Doctoral Dissertation Research: The Effects of Identity and Social Networks on Investment and Market Participation Behavior: An Analysis from Rural South India

博士论文研究：身份和社交网络对投资和市场参与行为的影响：来自南印度农村的分析。本论文定量分析了身份和社会网络对教育投资、农业技术采纳和市场参与决策的影响，并沿着沿着区分了这些影响（社会学习，歧视，社会规范，模仿，价格规模的影响和对代理人的资源限制的影响）使用的数据来自六个村庄在印度的马哈拉施特拉邦和安得拉邦。为这项研究选定的那些村庄以前在1975年至1985年期间曾由国际热带作物研究所在其村庄一级研究中进行过研究，因此可以获得关于家庭和村庄一级特点的详细背景资料和数据。 因此，本论文的研究解决了发展经济学文献中的一个重要空白。最近的经济学研究承认（非生产性）身份和社交网络在经济决策中的重要性。社交网络和身份通过对偏好、限制和期望的影响，介导从资源到成果的映射，从而影响经济决策，如对儿童教育的投资、技术采用决策和市场参与决策。然而，现有的研究都没有区分这些影响沿着以下途径：从一个人的接触中学习（通常称为社会学习），歧视（即，价格或获得某些商品的机会对于不同的基于身份的群体的成员是不同的），价格规模效应（即，商品的价值/成本取决于用户的数量），社会规范（即，通过制裁在社会上强制执行的某个地方的规则），模仿（即，复制人们观察到的某些人的行为），以及放宽劳动力、资本和信贷约束（例如，朋友和家人在收获时在土地上伸出援助之手，借给机器或提供贷款）。从政策的角度来看，区分这些影响沿着这些不同的途径是重要的，因为刺激教育投资、增加和更快地吸收有前途的技术以及市场参与的正确战略从根本上取决于家庭一级的决策过程结构。 具体而言，本研究的目标有三个方面：第一，通过采用一套较新的数据收集技术，促进发展经济学领域的方法发展，例如：（1）样本内随机技术，将属于随机样本的个人与同一样本内的其他个人随机配对，并询问过去和现在的联系，他们愿意建立一个（假设的）与他/她的联系以及他们对他/她的能力和社会文化属性的看法; 2）收集关于主观期望的信息; 3）通过进行小型实验调动信息的能力;第二，发展一种识别策略，分离出数据中不同的社会网络和身份效应;第三，公开数据和数据收集方法，从而确保本研究的政策影响超出印度案例。

Doctoral Dissertation Research: The Effects of Identity and Social Networks on Investment and Market Participation Behavior: An Analysis from Rural South IndiaProposal No. 0649330 PI: Christopher B. Barrett CO-PI Annemie Maertens This field dissertation research analyzes quantitatively the effects of identity and social network on educational investment, agricultural technology adoption and market participation decisions and differentiates these effects along pathways (social learning, discrimination, social norms, imitation, price-scale effects and effects on the agent's resource constraints) using data from six villages in the Indian states of Maharashtra and Andhra Pradesh. The villages selected for this study have been studied before by ICRISAT in its Village Level Studies (VLS) during the period 1975-1985, and as such detailed background information and data on both household and village level characteristics are available. This dissertation research thereby addresses an important gap in the development economics literature. Recent studies in economics acknowledge the importance of (non-productive) identity and social networks in economic decision-making. Social networks and identity mediate the mapping from resources into outcomes through their effects on preferences, constraints and expectations, thereby influencing economic decisions such as investment in children's education, technology adoption decisions and market participation decisions. However, none of the existing studies have differentiated these effects along the following pathways: learning from one's contacts (often referred to as social learning), discrimination (i.e., prices or access to certain goods is different for members of different identity-based groups), price-scale effects (i.e., the value/cost of a good depends on the number of users), social norms (i.e., a rule in a certain locality that is socially enforced through sanctions), imitation (i.e., copying the actions of certain people one observes), and easing of labor, capital and credit constraints (e.g., friends and family give a helping hand on the land during harvest, lend machinery or provide loans). Differentiating these effects along these different pathways is important from a policy perspective as the right strategy to stimulate educational investment, increased and quicker uptake of promising technologies, and market participation depends fundamentally on the structure of the decision process at the household level. Concretely, the goal of this research is threefold: first, to contribute to the methodological development of the field of development economics by introducing a set of relatively new data-collection techniques, such as 1) the within-sample random technique where individuals who are part of a random sample are randomly matched with other individuals within the same sample and asked about past and present links, their willingness to establish a (hypothetical) link with him/her and their perception of his/her ability and social and cultural attributes and 2) the collection of information on subjective expectations and 3) the capacity to mobilize information by conducting small experiments; second, to develop an identification strategy to separate out the different social network and identity effects in the data; and third, to make the data and the data collection methodology publicly available, thereby ensuring that the policy implications of this study extend beyond the Indian case.