An Experimental Evaluation of a Matching Market Mechanism

在过去的三十年里，市场设计的见解对没有价格的匹配市场产生了巨大的影响。经济学家帮助重新设计了新医生到医院的机制，将学生分配到许多大学区的学校，以及将肾脏捐赠者与接受者进行最佳匹配的机制。尽管这些市场的市场设计取得了明显的成功，但关于这些改革对市场和匹配结果的影响的实证证据却非常少。该项目将提供迄今为止最可信的证据，证明在陆军军官与职位匹配的背景下，使用通用市场设计机制对匹配结果的影响。每年有一万多名军官申请转到军队内的新部队。该项目将衡量使用共同市场设计算法或现状机制对匹配结果（如官员留任和绩效）的影响。每年有一万多名军官申请转到军队内的新单位。目前，这种军官与部队的匹配是由美国陆军人力资源司令部（HRC）手动执行的。本项目采用随机对照试验来衡量用延迟接受算法取代现状分配过程对匹配结果的影响，如军官的留任和绩效。具体来说，由军官级别和专业化（例如工程船长）定义的脱节市场被随机分配，使用延迟接受算法或现状机制来确定匹配。这将提供第一个随机对照试验的证据，证明市场设计干预对匹配和市场结果的影响。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Over the last three decades, insights from market design have had a tremendous impact onmatching markets without prices. Economists have helped re-design the mechanisms formatching new doctors to hospitals, for assigning students to schools in a number of large schooldistricts, and for optimally matching kidney donors to recipients. Despite the apparent success ofmarket design in these markets, there is remarkably little empirical evidence of the impact ofthese reforms on market and match outcomes. This project will provide the most credibleevidence to date of the impact of matching using a common market design mechanism on matchoutcomes in the context of matching Army officers to positions. Each year, over ten thousandofficers apply for transfers to new units within the military. This project will measure the impactof using a common market design algorithm or the status quo mechanism on match outcomes,like officers’ retention and performance.Each year, over ten thousand officers apply for transfers to new units within the military.Currently, this officer to unit matching is implemented manually by the US Army’s HumanResources Command (HRC). This project uses a randomized controlled trial to measure theimpact of replacing the status quo assignment process with the deferred acceptance algorithm onmatch outcomes, like officers’ retention and performance. Specifically, disjoint markets, definedby officers’ rank and specialization (e.g. Captains in engineering) were randomly assigned todetermine matches using either the deferred acceptance algorithm or the status quo mechanism.This will provide the first evidence from a randomized controlled trial about the impact of amarket design intervention on match and market outcomes.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.