A widely used approach to valuing transport infrastructure is to look at its effects on housing prices, in treated vs control regions, before and after it is built. But anticipation effects mean that housing prices may start changing as soon as a project is announced. This creates complications for an analysis of the capitalization effects on prices. A long span of data is potentially required—because the time from a project’s announcement to completion may be many years—and the assumption of common price trends between the treated and control regions must be maintained over this time span. An alternative approach is to use rents. This may be advantageous as rents are more directly related to the infrastructure’s service-flow. Though account needs to be taken of the fact that those homes which are rented are often not representative of the housing stock as a whole. We investigate these issues for a new train line in Sydney and contrast the results from using both prices and rents.

评估运输基础设施的一种广泛使用的方法是，在建设之前和之后，查看其对已处理地区与控制地区房价的影响。但是，预期效应意味着，一旦宣布一个项目，房价可能会开始发生变化。这为分析资本化对价格的影响带来了麻烦。可能需要大范围的数据-因为从项目宣布到完成的时间可能是很多年-并且必须在此时间段内保持被处理区域和控制区域之间的共同价格趋势的假设。另一种方法是使用租金。这可能是有利的，因为租金与基础设施的服务流更直接相关。尽管需要考虑这样一个事实，即那些租用的房屋通常不能代表整个房屋存量。我们针对悉尼的一条新铁路线调查了这些问题，并对比了使用价格和租金得出的结果。