COLLABORATIVE RESEARCH: PRICES, PEERS, AND PERCEPTIONS: FIELD EXPERIMENTS ON TECHNOLOGY ADOPTION IN THE CONTEXT OF IMPROVED COOKSTOVES

在许多情况下，对可能改善福利的技术的采用仍然低得令人沮丧。改进的炉灶是一个主要的例子：虽然清洁燃烧的炉灶技术具有潜在的健康、环境和社会效益，但推广这些技术的努力还不够，在许多发展中国家，在明火上使用生物质烹饪的做法仍然占主导地位。本提案的中心目标是研究经济激励（“价格”）、社会学习（“同伴”）和主观信念（“感知”）如何相互作用，影响技术采用动态。为此，我们进行了一项实地试验，向有或没有社会关系的家庭和已经获得炉灶的家庭提供不同价格水平的新炉灶。我们的研究结果将为今后向更大的人口群体传播提高福利的技术提供信息。我们的家庭技术采用和使用决策的概念模型强调了价格、同伴和观念之间的多种潜在相互作用。通过我们的实验将解决的关键研究问题包括价格如何影响新技术的感知质量，这些感知如何通过接触具有该技术经验的同行而改变，以及感知如何随着时间的推移根据自己的经验和（客观和主观）技术性能而变化。我们实施了一种新的识别策略来识别这些影响，使用预先存在的和外源控制的免费炉子分布，结合不相关的集群随机分配到不同的炉子补贴水平。通过明确地测量感知与实验中的其他结果变量（包括炉子使用的调查和物理指标以及对个人暴露于污染物的影响），研究人员将能够测试有关价格和同伴（先前采用在信念形成中相互作用）的特定模型，这是技术采用文献中的一个关键问题。

Adoption of potentially welfare-improving technologies remains frustratingly low in many contexts. Improved cookstoves are a prime example: while cleaner-burning stove technologies have potential health, environmental, and social benefits, efforts to disseminate these technologies have fallen short and the practice of cooking with biomass over open fires remains dominant throughout much of the developing world. The central aim of this proposal is to study how economic incentives ("prices"), social learning ("peers"), and subjective beliefs ("perceptions") interact to influence technology adoption dynamics. We do so through a field experiment that offers new stoves at different price levels to groups of households with and without social ties to households that have already received stoves. Our results will inform and future efforts to disseminate welfare-enhancing technologies to larger population groups.Our conceptual model of households' technology adoption and use decisions highlights multiple potential interactions among prices, peers, and perceptions. Key research questions that will be addressed through our experiments include how price affects perceived quality of a new technology, how these perceptions are modified by exposure to peers that have experience with the technology, and how perceptions change over time based on one's own experience and (objective and subjective) technology performance. We implement a novel identification strategy for identifying these effects, using the preexisting and exogenously controlled distribution of free stoves in combination with uncorrelated, cluster-randomized assignment to different stove subsidy levels. By explicitly measuring perceptions in conjunction with other outcome variables in the experiment (including both surveys and physical indicators of stove use and impacts on personal exposure to pollutants), the researchers will be able to test specific models about how prices and peers(prior adoption interact in belief formation) a key issue in the technology adoption literature.