across political systems, even when faced with poorly performing bureaucracies. Beyond the theoretical and analytical contributions of this research, two practical elements stand out: 1) the size and detail of the decree database, and 2) the valuable research experience gained by four undergraduate assistants. The database--which will include over 50,000 Latin American executive decrees, including myriad characteristics of each policy--will be an instrumental resource for other scholars of comparative politics, public administration, and public policy. These data will be publicly available upon completion of the project via the Dataverse Network Project. Additionally, the project offers four undergraduate research assistants the opportunity to gain experience in collecting data, reading a codebook and coding those data, and then systematizing and interpreting the aggregated data. In sum, this project seeks to make theoretical and empirical contributions to the understanding of comparative executive and bureaucratic politics, while offering learning opportunities for undergraduate students and providing valuable data for other scholars.