Party organisation and the success of regional parties at the sub-national level in India.

ESRC博士后奖学金的主要目的是巩固我的博士研究。拟议的研究是基于一个简单的研究问题：为什么区域性政党在一个国家的某些地区取得成功，而在其他地区却没有。对这一问题最常见的答案是基于政治和经济权力下放和/或基于区域的种族分裂的存在。现有研究表明，当地区拥有控制重要政治和经济问题的地区议会时，地区政治精英更有可能组建地区政党，地区选民更有可能投票给他们。同样，其他人认为，当地区具有强烈的地区认同感时，地区精英和选民更有可能支持地区政党。承认这些解释的重要性，我认为，迄今为止探索的解释所提出的研究问题是基于党的组织。我认为，当全国性政党给他们的地区分支机构自治，地方政党发现很难成功。这是因为区域分支能够吸收区域精英和选民中存在的任何潜在的区域主义。即使该地区在重要问题上拥有政治和经济权力，和/或具有强烈的区域特征，情况也是如此。为了证明政党组织的重要性，我使用了来自世界上最大的民主国家印度的经验证据和数据。从经验和方法上讲，印度也是一个很好的测试案例。首先，关于地区政党成功的研究在印度尤其重要，因为2014年有近2.4亿公民投票支持地区政党。通过我的研究，我的目的是解释为什么一些印度人投票给地区政党，而其他人不投票。从方法论上讲，印度是一个联邦国家，印度各邦的政治和经济权力下放程度越来越大。此外，印度的大多数邦都是在地区认同的基础上形成的。尽管有这些特点，一些印度邦由地区政党主导，而其他邦的政治竞争仅限于全国性政党。我认为，这种变化在印度联邦是因为在国家政党内的区域分支自治水平的变化。在全国性政党将自治权赋予其地区分支机构的地区，地区政党很难取得成功。与此相反，如果没有一个主要的全国性政党给予其地区分支机构自治权，则地区政党更有可能取得成功。没有数据可以系统地衡量印度全国性政党中地区分支的自治程度。为此，我开始对印度两个邦，即古吉拉特邦和马哈拉施特拉邦的两个主要民族政党--国大党和印度人民党（BJP）--的地区分支自治级别进行手工编码。然而，由于时间，资源和研究技能的限制，我无法在博士期间完成编码练习。奖学金将使我能够完成1989年后印度所有主要邦的编码工作。与我在读博士期间进行的手工编码练习不同，我希望使用R语言中的定量文本分析来编码区域分支自治的级别。自动编码工作将产生一个1989年后印度15个邦的两个全国性政党内的区域分支自治小组数据集。然后，数据集将被用作（1）主要因变量，以了解为什么一些国家政党在印度一些邦给予其地区分支机构更多的自主权，而不是其他人。(2)它还将被用作一个主要的独立变量，以确定国家政党在区域一级的高度区域分支自治是否与区域政党获得的支持较少有因果关系。

The primary aim of the ESRC post-doctoral fellowship is to consolidate my PhD research. The proposed research is based on a simple research question: why are regional parties successful in some parts of a country, but not others. The most common answer to this question is based on political and economic decentralisation and/or the presence of regionally-based ethnic cleavages. Existing research shows that when regions have regional parliaments with control over important political and economic issues, regional political elites are more likely to form regional parties and regional voters are more likely to vote for them. Similarly, others have argued that when regions have a strong regional identity, regional elites and voters are more likely to support regional parties. Acknowledging the importance of these explanations, I argue that an unhitherto explored explanation to the proposed research question is based on party organisation. I argue that when national parties give their regional branches autonomy, regional parties find it hard to succeed. This is because the regional branch is able to absorb any underlying regionalism present amongst the regional elites and voters. This is the case even if the region has political and economic authority over important issues, and/or a strong regional identity. To demonstrate that party organisation matters, I use empirical evidence and data from India, the world's largest democracy. India is also a good test case empirically and methodologically speaking. First, a study on regional party success is particularly relevant in India because close to 240 million citizens cast their votes for a regional party in 2014. Through my research, I aim to explain why some Indians vote for regional parties, where others do not. Methodologically speaking, India is a federal country with growing political and economic decentralisation for all the Indian states. Moreover, most Indian states are formed on the basis of a regional identity. Despite these characteristics, some Indian states are dominated by regional parties where others have political competition restricted to national parties. I argue that this variation across the Indian federation is because of the variation in the level of regional branch autonomy within the national parties. In regions where national parties have given autonomy to their regional branches, regional parties find it hard to succeed. In contrast, where none of the main national parties give autonomy to their regional branches, regional parties are more likely to succeed.There exists no data that systematically measures the level of regional branch autonomy within national parties in India. To this end, I began manually coding the level of regional branch autonomy within the two main national parties-the Congress Party and the Bharatiya Janata Party (BJP)-for two Indian states, namely Gujarat and Maharashtra. However, due to limitations of time, resources, and research skils, I was not able to complete the coding exercise during my PhD. The fellowship will allow me to complete the coding exercise for all the major Indian states in the post-1989 period. Unlike the manual coding exercise I undertook during my PhD, I wish to employ Quantitative Textual Analysis in R to code the level of regional branch autonomy. The automated coding exercise will result in a panel dataset of regional branch autonomy within the two national parties for the 15 Indian states in the post-1989 period. The dataset will then be used as a (1) primary dependent variable to understand why some national parties give more autonomy to their regional branches in some Indian states, but not others. (2) It will also be used a primary independent variable to see whether high regional branch autonomy for national parties at the regional level is causally related to less support for regional parties.