The Microfoundations of Housing Market Dynamics

【摘要】房价动态的微观基础（0721136）：Patrick Bayer和Christopher timmins住宅建筑的波动在经济商业周期中起着核心作用。在低利率的推动下，1997年至2006年期间美国房价飙升，以及随后的建筑热潮，被普遍认为是这一时期经济扩张的推动力。而且，对房地产市场大幅回调的担忧在美联储随后一段时间的货币政策决定中发挥了核心作用。然而，尽管研究人员已经记录了房地产市场动态的许多重要宏观经济方面，但对房地产繁荣-萧条周期中不同时刻的个人购房者和卖家的行为知之甚少。特别是，一些关于本地房地产繁荣-萧条周期的性质和长度的根本决定因素的问题仍然悬而未决。该项目的智力价值源于其设计的两个新颖特征，这使得分析能够提供有关住房市场动态的新见解。首先是建立一个新的数据集，该数据集使用了15年来数百万笔住房交易的详细信息。第二是开发一个易于处理的社区选择模型，允许在动态环境下研究个人的房屋购买和出售决策。项目中的实证应用遵循两条主要的调查路线，力求提高我们对(i)房价波动的根本原因和（ii）推动城市种族隔离动态的根本因素的理解。第一行研究的主要目标是在住房市场的微观经济基本要素（即，控制个人买卖决定的因素）和总体市场动态（例如，价格的积极持续，当地住房泡沫等）之间提供明确的联系。特别重要的问题是，个人购房者和卖家是否对当地房地产周期不同阶段的房地产市场动态有理性预期。换句话说，个人是否会在当地房地产繁荣时期变得非理性繁荣，而在低迷时期变得过度悲观？第二项研究包括三篇论文，旨在提高我们对推动种族隔离动态的微观经济力量的理解。前两篇论文分别对住房交易中的种族歧视和种族小费进行了新的实证检验。每篇论文都使用交易数据的详细性质来设计测试，这些测试受到的混淆问题远少于迄今为止文献中出现的问题。这一研究领域的第三篇论文使用社区选择的估计动态模型来更好地理解偏好与期望在价格和社区动态中在种族隔离的演变和持续中的作用。这个项目有许多更广泛的影响。作为该项目的一部分，开发的新数据集和方法可能适用于研究住房市场和城市中更广泛的动态现象。此外，进一步了解住房市场波动的性质和原因，对联邦和地方两级制定政策都有直接好处。

AbstractThe Microfoundations of Housing Price Dynamics (0721136)Patrick Bayer and Christopher TimminsFluctuations in home building play a central role in economic business cycles. Fueled by low interest rates, the upsurge in housing prices in the US between 1997-2006, and the subsequent building boom, is generally considered to have been a driving force in the expansion of the economy over that period. And, worries about a major housing market correction have played a central role in the determination of monetary policy by the Fed in the subsequent period. Yet while researchers have documented many of important macroeconomic aspects of housing market dynamics, little is understood about the behavior of individual home-buyers and sellers at various points in a real estate boom-bust cycle. In particular, a number of questions about the fundamental determinants of the nature and length of local real estate boom-bust cycles remain open. The intellectual merit of the project derives from two novel features of its design, which permit the analysis to deliver new insights concerning housing market dynamics. The first is the construction of a new dataset that uses detailed information on millions of housing transactions over a 15-year period. The second is the development of a tractable model of neighborhood choice that permits the study of individual home-buying and selling decisions in a dynamic context. The empirical applications in the project follow two main lines of inquiry, seeking to improve our understanding of (i) the underlying causes of housing price fluctuations and (ii) the underlying factors driving the dynamics of racial segregation in cities. The principal goal of the first line of research is to provide a clear linkage between the microeconomic primitives of the housing market (i.e., the factors governing individual buying and selling decisions) and the aggregate market dynamics (e.g., the positive persistence of prices, local housing bubbles, etc). Of particular importance is the question of whether individual home buyers and sellers have rational expectations about housing market dynamics at different points in a local real estate cycle. In other words, do individuals become irrationally exuberant during local real estate booms and unduly pessimistic during downturns? The second line of research includes three papers that aim to improve our understanding of the microeconomic forces driving the dynamics of racial segregation. The first two papers develop new empirical tests for racial discrimination in housing transactions and racial tipping, respectively. Each paper uses the detailed nature of the transaction data to devise tests that are subject to far fewer confounding issues than those that have appeared in the literature to date. The third paper in this line of research uses the estimated dynamic model of neighborhood choice to better understand the roles of preferences versus expectations about price and neighborhood dynamics in the evolution and persistence of racial segregation.This project has a number of broader impacts. The new dataset and methodologies developed as part of the project are potentially applicable to the study of a much wider set of dynamic phenomena in housing markets and cities. Moreover, an increased understanding of the nature and causes of housing market fluctuations has direct benefits for the setting of policy at both the federal and local levels.