The problem of identifying land use spillovers: Measuring the effects of open space on residential property values
The problem of identifying land use spillovers: Measuring the effects of open space on residential property values
复制标题
DOI：
10.1111/0002-9092.00193
复制
发表时间：
2001-08-01
影响因子：
4.2
通讯作者：
Bockstael, NE
中科院分区：
文献类型：
--
作者：
Irwin, EG;Bockstael, NE
开放空间通常被认为是吸引城市和郊区居民到位于大都市边缘之外的郊区的主要因素。休憩用地所提供的乡郊设施包括景色优美、提供康乐活动的机会，以及避免因发展而带来的不便，例如交通挤塞和空气污染。几个支付意愿的研究已经证明了开放空间的积极舒适价值（霍尔斯特德;比斯利，沃克曼和威廉姆斯），但开放空间的价值的证据显示偏好的方法-即从享乐模型-是有限的和混合的。在这篇文章中，我们考虑使用享乐定价模型与住宅物业销售的开放空间溢出估计的问题，并提供了一个解释，为什么开放空间效应的正舒适价值，即使它存在，可能并不总是经验检测。我们考虑的估计问题是识别问题，这是在一个特征住宅价格模型中出现的开放空间土地是私人持有和可开发的。1在这种情况下，被视为开放空间的地块是住宅用地市场的一部分，因此受到决定一个地点的住宅价值的相同经济力量的影响。因此产生了两个识别问题。第一个是标准型的计量经济学识别问题，由于内生解释变量。在测试地块i的住宅价值是否受到相邻地块j是否被开发的影响时出现了问题。地块j被开发的可能性是其住宅使用价值的函数，而住宅使用价值本身又是i是否被开发的函数。第二个问题的出现是因为导致地块在住宅使用中或多或少有价值的因素是空间相关的。如果这些空间相关的解释变量中的任何一个从估计中被省略，那么测量周围开放空间的任何变量都将与误差项相关。在这两种情况下，有偏的系数估计将导致。虽然我们专注于开放空间在这篇文章中，识别土地使用溢出是一个更广泛的问题，在享乐建模，适用于任何措施的土地使用外部性，其中特定的土地使用被测量的影响，由住宅市场。出现的识别问题是”反射问题”和”相关不可观测”问题的变体，这些问题在关于主体交互模型的文献中讨论过（Manski; Brock和Durlauf）。2.据我们所知，这一问题在《公约》中没有得到解决。
Open space is often cited as a primary attractor of urban and suburban residents to exurban areas located just beyond the metropolitan fringe. Included in the rural amenities afforded by open space are scenic views, recreational opportunities, and an absence of the disamenities associated with development, such as traffic congestion and air pollution. Several willingness-to-pay studies have demonstrated the positive amenity value of open space (Halstead; Beasley, Workman, and Williams), but evidence of the value of open space from revealed preference methods--namely, from hedonic models--is limited and mixed. In this article, we consider the issue of estimating open space spillovers using a hedonic pricing model with residential property sales and offer an explanation for why the positive amenity value of open space effects, even if it exists, may not always be empirically detected. The estimation problems that we consider are ones of identification, which arise in a hedonic residential price model when the open space land is privately held and devel-opable. 1 Under these circumstances, land parcels counted as open space are part of the market for residential land and there-fore subject to the same economic forces that determine a location's residential value. Two identification problems arise as a result. The first is the standard type of econometric identification problem due to endogenous explanatory variables. The problem arises in testing whether the residential value of parcel i is affected by whether a neighboring parcel j is developed. The likelihood that parcel j is developed is a function of its value in residential use, which is itself a function of whether i is developed. The second problem arises because the factors that cause parcels to be more or less valuable in residential use are spatially correlated. If any of these spatially correlated explanatory variables are omitted from the estimation, then any variable measuring surrounding open space will be corre-lated with the error term. In both instances, biased coefficient estimates will result.Although we focus specifically on open space in this article, identification of land use spillovers is a broader issue within hedonic modeling that applies to any measure of land use externalities in which the particular land use being measured is influenced by the residential housing market. The identifi-cation issues that arise are variants of the" reflection problem" and the problem of" cor-related unobservables" discussed in the liter-ature on agent interaction models (Manski; Brock and Durlauf). 2 This problem has not been, to our knowledge, addressed in the