School Choice Mechanisms
Project Outcomes Statement
Adam Kapor, Christopher Neilson, and Seth Zimmerman have completed research for their NSF-supported project "School Choice Mechanisms" (NSF Award #1629226). Research conducted for this project helped advance the technical understanding of school choice mechanism design, and also contributed to policy improvements in New Haven, Connecticut, the district that was the focus of the empirical work.
On the technical side, we wrote and published a scientific paper titled "Heterogeneous Beliefs and School Choice Mechanisms" (American Economic Review, forthcoming). This paper combines surveys of families participating in school choice with administrative data from the choice system in New Haven, Connecticut to explore what households do and do not understand about the choice process, and the implications of imperfect information for policy design.
Many districts in the United States and abroad use centralized choice systems to assign students to schools. In centralized choice, students submit rank-ordered lists of schools to a central authority, which then uses a process known as assignment mechanisms to place students in schools. Until 2019, New Haven used an assignment mechanism that rewarded participants who engaged in informed strategic behavior when filling out their applications. Strategic behavior here is taken to mean submitting an application that consists of something other than one's true preferences, listed in order. Versions of this mechanism, known as the "Immediate Acceptance" or "Boston" mechanism, are also used in cities like Cambridge, Massachusetts, Charlotte, North Carolina, and Beijing. A common alternate approach, known as the "Deferred Acceptance" mechanism, aims to make reporting one's true preferences in order a dominant strategy. The tradeoff between these two approaches to choice is that while the Boston mechanism may raise welfare by allowing participants to express preference intensity through strategic play, the additional challenges of strategic play may lead to welfare-reducing mistakes. Applicants from low socioeconomic status backgrounds may be more likely to make costly mistakes if the resources available to them for learning about the choice process are limited.
Our survey data show that choice participants engage in strategic behavior, but do so on the basis of beliefs about admissions chances that are often incorrect. Using the survey data as input to empirical model of school choice shows that welfare losses from application mistakes in the Boston Mechanism outweigh the ability to express preference intensity, and therefore that a switch to the simpler Deferred Acceptance approach to choice would likely be welfare-improving.
On the policy side, our findings contributed to a change in New Haven's approach to choice. After completing an initial draft of our paper, we discussed our findings with the district leadership team, and they decided to change New Haven’s choice mechanism from Boston to Deferred Acceptance, beginning with the 2019 choice process (Peak, 2019). Going forward, we will continue to collaborate with the school district to evaluate the effects of the policy change and to assist the district in developing its internal capacity for the design and evaluation of choice policy.
Kapor, Adam, Christopher Neilson, and Seth Zimmerman. "Heterogeneous Beliefs and School Choice Mechanisms." American Economic Review, forthcoming.
Peak, Christopher. "Profs Tapped to Fix School-Choice Lottery." New Haven Independent, January 8th, 2019. Available at https://www.newhavenindependent.org/index.php/archives/entry/professors_school_choice_lottery/. Accessed February 12th, 2020.
Supported by the National Science Foundation grant #1629226
More from NBER
In addition to working papers, the NBER disseminates affiliates’ latest findings through a range of free periodicals — the NBER Reporter, the NBER Digest, the Bulletin on Retirement and Disability, the Bulletin on Health, and the Bulletin on Entrepreneurship — as well as online conference reports, video lectures, and interviews.