National Academies Press: OpenBook

Effective Methods for Environmental Justice Assessment (2004)

Chapter: Chapter 12 - Land Prices and Property Values

« Previous: Chapter 11 - Visual Quality
Page 269
Suggested Citation:"Chapter 12 - Land Prices and Property Values." National Academies of Sciences, Engineering, and Medicine. 2004. Effective Methods for Environmental Justice Assessment. Washington, DC: The National Academies Press. doi: 10.17226/13694.
×
Page 269
Page 270
Suggested Citation:"Chapter 12 - Land Prices and Property Values." National Academies of Sciences, Engineering, and Medicine. 2004. Effective Methods for Environmental Justice Assessment. Washington, DC: The National Academies Press. doi: 10.17226/13694.
×
Page 270
Page 271
Suggested Citation:"Chapter 12 - Land Prices and Property Values." National Academies of Sciences, Engineering, and Medicine. 2004. Effective Methods for Environmental Justice Assessment. Washington, DC: The National Academies Press. doi: 10.17226/13694.
×
Page 271
Page 272
Suggested Citation:"Chapter 12 - Land Prices and Property Values." National Academies of Sciences, Engineering, and Medicine. 2004. Effective Methods for Environmental Justice Assessment. Washington, DC: The National Academies Press. doi: 10.17226/13694.
×
Page 272
Page 273
Suggested Citation:"Chapter 12 - Land Prices and Property Values." National Academies of Sciences, Engineering, and Medicine. 2004. Effective Methods for Environmental Justice Assessment. Washington, DC: The National Academies Press. doi: 10.17226/13694.
×
Page 273
Page 274
Suggested Citation:"Chapter 12 - Land Prices and Property Values." National Academies of Sciences, Engineering, and Medicine. 2004. Effective Methods for Environmental Justice Assessment. Washington, DC: The National Academies Press. doi: 10.17226/13694.
×
Page 274
Page 275
Suggested Citation:"Chapter 12 - Land Prices and Property Values." National Academies of Sciences, Engineering, and Medicine. 2004. Effective Methods for Environmental Justice Assessment. Washington, DC: The National Academies Press. doi: 10.17226/13694.
×
Page 275
Page 276
Suggested Citation:"Chapter 12 - Land Prices and Property Values." National Academies of Sciences, Engineering, and Medicine. 2004. Effective Methods for Environmental Justice Assessment. Washington, DC: The National Academies Press. doi: 10.17226/13694.
×
Page 276
Page 277
Suggested Citation:"Chapter 12 - Land Prices and Property Values." National Academies of Sciences, Engineering, and Medicine. 2004. Effective Methods for Environmental Justice Assessment. Washington, DC: The National Academies Press. doi: 10.17226/13694.
×
Page 277
Page 278
Suggested Citation:"Chapter 12 - Land Prices and Property Values." National Academies of Sciences, Engineering, and Medicine. 2004. Effective Methods for Environmental Justice Assessment. Washington, DC: The National Academies Press. doi: 10.17226/13694.
×
Page 278
Page 279
Suggested Citation:"Chapter 12 - Land Prices and Property Values." National Academies of Sciences, Engineering, and Medicine. 2004. Effective Methods for Environmental Justice Assessment. Washington, DC: The National Academies Press. doi: 10.17226/13694.
×
Page 279
Page 280
Suggested Citation:"Chapter 12 - Land Prices and Property Values." National Academies of Sciences, Engineering, and Medicine. 2004. Effective Methods for Environmental Justice Assessment. Washington, DC: The National Academies Press. doi: 10.17226/13694.
×
Page 280
Page 281
Suggested Citation:"Chapter 12 - Land Prices and Property Values." National Academies of Sciences, Engineering, and Medicine. 2004. Effective Methods for Environmental Justice Assessment. Washington, DC: The National Academies Press. doi: 10.17226/13694.
×
Page 281
Page 282
Suggested Citation:"Chapter 12 - Land Prices and Property Values." National Academies of Sciences, Engineering, and Medicine. 2004. Effective Methods for Environmental Justice Assessment. Washington, DC: The National Academies Press. doi: 10.17226/13694.
×
Page 282
Page 283
Suggested Citation:"Chapter 12 - Land Prices and Property Values." National Academies of Sciences, Engineering, and Medicine. 2004. Effective Methods for Environmental Justice Assessment. Washington, DC: The National Academies Press. doi: 10.17226/13694.
×
Page 283
Page 284
Suggested Citation:"Chapter 12 - Land Prices and Property Values." National Academies of Sciences, Engineering, and Medicine. 2004. Effective Methods for Environmental Justice Assessment. Washington, DC: The National Academies Press. doi: 10.17226/13694.
×
Page 284
Page 285
Suggested Citation:"Chapter 12 - Land Prices and Property Values." National Academies of Sciences, Engineering, and Medicine. 2004. Effective Methods for Environmental Justice Assessment. Washington, DC: The National Academies Press. doi: 10.17226/13694.
×
Page 285

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

275 CHAPTER 12. LAND PRICES AND PROPERTY VALUES OVERVIEW Scholars and practitioners have long recognized that highway projects influence land and property values. Importantly, the influence of a highway project on land or property values is based on many of the other impacts that are discussed earlier in this guidebook. Highways can influence the accessibility of various parcels and may produce disamenities such as noise, air quality, and visual impacts. Construction disruption can also influence property prices near a transportation facility. The net effect of these and all other impacts can be reflected in changes in the value of the property. Thus changes in property value are not distinct from the other impacts that have been discussed in this guidebook. Instead, property values reflect the broad range of impacts from highway projects, and so provide another window into understanding both the effects of transportation system change and the implications for environmental justice. Property prices reflect the full range of positive and negative impacts of transportation system changes. More generally, property prices reflect all location-specific characteristics of a parcel, including characteristics that are not related to transportation, such as desirable views, proximity to good schools, crime rates in the area, and noxious nearby land uses, such as toxic waste dumps. The fact that property prices reflect all impacts, both positive and negative, has both advantages and disadvantages. On the plus side, once the impact of a highway project on property prices has been determined, one has a good summary measure of the net impact of all of the possible effects of the highway. In some cases, rather than assessing each of the various effects discussed in this guidebook individually, an environmental justice analysis might focus on the total impact of the highway by examining impacts on property values. On the negative side, like many summary measures, a property value analysis can obscure information on specific effects of the highway. If the property value analysis does not distinguish between the various effects of the highway, one might be unable to comment on how the influence of a highway can be disaggregated into impacts based on accessibility, noise, air quality, and other effects of the highway.1 STATE OF THE PRACTICE The link between property values and transportation has long been recognized. Models of urban development often incorporate the influence of transportation on land values. See, for example, descriptions of classic models of urban form in Alonso (1964; 1972) and Fujita (1989). Empirical studies of highways and property values date to the early years of the Interstate Highway System (Adkins 1959; Mohring 1961). 1 This is not necessarily the case. A property value analysis can be designed to illuminate the independent effects of, for example, changes in accessibility, noise, air quality, and other impacts of the highway. In that case, the result of a property value analysis is not a summary measure but, rather, an analysis of the influence of several types of impacts on property values.

276 The methods for assessing the link between property values and highways are similarly well established. The methods described below are based largely on appraisal techniques and hedonic analyses that have been applied for decades. Every metropolitan area and small town has property appraisers, and their expertise and methods can be adapted to understand the impact of highway projects on land and property values. Hedonic analysis of property prices was pioneered in the 1970s and is now the subject of a large literature. While both appraisal techniques and hedonic (or regression) studies of property values can be applied to many phenomena other than transportation projects, both can and have been adapted many times to highways in ways that provide a strong foundation for incorporating land and property value impacts into an assessment of environmental justice. The state of the practice in this area is evolving rapidly. Until recently, property value data were difficult to obtain and hard to apply to geographically oriented studies of the sort required by an environmental justice analysis. For that reason, despite a wealth of theory, applied property value studies were rare until a decade ago. Both technology and data availability have changed that. There is a wealth of data sources for property values. In most states, local tax assessor offices collect data on property sales. Those data are increasingly available, either at no charge from the assessor’s office or (more commonly) for a fee from real estate data companies such as those described in Method 3 of this chapter. Local newspapers typically track the health of local real estate markets, and universities now often have units devoted to collecting and disseminating real estate market data. These data can be supplemented, as needed, with information on property values from tax appraisals or other sources, such as the U.S. Census public use micro sample. The rich availability of data combined with modern geographic information system (GIS) technology creates a powerful tool. Property sales data can be matched to specific parcels; and, with GIS, precise estimates of distance from a highway project or distance from other sources of impacts can be developed. The net effect is that the use of property values to understand the impact of public projects is poised to grow rapidly. Public agencies will increasingly be able to use information about property values or real estate markets to understand the impacts of their projects, highway projects included. SELECTING AN APPROPRIATE METHOD OF ANALYSIS Certain special issues must be considered when performing any environmental justice assessment of property values. These topics are described below. Retrospective versus predictive analyses. Most property value analyses of highways have so far been retrospective in nature. Typically, a researcher examines how access to an existing or newly built highway influences property values nearby. While that adds to our knowledge about the link between highways and property prices, a retrospective study is often not sufficient for environmental justice questions that arise as part of the project analysis phase. For many environmental justice questions, a predictive analysis is needed. The question is often, “if this highway project goes forward, what will be the environmental justice implications?” For property value-based environmental justice assessments, the objective often is predicting the

277 future impact of highway projects not yet built and then assessing how those property value impacts affect minority, low-income, or other protected populations. Fortunately, predictive analyses can often be an easy adaptation from retrospective analyses, and as land price and property value analyses become increasingly common, that adaptation will become even more feasible. Briefly stated, the retrospective analyses are the knowledge base for moving to predictive analyses. Future impacts of highway projects can be understood by applying what is known about past impacts of similar highway projects on similar properties. The methods described below can be used to predict how a highway project will impact property values. Often past studies will provide coefficients, slopes, or average effects that can be applied to similar projects to infer how future projects will influence property values. Owners and renters. Property value impacts affect both owners and renters. Economic theory suggests that the purchase price of a structure is directly linked to the rental value. For that reason, studies of property value impacts can and should be applied to both owner-occupied and rental property. For housing markets, most data sources report either the appraised value or the sale value of the property and as such apply most directly to the market for owner-occupied housing. If a property is rented, one can infer the magnitude of the impact on rental property by relating the rental price to the overall purchase price of the structure. What will likely be more important for an environmental justice analysis is to distinguish between benefits and costs that are borne differentially by landlords and tenants. In a simple situation, negative property value impacts on owners (landlords) will result in lower rents because the demand for the rental property will drop. In such situations, landlords will be harmed, because they will lose property value. Renters will both pay lower rents and experience disamenities, and so the perceived effects to renters are based on their individual values and perception. Thus, counterintuitively, the landlords might be more adversely affected by a negative highway impact than renters, depending on the renters’ values. To clarify this, consider the following example. A house near a new highway is subject to increases in highway noise. This disamenity reduces the number of persons willing to rent the house, and that drop in demand results in a lower rental price. The renter, living near the highway, experiences a noise disamenity, but if the rental market is competitive, the lower rental price will compensate the renter for the noise disamenity. The renter, on net, might be no worse off depending on his or her values and needs. The owner of the house, on the other hand, sees rental income (and the sale value of the house) drop, and so the landlord is adversely affected. The converse situation could also hold. Suppose that, for the same house near the same new highway, the renter has signed a long-term lease. The rent will not drop to reflect the noise disamenity until, at the earliest, the lease is up for renewal. Even then, if the rental market is not perfectly competitive, or if renters have imperfect information about the nature of the noise disamenity, rents might not drop to fully reflect the noise disamenity. Then the renter would be,

278 on net, disadvantaged by the impact of the highway, while the landlord could, if rental income does not drop at all, see no impact. The crucial question is whether and how quickly property prices adjust to reflect the impact of the highway, be that impact positive or negative. While understanding that is important in interpreting any property value analysis, it is especially important in understanding how land value impacts may be experienced by renters and landlords. Residential property and commercial property. Most property value studies, and most property value data sets, look at either residential or commercial properties. Combining the two in the same analysis is rare. Because environmental justice studies will sometimes need to understand the impact of highways on both residential and commercial properties, we recommend looking at both markets when that is appropriate. The analyses of the residential and the commercial markets will likely use different data and might yield different findings. For example, the literature has shown that being very close to a highway (typically less than a quarter mile) can be, on net, a disamenity that will depress the value of residential property (e.g., Langley 1981). The noise and (possibly) air quality impacts of the highway apparently outweigh accessibility advantages for residential properties at those distances. Yet the evidence does not as strongly suggest that commercial properties will experience the same disamenities. Apparently firms either benefit more from being very close to highways or the negative impacts (such as noise) matter less for commercial or office uses. The double counting critique. The impacts on property values should not be added to other impacts that are also influencing property values. Consider the following example. Suppose one found that persons near a highway would be bothered by increased traffic noise, and suppose surveys of those residents yielded a dollar value estimate of the noise impact. Suppose a transportation agency also conducted an assessment of how the highway would affect property values in the area adversely impacted by increased highway noise, and suppose that study showed that property values near the new highway would drop. One should not add both the estimated value of the noise impact from the survey and the lost property value together. To do so would count the impact of noise twice, since the lower property values are due in part (likely in large part) to the increased noise. An agency should choose one way to measure the impact of increased highway noise—the survey method or the property value assessment. Or the agency could consider both methods as alternative measures, and choose some way to average or otherwise use information from both methods. But adding the dollar value impact from both methods together will overstate the noise impact from the highway, because that impact would be counted twice—once by asking residents in a survey how they perceive they are impacted and a second time by assessing how home buyers pay less for homes in the noise contour of the new highway. Property value analysis can illuminate and verify analyses from other methods. Often times that will be good practice. Yet agencies should be aware not to double count impacts by adding the estimated impact from property value analysis to assessments of specific impacts that are driving the property value analysis. Remembering that property values are a summary measure and that changes in property values are derivative of the full range of effects of a highway project should

279 clarify this cautionary note. With that caution in mind, property analysis is one of the most powerful tools available in understanding the influence of highway projects on nearby residents and businesses, and the applicability of this tool, which already is high, will increase in the future as new data and GIS tools become available. Importance of impacts. In some cases, the collection of sufficient data to carry out an analysis of the probable effects of a transportation project on nearby property values may be quite costly. It may be prudent to begin with a simpler approach, like the first method listed below, to gain insight into the approximate magnitude of these effects. If the effects are unlikely to be sizable, it generally is not necessary to collect the data necessary to apply the more rigorous Method 3. METHODS Table 12-1 provides a summary of the methods presented in this chapter. Table 12-1. Summary of methods for analyzing land prices and property values Method Assessment level Appropriate uses Use when Data needs Expertise required 1. Market studies and expert opinion Screening/ detailed Project/ corridor Small project or where a more resource-intensive analysis is either not feasible or not necessary Low Data collection and interview 2. Property comparisons and appraiser opinion Detailed Project/ corridor Projects and alternatives assessments where sophisticated models are not available Medium Property appraisal 3. Hedonic regression Detailed Project/ corridor/ system Large projects or policies where changes in land and property values are expected High Statistical methods Method 1. Market studies and expert opinion This method can be the simplest and least data intensive of the methods described in this chapter. The application of this method can be as simple as having a real estate expert advise on the likely property value impacts of a highway project and then incorporating that assessment into an environmental justice analysis. Yet, as with almost any method, agencies can use variations that are more systematic and hence can provide more reliable inferences.

280 There are, broadly speaking, two types of expert judgment that can be obtained. In the first case, one can consult real estate experts, usually local professionals who are knowledgeable about local property markets. In the second case, an agency can contract with a firm or expert to conduct a market study. The two cases are likely to be distinguished by the rigor and amount of systematic analysis employed, even though both will rely heavily on expert understanding of the property markets. In some cases, a market study might combine expert analysis with the techniques from Method 2 (property appraisals) and Method 3 (hedonic regression). Experts might form their opinion, in part, based on appraisals or hedonic studies. So there can be cases where it is difficult to cleanly separate which of the three methods described in this chapter is being used, and the boundaries between the methods can blur at times. Yet having said that, the three methods described in this chapter—expert opinion, appraisals, and hedonic regression—each involve increasing levels of sophistication and are each appropriate in different circumstances, and so it is sensible to describe each separately. When to use. The method of expert opinion is most appropriate in cases where a more resource- intensive analysis is either not feasible or not necessary. For small projects, with likely small environmental justice implications, an expert analysis might be sufficient. Similarly, smaller agencies with limited technical capabilities will find that careful use of expert opinion can go a long way toward illuminating the potential environmental justice impacts of a project. This method can be used as a preliminary stage in the analysis, to identify the location of possible positive or negative impacts. If, for example, the use of expert opinion suggests that the property value impacts might lead to importantly large environmental justice concerns, the other two methods (appraisals or hedonic regression) can be used to understand the magnitude of the property value impacts and how those positive or negative impacts will be incident on different groups. Analysis. The data collection should yield information about where property values will be expected to increase and where they will be expected to decrease as a result of the highway project. In some cases, agencies might even have estimates of magnitudes of property value increases or decreases at particular locations. The analysis involves comparing that information to the spatial distribution of low-income or minority tenants (owners and renters of residences and also possibly business space) to understand if those groups are disproportionately affected by the highway project. Data needs, assumptions, and limitations. The data collection involves canvassing real estate experts to assess the likely impact of a highway project on nearby properties. This can include focus groups or systematic surveys of experts. Less systematically, an agency might simply discuss the impact with a few experts and carefully document the results. In some cases, such less systematic assessments might be sufficient. It is also necessary to obtain data on minority, low-income, and other protected populations’ characteristics, such as is available from the census. Information on housing unit occupancy and owner/renter tenancy are also available from the census. This method provides general

281 information on the potential for a project to affect property values. While expert opinion might not be able to precisely quantify the magnitude of land price or property value impacts, expert opinion can be quite useful in assessing where property values might be positively or negatively influenced by a highway project. Typically, this method will be regarded as less systematic than the other two methods described below. Certainly, agencies should carefully document how experts arrived at their opinion, so that the agency can defend the analysis, if necessary, in the future. Results and their presentation. Generally, the results of this method will be contained in a report summarizing the insights of the experts contacted. Two levels of analysis are possible. In the simplest case, the experts may provide an indication of the properties that would be affected if the highway project were to be carried out. A more extensive analysis may reveal the experts’ opinions as to the approximate magnitude of effects on property values. Assessment. The reliability of this method depends crucially on the knowledge of the experts and the care taken to interpret their opinions. Agencies should seek to understand not only the experts’ opinions, but also how reliable (or precise) they believe their analysis to be. Overall, this method is best used either when the scope of the project or the expected environmental justice implications are of a small enough magnitude that large resource investments are not justified or when the agency does not have the means to conduct a study using the other methods described below. Method 2. Property comparisons/appraiser opinion The method of property comparisons is widely used by property appraisers to determine values of houses and commercial properties. To conduct the analysis, an appraiser finds recently sold properties within the same vicinity and with characteristics similar to the property being appraised. The properties comparable to the property to be appraised are known as “comps.” Their sale prices are adjusted to yield the appraised value of the property in question. In practice, the comparison criteria include dwelling age, various physical characteristics, and a broad range of location characteristics. When to use. Since this method is based largely on appraisers’ judgment, property value impacts of transportation projects of any size can be evaluated in this way. Appraisers’ judgment may well be the best estimates of property values when sophisticated models are not readily available. The agency should give special attention to finding comparison properties near transportation projects that are similar to the project being analyzed. Analysis. To use this method for assessing environmental justice impacts of transportation projects, appraisers must include a precise measure of how transportation accessibility is capitalized into property value. If there are numerous recent property sales, it may be easy to find comps with similar accessibility characteristics. Otherwise, a gradient must be developed, such that it can explain variation in property value by variations in accessibility. For instance, past empirical studies of the relationship between house values and distances to highways revealed a gradient of approximately $1 to $4 per foot, which implies that each foot of distance away from a highway will reduce house values by $1 to $4 (Boarnet and Chalermpong 2001). In this way,

282 even when comps with similar accessibility characteristics cannot be found, other comps can be used, and the values of those comps can be adjusted using an accessibility gradient. Data needs, assumptions, and limitations. Data to be collected include physical characteristics (such as number of rooms, floor area, quality of construction, piping condition), transportation accessibility, amenities (such as school quality), and disamenities (such as crime and noise). More importantly, location characteristics that involve transportation accessibility must be evaluated and recorded. These include, for example, distance to the transportation facility of interest, such as highway ramps or transit stations. Other neighborhoods in the region with similar transportation accessibility characteristics must then be identified. Appraisers can then search for comps using the information about properties to be appraised. Market rents and sale prices of these comps are then used to estimate the values of properties in question. Special care must be given to identify good comps. Since human judgments can vary widely, this method may not be suitable if similar properties are difficult to find. Without good comps, this method, which relies solely on appraisers’ judgment, may not yield an accurate forecast of changes in property values. Results and their presentation. For the potentially affected properties, each appraiser provides a report on the likely change in value, based on comparable properties. If two or more appraisers are used, the analyst can compile a brief report providing a range in impacts on the values of affected properties. Assessment. Difficulties in finding similar properties or land with similar accessibility and other characteristics may limit the use of this method. Some location characteristics that may affect property values, such as air and noise pollution, may be difficult to compare, and thus cannot be assessed. Yet the advantages of this method are that all metropolitan areas have a large number of property appraisers and property appraisal techniques have become highly standardized. By using a panel of several appraisers, a transportation agency can obtain a range of estimates that will help the agency cope with differences in interpretation that are inherent in this method. Method 3. Hedonic regression Hedonic regression is a statistical method for evaluating impacts of various kinds on house and residential property prices. These factors range from basic attributes of houses, such as size, number of rooms, and age, to location attributes, such as accessibility to highways and other amenities and disamenities, including the quality of public services, local tax burdens, and public safety. The method utilizes a wide range of data to estimate how a change in a given factor increases or decreases house prices. Environmental justice analyses of new highway projects can be carried out by applying this method to forecast the highway-induced changes in prices of houses that are owned or rented by people in protected population groups. For example, increased levels of traffic noise caused by a new highway can reduce the desirability of houses nearby and therefore reduce sales prices of these houses. If the owners or tenants belong to protected population groups, actions may need to

283 be taken to compensate them for the losses or disamenities. The losses can often be quantified using the output of the regression model. When to use. The regression method is appropriate when the project in question is of significant scale. A new grade-separated highway, for example, will not only change transportation accessibility for residents in the area, but also may have major environmental impacts. These changes will be translated into changes in house prices in the surrounding areas, which can be evaluated using the hedonic method. However, smaller projects, such as a road widening in a commercial district, will likely have minor impacts on house prices, and the method may not be able to detect any significant changes in house prices.2 Also, this method requires extensive data as well as technical skills, both of which may not be available in smaller agencies. Analysis. A multiple regression technique is used to isolate effects that various factors have on house prices. These factors are called explanatory (independent) variables because they are used to explain the variation in house and property prices (the dependent variable). The influence of each factor (independent variable) on property values is reflected by its coefficient; that is, if the coefficient of an independent variable, such as floor area, is positive, an increase in that variable (larger floor area) will increase the property value, all else being equal. Statistical knowledge and judgment are required for hedonic modeling to make decisions such as which factors should be used as explanatory variables. The estimation of different specifications of hedonic regressions is carried out by commercial software, such as SPSS, SAS, STATA, or GAUSS. The performance of each estimated model is reflected by two measures – the R2 (goodness of fit of the overall equation) and t-statistics of the coefficients. The R2 ranges from 0 to 1 and shows how well the independent variables together explain the variation in the dependent variable (property prices). The t-statistics show how precisely each coefficient is estimated. Environmental justice analyses for highway projects can be carried out as follows. First, identify and quantify changes in independent variables that result from the project. For example, a new highway may reduce the distance from a house to the nearest highway.3 The highway project might also increase the noise levels and create visual blight. Changes in property values can be determined from the coefficients of a hedonic model. For example, if the coefficient on distance to the nearest freeway is -$1 per foot, and the highway project will reduce the distance from a particular house to the nearest freeway by 2,000 feet, then that house would gain value equal to (-$1) x (-2,000) = $2,000. If noise levels at the house rise by 5 decibels (dB) due to the highway project and if the coefficient on noise in a hedonic regression indicates that property prices are reduced by $100 per dB, then the noise impact due to the highway project will reduce the house’s value by (-$5) x (100) = -$500. Thus, ignoring visual effects, the net effect of the changes in accessibility and noise caused by the completion of the highway project increases the house’s value by $1,500. 2 Note that a road widening in a commercial district might have larger effects on the prices of commercial property. 3 The straight-line distance to the nearest highway is one measure of accessibility. Others include street-network distance to the nearest highway and distance to the nearest mass transit station.

284 The hedonic regression method is used to explain the variation in property prices by their characteristics. For example, a house that is 500 feet away from a rail transit station is more expensive than the one that is 5 miles away partially because of its better accessibility to transit. Similarly, a house near the beach is more expensive than the one farther inland, and a house in a good school district is more expensive than the one in an under-performing school district. Each of the proceeding characteristics—access to transit, proximity to a beach, and school quality—are location-specific. GIS software, such as ArcView, is a useful tool that can help quantify the value of these variables.4 Agencies can use GIS software to generate many of the location-specific variables, which in turn can be used for regression analysis of property prices. The most obvious variables needed for transportation analysis are accessibility measures, often approximated by distance from a property to a transportation facility (straight-line and street- network distance have both been used). You can use GIS software to measure the distance from each house to the nearest transportation facility. Other location-related variables are available based on local jurisdictions. For example, average SAT scores, which can be used as a measure of school quality, are often available from school districts. GIS software can be used to match house locations to school district boundaries. Similarly, GIS software can match house locations to other jurisdictional boundaries, so that data available from various jurisdictions can be used, often as control variables, in an environmental justice analysis of highway projects. For example, you could assign a city’s average crime rate (which reflects the city’s public safety) to a house according to where the house is located. The procedures for creating independent variables using GIS are summarized below. Step 1 – Geocode the street addresses of houses. You can match (geocode) each house’s street address, obtained from a source such as tax assessor’s records, to the GIS street-network map. The geocoding process can be automated to match many addresses at the same time, and GIS software packages typically include routines for address matching and geocoding. Step 2 – Verify match accuracy. You can examine the accuracy of the address match by either using diagnostics from the GIS software or by comparing a small number of computerized matches to printed maps. For example, one can randomly draw 100 houses and compare the GIS match with published street maps. If relatively few of the GIS matches are inconsistent with the paper map, then the automated address matching can be judged to be relatively accurate. Step 3 – Create independent variables. The spatial join feature of ArcView can be used to determine the distance (straight-line or street-network) from a property (from Step 2) to a transportation facility. The same feature can also be used to assign amenity/disamenity characteristics that follow jurisdictional boundaries. This requires GIS maps for jurisdictions on which the data are based, such as school districts or municipalities. Some other aspects of data preparation are listed below: • The accuracy of the raw data should be verified. Data quality can vary across different sources and even across time periods for the same source. 4 ESRI’s ArcView for desktop computers can be obtained from http://www.esri.com/.

285 • Home sales prices should be adjusted for inflation, using, for example, the consumer price index for housing in the region or metropolitan area. This index can be obtained from the U.S. Bureau of Labor Statistics’ Web site, http://stats.bls.gov/. • Non-arms-length transactions should be excluded from the analysis to ensure that sales prices reflect market transactions. For example, transactions, in which buyers and sellers share the same last name should be dropped, or at least scrutinized closely, as those sales might reflect non-arms-length transfers of property from, for example, parents to children. The sales price in such cases might not reflect the market value of the home if, for example, the seller is willing to sell the house below market rate to a relative. Data needs, assumptions, and limitations. Due to the nature of the statistical method, the hedonic regression technique requires input data that cover a reasonably wide range of properties. House and property sales records are the main source of data for hedonic analysis. These data generally include property sale prices, rents, physical characteristics of property, street address, sellers and buyers, loan amount, etc. Such records can be obtained from local government agencies, such as tax assessor and collector’s offices or appraisal data support companies that compile such data from these agencies. Measures of accessibility can be generated from street addresses, using GIS software. In addition to these sources, information about amenities and disamenities can be gathered from local authorities, such as local police departments and state departments of justice for crime, and school districts for school quality. Finally, information about environmental impacts is available in environmental quality reports, which are commonly required in large highway projects. Visual effects, being unique to each project and location, might be difficult to assess in a generalizable way using a hedonic regression approach. Also, there may be variation in different types of effects across places. For example, some neighborhoods may value quietness more than others. As a result, the model estimated from data in one neighborhood may not be transferable for use in other places. Results and their presentation. This method produces regression equations that show the additive effects of various attributes on residential property values. If the appropriate data are available and included in the analysis, it is possible to estimate the effects of a change in a transportation facility on residential properties located at varying distances from the facility. These results can be easily presented to a nontechnical person in a way that gives him or her considerable insight. Assessment. The hedonic method is more objective and more firmly grounded in theory than the other methods described in this chapter. Changes in property values can be attributed directly to each different effect from a transportation project. However, there are substantial technical and data requirements associated with this method. Moreover, some effects, such as visual quality, cannot be easily quantified, and thus are difficult to analyze using hedonic regressions.5 5 Rough measures, such as dummy variables for visually blighted areas, can be used as independent variables. For example, a dummy variable for visual blight can be specified, such that it takes a value of 1 if a highway can be seen from a house and 0 otherwise. In this way, the effect of visual intrusion from highways on house prices can be isolated.

286 ADDITIONAL INFORMATION The following four tables provide information useful in performing the methods presented in this chapter. Table 12-2 provides a summary of findings from recent studies on house prices and local amenities and disamenities. In general, these studies found that good access to major highways has a positive effect on house prices, but when the house is very near to the facility, noise, dust, and other disamenities may counter the value of accessibility to some extent. Table 12-2. Selected recent studies of house prices and local amenities or disamenties Study Findings Studies of highways or other transportation infrastructure Boarnet and Chalermpong (2001) Impact of new toll roads on house prices in Orange County, CA; techniques used include hedonic estimation and repeat sales Langley (1981) Impact of the Washington, DC Beltway on house prices; found that house prices increase with the distance from the highway out to a distance of 1,125 feet, and then decrease beyond that distance Gatzlaff and Smith (1993) Effect of Miami rail transit on house prices; used repeat sales technique to construct house price index Huang (1994) Surveyed the literature on the impacts of transportation infrastructure, specifically transit stations and highways, on property value Kockelman and ten Siethoff (2002) Property values and highway expansion: an investigation of timing, size, location, and use effects Studies of other (nontransportation) environmental amenities or disamenities Gayer (2000) Effect of Superfund hazardous waste sites on house prices in the greater Grand Rapids, MI, area Gayer et al. (2000) Effect of Superfund hazardous waste sites on house prices in the greater Grand Rapids, MI, area Kiel (1995) Effect of Superfund hazardous waste sites on house prices in Woburn, MA Leggett and Bockstael (2000) Effect of fecal coliform bacteria in Chesapeake Bay on the price of waterfront property Palmquist and Danielson (1989) Soil quality and farmland value in North Carolina Smith and Huang (1993) Air quality (metaanalysis of 26 studies) Smith and Huang (1995) Air quality (metaanalysis of 37 studies) Zabel and Kiel (2000) Air quality and house prices in Chicago, Denver, Philadelphia, and Washington, DC, metropolitan areas

287 Table 12-3. Residential gradients from recent studies Author(s) Year Data Metro area Rent gradient1 Highway studies Boarnet and Chalermpong (2001) 1988- 2000 House sale price and distance to nearest highway ramp Orange County, CA -$0.88 to - $4.49/ft. (1982 dollars) Voith (1993) 1970- 1988 House sale price and auto commute time Montgomery County, PA -$955 to -$1,168 (1990 dollars) per minute Langley (1981) 1962- 1978 House sale price index from sale-resale pairs, comparing houses within and outside of 1,125 foot buffer from highway Washington, DC, Metro Area -$3,000 to -$3,500 per house for properties within 1,125 feet of highway Li and Brown (1980) 1971 House sale price and distance to expressway interchange Boston, MA -$1,642 to -$1,815 per house for each doubling of distance Transit studies Haider and Miller (2000) 2000 Prices of houses in and out of 1.5 km distance from subway line Greater Toronto area, Canada C$4,000/house2 Sedway Group (1999) 1999 House price and distance from transit station San Francisco Bay Area, CA (Alameda and Contra Costa Counties) -$3,200 to -$3,700/mile Cambridge Systematics (1998) 1997 House price and distance from transit station San Francisco Bay Area, CA (Urban/CBD properties) -$2.88 to -$69.12/ft 3 Voith (1993) 1970- 1988 House sale price and accessibility to train station Montgomery County, PA $7,279 to $9,605/house 1 Unless otherwise indicated, this is the drop in value for each unit distance from a transportation facility. 2 In this study, the authors separate housing stock into two groups: one includes houses within a 1.5km distance of a subway line and the other includes the rest of the housing stock. All else equal, the authors found that a house within the 1.5 km distance of a subway line sells for C$4,000 (4,000 Canadian dollars) more, on average. 3 The gradients are for single-family housing units in urban areas. The gradient is steep closer to the BART station (within 1,000 ft.) and flat farther from the station (more than 2,000 ft.). In other words, house values drop quickly near station but slowly far away from station. Note, also, that the data used in this study were gathered from properties that are located in urban areas (not suburban locations) and within 2,500 feet of stations only. The magnitude of gradient is much higher than in the Sedway study, where the data are from suburban counties.

288 House price gradients that were obtained from various studies on transportation and residential property values are listed in Table 12-3. We note that some of these studies are quite dated. The various price gradients estimated in these studies corroborate the less specific findings presented in Table 12-2 in that the Orange County and Boston studies indicate a considerable negative price effect of distance from a major highway. The trade-off between the positive influence of shorter commute times versus the negative effect of direct proximity to such a facility is evident in the two studies of the Washington, DC, area. Similar information for commercial property is presented in Table 12-4. In the case of commercial properties, access to major highways and thus customers has a strong positive impact on property values. In some cases, these studies found a greater than linear decline in values with greater distance from an access point. Table 12-5 lists a series of gradients obtained from a recent TCRP report. That study found that direct proximity to a rail transit station within the city has a major positive effect on both single- and multiple-family residential property values. Within the city, impact on per-square-foot rental prices for offices space is nearly twice that on retail space. In suburban areas, retail space benefits much more due to direct proximity to a station. Table 12-4. Gradients for commercial property Author(s) Year Data Metro area Rent gradient Highway studies Kockelman and ten Siethoff (2002) 1982- 1999 Property and land value (10 types of use) and distance to frontage road network Austin, TX -$510,000/ acre/sq.mi.1 Transit Cambridge Systematics (1998) 1998 Rent per sq ft per month of retail and office properties San Francisco Bay Area, CA -$.05/1000 sq ft/mo/mi Sedway Group (1999) 1999 Land price per sq ft for office properties San Francisco Bay Area, CA -$117/sq ft/ mi 1 The gradient in this case is quadratic (dollar value per distance squared), rather than linear (dollar value per distance). In the case of a quadratic gradient, the negative effect of being away from a frontage road increases more rapidly with distance than in the case of a linear gradient. For example, for the gradient of $510,000/acre/sq mi, being 0.1 mile away from the road network will reduce property value by $5,100 per acre. If we double the distance to 0.2 mile, the effect on land value will be quadrupled, to -$20,400 per acre. In the linear case, however, doubling the distance will only double the negative effect on land value.

289 Table 12-5. Property value increases near BART stations (1997 dollars) Land use type Distance from BART station (ft) Central business district/Urban Suburban Single family Per Unit Per Unit 0-500 $48,960 $9,140 500-1,000 $14,400 $7,930 1,000-1,500 $8,640 $3,040 2,000-2,500 $5,760 $5,500 Multifamily Per unit/month Per unit/month 0-1,300 $50,00 $42.30 1,300-2,500 $0.00 $0.00 Offices Per sq ft/month Per sq ft/month 0-1,300 $0.13 $0.00 1,300-2,000 $0.07 $0.28 2,000-2,500 $0.00 $0.00 Retail Per sq ft/month Per sq ft/month 0-500 $0.07 $0.24 500-1,000 $0.00 $0.24 1,000-2,500 $0.00 $0.00 Note: This table summarizes how property values change with proximity to Bay Area Rapid Transit (BART) stations. Source: Cambridge Systematics 1998. RESOURCES 1) Cambridge Systematics, Inc. 1998. Economic Impact Analysis of Transit Investments: Guidebook for Practitioners. Transit Cooperative Research Program Report 35, Transportation Research Board, National Research Council. Available at http://www4.trb.org/trb/crp.nsf/. This comprehensive guidebook describes various technical methods for measuring the economic impacts of transit investments, including changes in adjacent property values. It also includes a summary of research on the increases in property values found near BART stations in the San Francisco Bay Area. 2) Sources of property value information: • DataQuick, Inc. is a provider of real estate and land data. The Web site is http://www.dataquick.com/. • Domain is a look-up Web site that provides sales information by location. Available at http://www.domania.com/.

290 • Web sites in many states provide summaries of home sales data. For example, the Texas A&M Real Estate Center provides data for Texas metropolitan areas from 1979 to present. The Web site is http://recenter.tamu.edu/data/. • Regional consumer price indexes can be obtained from the U.S. Bureau of Labor Statistic’s Web site. Available at http://stats.bls.gov/. REFERENCES Adkins, W. G. 1959. “Land Value Impacts of Expressway in Dallas, Houston, and San Antonio, Texas.” Bulletin 227, pp. 50-65. Washington, DC: Highway Research Board. Alonso, William. 1964. Location and Land Use: Toward a General Theory of Land Rent. Cambridge, MA: Harvard University Press. Alonso, William. 1972. “A Theory of Urban Land Market.” Readings in Urban Economics. Edited by Mathew Edel and Jerome Rothenberg. New York, NY: McMillan Press. Appraisal Institute. 1996. The Appraisal of Real Estate. Chicago, IL: American Institute of Real Estate Appraisers. Boarnet, M. G., and S. Chalermpong. 2001. “New Highways, House Prices, and Urban Development: A Case Study of Toll Roads in Orange County, CA.” Housing Policy Debate Vol. 12, No. 3, pp. 575-606. Fujita, Masahisa. 1989. Urban Economic Theory: Land Use and City Size. Cambridge, England: Cambridge University Press. Gatzlaff, D. H., and M. T. Smith. 1993. “The Impact of the Miami Metrorail on the Values of Residences near Station Locations.” Land Economics, Vol. 69 No. 1, pp. 54-66. Gayer, Ted. 2000. “Neighborhood Demographics and the Distribution of Hazardous Waste Risks: An Instrumental Variables Estimation.” Journal of Regulatory Economics, Vol. 17. No.2, pp. 131-155. Gayer, T., J.T. Hamilton, and W.K. Viscusi. 2000. “Private Values of Risk Tradeoffs at Superfund Sites: Housing Market Evidence on Learning about Risk.” Review of Economics and Statistics, Vol. 82, No. 3, pp. 439-451. Haider, M., and E.J. Miller. 2000. “Effects of Transportation Infrastructure and Locational Elements on Residential Real Estate Values. Application of Spatial Autoregressive Technique.” Transportation Research Record 1722. Washington, DC: Transportation Research Board, National Research Council, pp. 1-8. Huang, W. 1994. “ The Effects of Transportation Infrastructure on Nearby Property Values: A Review of the Literature.” The Institute of Urban and Regional Development Working Paper 620, University of California, Berkeley. Kockelman, K., and B. ten Siethoff. 2002. “Property Values and Highway Expansions: An Investigation of Timing, Size, Location, and Use Effects.” Forthcoming in Transportation Research Record. Washington, DC: Transportation Research Board, National Research Council.

291 Kiel, KA. 1995. “Measuring the Impact of the Discovery and Cleaning of Identified Hazardous Waste Sites on House Values.” Land Economics, Vol. 71, No. 4, pp. 428-435. Langley, J. C. 1981. “Highways and Property Values: The Washington Beltway Revisited.” Transportation Research Record 812. Washington, DC: Transportation Research Board, National Research Council, pp. 16-20. Leggett, C., and N. Bockstael. 2000. “Evidence of the Effects of Water Quality on Residential Land Prices.” Journal of Environmental Economics and Management, Vol. 39, No. 2, pp. 121-145. Li, M., and H.J. Brown. 1980. "Micro-Neighbourhood Externalities and Hedonic Housing Prices." Land Economics, Vol. 56, pp. 125-141. Mohring, Herbert. 1961. “Land Values and the Measurement of Highway Benefits.” Journal of Political Economy, Vol. 79, pp. 236-249. Palmquist, R.B., and L. Danielson. 1989. “A Hedonic Study of the Effects of Erosion Control and Drainage on Farmland Values.” American Journal of Agricultural Economics (February), pp. 55-62. Sedway Group. 1999. “Regional Impact Study.” Commissioned by Bay Area Rapid Transit District (BART), July 1999. San Francisco, CA. (From “Rail Transit and Property Values” in Information Center Briefing, Number 1, March 2001. Available at http://www.apta.com/ research/info/briefings/briefing_1.cfm. Smith, J.J. 2001. “Does Public Transit Raise Site Values Around Its Stops Enough to Pay for Itself (Were the Value Captured)?” Available at http://www.vtpi.org/smith.htm. Smith, V.K., and J.C. Huang. 1993. “Hedonic Models and Air Pollution: Twenty-Five Years and Counting.” Environmental and Resource Economics, Vol. 3, No. 4, pp. 381-394. Smith, V.K., and J.C. Huang. 1995. “Can Markets Value Air Quality? A Meta-Analysis of Hedonic Property Value Models.” Journal of Political Economy, Vol. 103, No. 1, pp. 209- 227. Voith, Richard. 1993. "Changing Capitalization of CBD-Oriented Transportation Systems: Evidence from Philadelphia, 1970-1988." Journal of Urban Economics, Vol. 33, No. 3, pp. 361-376. Zabel, J.E., and K.A. Kiel. 2000. “Estimating the Demand for Air Quality in Four U.S. Cities.” Land Economics, Vol. 76, No. 2, pp. 174-194.

Next: Chapter 13 - Cultural Resources »
Effective Methods for Environmental Justice Assessment Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB’s National Cooperative Highway Research Program (NCHRP) Report 532: Effective Methods for Environmental Justice Assessment is designed to enhance understanding and to facilitate consideration and incorporation of environmental justice into all elements of the transportation planning process, from long-range transportation systems planning through priority programming, project development, and policy decisions. The report offers practitioners an analytical framework to facilitate comprehensive assessments of a proposed transportation project’s impacts on affected populations and communities.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!