|
Article Excerpt
Abstract
This paper provides the first study on the impact of noise barriers on the price of adjacent houses based on a repeat sale analysis (RSA). RSA allows us to empirically examine the differential between the prices of houses sold before and after an event that may have affected their value, and after other relevant variables such as the evolution of the real estate market and major renovations performed on the house are controlled. This paper focuses on the neighborhood of Laval, a suburb of Montreal, where a large noise barrier was built in 1990 along a highway. The data set contains transaction information on 134 houses that were sold at least twice from 1980-2000. The empirical result will show that the noise barrier induced a decrease of 6% in the house prices in our sample in the short run, while it had a stronger negative impact of 11% in the long run.
Introduction
Quality of life is a growing preoccupation in our society. People are increasingly concerned about the level of noise in their environments. Thus, it is not surprising that an increasing investment has been made in noise barriers in recent years. (1) In order to determine whether this investment contributes to increasing a society's welfare, one has to be aware of all its effects. In general, noise barriers are efficient at reducing the level of noise (Ouimet, 1994), but they have other consequences, such as affecting the prices of adjacent houses. It is well established that an increase in noise levels leads to a decrease in the price of houses (for a recent survey, see Boardman, et al., 2005), so a reduction in noise should have the converse effect. However, some people have argued that the aesthetic impact of walls, or their impact on luminosity, could lead to reduced property values, especially for houses located very close to the shielded dwelling (Kamerud and von Buseck, 1985). Therefore, altogether, the net effect of a noise barrier on the price of adjacent houses is theoretically ambiguous, and has to be resolved empirically.
To our knowledge, only two existing papers have tackled this question, but, as we will show below, none of them has done so in a proper fashion nor reported unambiguous results. This paper is the first study on the impact of noise barriers on the price of adjacent houses based on a repeat sale analysis (RSA), arguably the best methodology for addressing this question. In an RSA, one seeks to identify the impact of an event on the price of houses sold before and after the event. If there is a significant change in prices between the two transactions, this may be attributed to the event. Of course, for that to be true, the researcher must have controlled for other changes that may have had an effect on a house's price between two sales, such as the evolution of the real estate market and major renovations performed on the house. In the past, RSA was used to determine the impact of the construction of a train station (Gatzlaff and Smith, 1995), a highway (Palmquist, 1982), or the discovery of a toxic site (Kohlhase, 1991; Mendelsohn, et al., 1992), on the value of adjacent houses, but it has never been used to evaluate the impact of a noise barrier.
We collected our data from the neighborhood of Laval, a suburb of Montreal, where a large noise barrier was constructed in 1990 along a highway. We were able to obtain information on 154 houses sold at least twice from 1980-2000. In addition, we were able to obtain data on the real estate market in the area for the whole period, as in most RSAs, but also on the demographic characteristics of the area and major renovations carried out on these houses throughout the time span. To our knowledge, this was one of the first times that information on major renovations was available for an RSA, allowing us to conduct an "augmented" RSA. (2) We concluded that the noise barrier induced a decrease of 6% in house prices in our sample in the short run, and 11% in the long run.
The rest of this paper is organized as follows: Section 2 presents a survey of existing studies on the impact of noise barriers. Section 3 discusses our empirical strategy and data. Section 4 presents our empirical results. Section 5 concludes.
A Survey of the Existing Literature
As mentioned earlier, there is a general consensus in the literature as to the negative relationship between the noise level and the price of adjacent houses. Most of the studies that have tackled this question have used the hedonic pricing method. This method implies that regressions are run when the price of a house is related to a vector of characteristics (physical characteristics of the house, market conditions, and neighborhood), including a variable capturing a certain environmental attribute, such as the noise level. Recent surveys (e.g., Boardman, et al., 2005) reported a "noise depreciation sensitivity index" of around 0.65, which means that if the noise level increases by one Noise Exposure Forecast (NEF), then the price of an affected house decreases by an average of 0.65 percent.
Some authors (Palmquist, 1982; Mendelsohn, et al., 1992) have criticized this method, arguing that with samples of houses from different neighborhoods, it is difficult to distinguish between the effect of any environmental attribute on the price of adjacent houses and the effect of unobserved characteristics on that price (e.g., criminality in the neighborhood). These authors have suggested a variant of the standard hedonic pricing method, or the RSA we described above. Using methodologies along these lines, the authors of two papers have looked specifically at the impact of noise barriers on the price of adjacent houses. We will describe and comment on them below.
Kamerud and von Buseck (1985) studied two sites--Troy Meadows and Lakewood, both located near the same highway in Michigan, USA. In 1974, a natural sound barrier (earth berm) was built at Troy Meadows, and this led to real noise reduction estimated...
|
