Doctoral Dissertation Research in Political Science: Social Networks and Electoral Strategies in the Philippines

在新兴民主国家，如何鼓励政治家选择支持增长或发展的政策？越来越多的人认为，政治因素——尤其是善治和问责制——是经济成功的关键决定因素，进而也是发展援助有效性的关键决定因素。然而，问题在于，促进增长的政策，包括提供必要的公共产品，并不总是对寻求连任的政治家有吸引力（Bueno de Mesquita， Smith等人，2003；Vicente和Wantchekon 2008）。本文发展了一种新兴民主国家政治的新理论，其中政治家通过事先购买选票的策略赢得地方政府职位，其中政治家向个人提供礼物或金钱以换取他们的选票（Schaffer和Schedler 2007），以及事后裙带主义策略，通过政治家提供有针对性的私人物品以获得选举支持（Kitschelt和Wilkinson 2007； Stokes 2007）。社交网络的密度是选择这两种策略的重要决定因素。密切的社会联系和反复的互动有助于监督和核实政治家与选民个人之间的政治交流。该理论认为，社会网络密集的村庄将采用事后策略，而缺乏这种网络结构的村庄将采用事前策略。同样，研究人员假设，个人在社会网络中的地位会影响选民是事前还是事后的政治交换目标：与关系良好的人相比，关系较差的人更容易成为贿选的目标，而不太容易成为裙带关系交换的目标。本论文将在菲律宾第二大省份伊莎贝拉省进行一系列家庭和政治家调查，根据世界银行最近进行的政治和地方服务提供基线调查收集的数据。该调查采用了两种方法创新：1)收集社会网络数据来评估网络结构，并确定个人在网络中的位置；2)采用新的调查方法，允许被调查者不让枚举员看到他们的回答（无与伦比的点票技术），这有望减少对贿选的漏报。利用社会网络分析和新的调查方法，可以评估社会网络在确定与不同政治战略有关的监测和核查费用方面的作用。该项目是第一个研究偶然政治交换事前与事后机制的决定因素的项目。了解这些机制可以帮助决策者和当地非政府组织在菲律宾设计更有效的选民教育计划，并更好地满足特别容易受到贿选和其他非法行为影响的群体的需求。例如，菲律宾目前许多反贿选活动都是通过教会和公民团体组织起来的。然而，如果与社区社会联系较少的人比那些融入社会程度较高的人更容易成为贿选的目标，这些项目将无法触及最弱势群体。此外，通过现有社会群体实施的项目甚至可能产生意想不到的后果，即进一步扩大人脉广泛和人脉较少的个人之间的选民意识差距。本研究还将有助于提高发展援助和经济政策改革的有效性（如Burnside和Dollar 2000）。援助文献中的一个谜题是，为什么许多国家即使采用了自由选举和任期限制等制度，也无法促进政治问责制。这项研究可以有助于世界银行和其他援助组织设计经济改革干预措施，查明破坏政治问责制和阻碍有利于增长的经济政策成为可行的选举战略的政治和社会结构。

How can politicians be encouraged to choose pro-growth or development-oriented policies in new democracies? There is a growing consensus that political factors--in particular, good governance and accountability--are key determinants of economic success and, in turn, development aid effectiveness. The problem, however, is that growth-promoting policies, including providing necessary public goods, are not always attractive to politicians seeking re-election (Bueno de Mesquita, Smith et al. 2003; Vicente and Wantchekon 2008). This dissertation develops a new theory of politics in emerging democracies in which politicians win office at the local government level through ex ante strategies of vote buying, in which politicians give gifts or money to individuals in exchange for their vote (Schaffer and Schedler 2007), and ex post strategies of clientelism, through which politicians offer targeted private goods for electoral support (Kitschelt and Wilkinson 2007; Stokes 2007). The density of social networks is an important determinant of the choice between these two strategies. Close social ties and repeated interactions facilitate monitoring and verifying political exchange between politicians and individual voters. The theory implies that villages with dense social networks will be characterized by ex post strategies while villages lacking these network structures will use ex ante strategies. Similarly, the researcher hypothesizes that individual position in the social network affects whether voters are targeted for ex ante or ex post political exchange: the less connected are more targeted for vote-buying and less targeted for clientelistic exchange than the well-connected.This dissertation will undertake a set of household and politician surveys in Isabela Province, the second largest province in the Philippines, building on data collected from a baseline survey of politics and local service delivery recently conducted by the World Bank. This proposed survey employs two methodological innovations: 1) collecting social network data to evaluate network structure and to identify the position of individuals within the network; and 2) employing new survey methods that allow respondents to shield their response from enumerators (unmatched count technique) that are expected to reduce the under-reporting of vote buying. The use of social network analysis and of new survey methods allows for an assessment of the role of social networks in determining the monitoring and verification costs associated with the different political strategies.This project is the first to study the determinants of ex ante vs. ex post mechanisms of contingent political exchange. Understanding these mechanisms can help policymakers and local NGOs to design more effective voter education initiatives in the Philippines and better address the needs of groups that are particularly vulnerable to vote buying and other illegal practices. For example, many current anti-vote buying initiatives in the Philippines are organized through churches and civic groups. However, if people with fewer social ties to the community are targeted for vote buying at higher rates than those that are more integrated, such programs will not be able to reach the most vulnerable groups. Additionally, programs implemented through existing social groups may even have the unintended consequence of further widening the gap in voter awareness between well-connected and less-connected individuals. This study will also help improve the effectiveness of development aid and economic policy reform (e.g., Burnside and Dollar 2000). One puzzle in the aid literature is why many countries are unable to foster political accountability, even after adopting institutions such as free elections and term limits. This study can contribute to the design of economic reform interventions by the World Bank and other aid organizations by identifying political and social structures that undermine political accountability and prevent growth-friendly economic policies from being viable electoral strategies.