Doctoral Dissertation Research in Political Science: Political Institutions and Economic Organization

本项目利用公司和行业层面的数据，衡量智利和墨西哥政治民主化对私营部门经济组织的影响。经济组织被广泛地定义为包括企业内部的权力组织（公司治理）和企业之间的竞争程度。经济组织之所以重要，是因为它影响到一个国家的整体竞争力和经济发展前景。建立政治制度和经济发展之间长期关系的论点是众所周知的（Engerman和Sokoloff, 1997; Acemoglu， Johnson和Robinson， 2001），最近的学术研究使用复杂的面板数据技术将民主化与增长联系起来（Persson和Tabellini, 2006； Papaioannou和Siourounis， 2007）。然而，这两篇文献都没有探索因果链中的中间环节。这个项目试图通过研究政治制度对经济组织的影响来填补文献中的空白。该项目的智力价值源于其独特的研究设计和理论框架。它扩展了当前的研究，通过使用一种被称为具有切换复制的中断时间序列的比较研究设计来研究国家内部政治制度的变化如何影响公司治理和市场结构（Shadish, Cook, and Campbell, 2001）。该设计利用了两国之间交替的民主化时期：墨西哥公司在智利民主化期间充当“控制”组，随后在20世纪90年代末智利公司成为“控制”组时代表“被处理”组。由于几个原因，这种设计在近似民主化对经济组织的影响方面特别强。最重要的是，这项研究不是研究国家层面的因变量，如经济增长，而是采用企业和行业层面的因变量，这使研究人员能够解释与最初的部门或技术异质性相关的其他解释。它还允许探索政治改革在不同行业或部门的潜在差异影响（如Aghion， Alesina和Trebbi， 2007）。此外，这两个国家都经过精心挑选，遵循最相似的案例设计，试图保持尽可能多的替代解释不变。对于本项目最重要的是，智利和墨西哥几乎同时发生了可能影响经济组织的经济改革。这项研究整合了经济组织的两个组成部分，这两个组成部分通常是单独研究的。关于公司治理，我的研究将收集和分析公司层面的股权集中度和股票市值的度量。为了衡量政治制度与经济竞争之间的关系，本研究将收集并分析集中度、赫芬达尔-赫希曼指数和价格成本比。在一个统一的框架下研究公司治理和市场结构是有意义的，因为这两个组成部分都受到公司法和国家政府制定的其他制度的约束，这使得它们在政治改革后受到共同的压力。这项研究的广泛影响包括对两个重要辩论的贡献。首先，它通过将注意力吸引到政治上，提高了我们对新兴市场私营部门发展决定因素的认识。普遍的观点是，“制度对经济发展很重要”，但对于哪些具体制度特别重要，以及将它们与经济增长联系起来的渠道，人们知之甚少。其次，这个项目有助于更广泛的关于民主的辩论（例如，参见Acemoglu和Robinson， 2006）。例如，民主改革后企业组织变化的证据将为民主削弱特殊利益集团的经济权力这一命题提供强有力的支持。相比之下，一个无效的结果将表明，法律上的政治制度，如民主制度，对于经济变革是不够的，因为根深蒂固的精英实际上拥有巨大的政治权力。

This project measures the effect of political democratization on private sector economic organization in Chile and Mexico using firm- and industry-level data. Economic organization is defined broadly to include the organization of power within firms (corporate governance), and the degree of competition between firms. Economic organization matters because it influences a country's overall competitiveness and its prospects for economic development.The arguments establishing a long-run relationship between political institutions and economic development are well-known (Engerman and Sokoloff, 1997; Acemoglu, Johnson, and Robinson, 2001), and recent scholarship ties democratization to growth using sophisticated panel data techniques (Persson and Tabellini, 2006; Papaioannou and Siourounis, 2007). However, both of these literatures leave intermediate links in the causal chain unexplored. This project seeks to fill a gap in the literature by studying the effects of political institutions on economic organization.The intellectual merit of the project derives from its unique research design and theoretical framework. It extends the current research by studying how changes in political institutions within countries affect corporate governance and market structure using a comparative research design known as an interrupted time series with switching replications (Shadish, Cook, and Campbell, 2001). The design exploits the alternating periods of democratization among the two countries: Mexican firms serve as the "control" group during Chile's democratization, and subsequently represent the "treated" group during the late 1990's, when Chilean firms become the "control" group.This design is particularly strong at approximating the effect of democratization on economic organization for several reasons. Most importantly, rather than studying a country-level dependent variable such as economic growth, this research employs firm and industry-level dependent variables, which enable the researcher to account for alternative explanations related to initial sectoral or technological heterogeneity. It also allows for an exploration of a potentially differential impact of political reform across different industries or sectors (as in Aghion, Alesina, and Trebbi, 2007). Furthermore, the two countries are carefully chosen to follow a most-similar case design, which attempts to hold constant as many alternative explanations as possible. Most importantly for this project, economic reforms likely to affect economic organization occurred in Chile and Mexico at nearly the same time.This study integrates two components of economic organization that are often studied in isolation. Regarding corporate governance, my study will collect and analyze firm-level measures of ownership concentration and stock market capitalization. To measure the relationship between political institutions and economic competition, the study will collect and analyze concentration ratios, Herfindahl-Hirschman indices, and price-cost ratios. It makes sense to study corporate governance and market structure under a unified framework, since both components are governed by corporate laws and other institutions produced by national governments, which makes them subject to common pressures following political reform.The broad impact of the study includes a contribution to two important debates. First, it improves our knowledge of the determinants of private sector development in emerging markets by drawing attention to politics. A consensus view has that "institutions matter" for economic development in general, but little is known about which specific institutions are particularly important, and channels linking them to growth. Second, this project contributes to broader debates about democracy (see, for example, Acemoglu and Robinson, 2006). For instance, evidence of changes in corporate organization following periods of democratic reform would provide strong support for the proposition that democracy reduces the economic power of special interest groups. By contrast, a null result would suggest that de jure political institutions such as democracy are insufficient for economic change due to the substantial de facto political power of entrenched elites.