ABSTRACT

Government intervention in the housing market in response to the 2007–2010 mortgage crisis was driven in part by research showing that foreclosures lower neighboring housing values and thus increase neighbors’ risk of foreclosure. Researchers have consistently identified a negative spillover effect of foreclosures on nearby housing values, but the magnitude of the effect varies widely across studies. Although this variation is due, in part, to differences in the geographic region, time period, and empirical strategy of prior research, we argue that the spillover effect on nearby housing prices exhibits hyper-local variation, which may be obscured by models that aggregate spillover effect estimates within existing geographic units. In this article, we employ geographically weighted regression to capture the extent of spatial and temporal variation of foreclosure spillover effects in three Ohio metropolitan statistical areas. We find extensive heterogeneity of foreclosure spillover effect estimates over time and across space, suggesting that such spillovers perhaps should not be thought of as universal phenomena. These findings raise the possibility that policies and programs designed to intervene in the housing market should analyze and use local variation in the negative externalities of foreclosure to best target scarce resources within and across communities.

摘要

政府为应对 2007-2010 年抵押贷款危机而对住房市场进行干预，部分原因是研究表明止赎降低了邻近房屋的价值，从而增加了邻居的止赎风险。研究人员一直在确定止赎对附近房屋价值的负面溢出效应，但影响的程度在不同的研究中差异很大。虽然这种变化部分是由于先前研究的地理区域、时间段和经验策略的差异，但我们认为对附近房价的溢出效应表现出超局部变化，这可能被汇总的模型所掩盖。现有地理单位内的溢出效应估计。在本文中，我们采用地理加权回归来捕捉三个俄亥俄州大都市统计区域止赎溢出效应的空间和时间变化范围。我们发现止赎溢出效应估计随时间和跨空间存在广泛的异质性，这表明此类溢出可能不应被视为普遍现象。这些发现提出了一种可能性，即旨在干预住房市场的政策和计划应该分析和利用止赎负外部性的当地差异，以最好地瞄准社区内部和社区之间的稀缺资源。暗示这种溢出可能不应该被视为普遍现象。这些发现提出了一种可能性，即旨在干预住房市场的政策和计划应该分析和利用止赎负外部性的当地差异，以最好地瞄准社区内部和社区之间的稀缺资源。暗示这种溢出可能不应该被视为普遍现象。这些发现提出了一种可能性，即旨在干预住房市场的政策和计划应该分析和利用止赎负外部性的当地差异，以最好地瞄准社区内部和社区之间的稀缺资源。