Preferences and Poverty Traps: Experimental Investigations of Risk, Time and Social Preferences in Two Poor Neighborhoods

实验经济学正在走出实验室。实地实验允许研究人员检查自然环境中的行为，并调查新的和以前未开发的受试者的偏好。这项研究有可能对与政策相关的问题产生重要的见解。该项目通过收集两个贫穷少数民族社区居民的风险、时间和社会偏好，为这一努力做出了贡献。这些社区在收入和许多其他人口结构上相似，除了种族：南达拉斯主要是非裔美国人社区，而东达拉斯主要是西语裔社区。我们使用实验和调查来衡量偏好和行为。决策表格是为低识字率人群设计的，游戏以图画形式呈现，文字最少。这项调查被用来收集关于人口统计、社交网络、身份、社会偏好、住房拥有率、对非营利组织的捐赠以及志愿者工作的信息。调查人员使用收集的数据来调查一些问题。首先，他们将这些人群的风险、时间和社会偏好与我们通常在实验室和现场非贫困人群中看到的偏好进行了比较。他们预计，这种比较将说明贫困陷阱的一些原因和后果。其次，他们比较了两个群体之间的风险、时间和社会偏好，以便更好地了解在人口中最贫穷的部分表现出的文化差异，以及两个社区之间在经济表现和发展速度方面的差异。第三，他们考察了风险、时间和社会偏好对调查中收集的自我报告和实际行为的相关性和因果关系(通过中介和适度分析)，包括住房拥有率、储蓄决定、教育决定、慈善捐赠和其他行为。这项研究的最后一个贡献是更深入地了解特定样本的偏好，这将被证明对未来几十年美国经济的增长非常重要：低收入的拉美裔美国人。这项研究将我们关于风险、时间和社会偏好的个体差异的知识扩展到一个新的研究对象池：低收入少数群体。将这些偏好与通常抽样的中产阶级人口的偏好进行比较，并考察它们对现实世界经济决策的影响，将有助于深入了解贫困陷阱的原因和后果。比较这两个受试者之间的这些偏好将提供对文化差异的洞察，并提供关于什么使个人能够摆脱贫困陷阱和发展社区的新见解。更广泛的影响：首先，这些结果为贫困社区的内部运作提供了一个新的、亟需的描述性描述，对改善低收入城市社区和帮助个人克服贫困陷阱具有影响。考察贫穷拉美裔人口的偏好本身，将增进我们对最有可能在未来几十年为美国经济增长和竞争力做出贡献的个人的理解。其次，这一努力允许进一步改进对偏好的实验测量，这些测量经过微调，以最有效地引出贫穷、低识字率人群的偏好。这些措施对更广泛的经济学和社会科学领域的其他研究人员很有用。第三，这项研究涉及几名本科生研究助理的参与，他们将获得宝贵的实践研究经验和指导。

Experimental economics is moving out of the lab. Field experiments allow researchers to examine behavior in a natural setting, and to investigate the preferences of new and previously-untapped subject pools. This research has the potential to generate important insights into policy-relevant questions. This project contributes to this endeavor by collecting risk, time and social preferences from residents of two poor, minority neighborhoods. These neighborhoods are similar in income and many other demographics, with the exception of race: South Dallas is primarily an African-American community, while East Dallas is primarily a Hispanic community. We use experiments and surveys to measure preferences and behavior. The decision forms are designed for a low-literacy population, with the games presented in pictorial form with minimal text. The survey is used to collect information on demographics, social networks, identity, social preferences, home ownership, and donations to nonprofits as well as volunteer work. The investigators use the data collected to investigate a number of questions. First, they compare the risk, time and social preferences elicited from this population with those we typically see in the lab and among non-poor populations in the field. They anticipate that this comparison will illustrate some of the causes and consequences of poverty traps. Second, they compare the risk, time and social preferences between the two populations in order to better understand cultural differences that show up in the poorest segment of the population, as well as differences in economic performance and development rates between the two communities. Third, they examine the correlations and causality (through mediation and moderation analysis) of risk, time and social preferences on self-reported and actual behavior as collected in the survey, including home ownership rates, savings decisions, educational decisions, charitable giving and other behaviors. A final contribution of the research is a deeper understanding of preferences of a particular sample that will prove important to the growth of the US economy over the next several decades: low-income Hispanics. This research extends our knowledge about individual variation in risk, time, and social preferences to a new subject pool; low-income minorities. Comparing these preferences with those of the middle-class populations typically sampled, and examining their impact on real-world economic decisions will provide insights into the causes and consequences of poverty traps. Comparing these preferences between these two subject pools will provide insights into cultural differences and provide new insights into what enables individuals to escape poverty traps and neighborhoods to develop. Broader Impacts: First, these results provide a new and much-needed descriptive account of the inner workings of poor neighborhoods, with implications for improving low-income urban neighborhoods and helping individuals to overcome poverty traps. Examining the preferences of the poor Hispanic population per se will increase our understanding of individuals most likely to be contributing to US growth and competitiveness in the coming decades. Second, this effort allow further refinement in experimental measures of preferences that are fine-tuned to be most effective in eliciting preferences of poor, low-literacy populations. These measures are useful to other researchers in the broader economics and social science community. Third, the research involves the participation of several undergraduate research assistants who will receive valuable, hands-on research experience and mentoring.