Free Response to Motion - District Court of Federal Claims - federal

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 1 of 39

WORKING PAPER SERIES

Airport-Related Noise, Proximity, and Housing Prices in Atlanta

Jeffrey P. Cohen and Cletus C. Coughlin

Working Paper 2005-060A http://research.stlouisfed.org/wp/2005/2005-060.pdf

August 2005

FEDERAL RESERVE BANK OF ST. LOUIS Research Division 411 Locust Street St. Louis, MO 63102
______________________________________________________________________________________ The views expressed are those of the individual authors and do not necessarily reflect official positions of the Federal Reserve Bank of St. Louis, the Federal Reserve System, or the Board of Governors. Federal Reserve Bank of St. Louis Working Papers are preliminary materials circulated to stimulate discussion and critical comment. References in publications to Federal Reserve Bank of St. Louis Working Papers (other than an acknowledgment that the writer has had access to unpublished material) should be cleared with the author or authors. Photo courtesy of The Gateway Arch, St. Louis, MO. www.gatewayarch.com

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 2 of 39

Airport-Related Noise, Proximity, and Housing Prices in Atlanta Jeffrey P. Cohen Assistant Professor of Economics Barney School of Business University of Hartford [email protected] Cletus C. Coughlin Deputy Director of Research and Vice President Research Division Federal Reserve Bank of St. Louis [email protected]

August 2005

Abstract Using hedonic pricing models, we focus on the effects of proximity and noise on housing prices in neighborhoods near Hartsfield-Jackson Atlanta International Airport. We find that proximity to the Atlanta airport is related positively to housing prices and that airport-related noise is associated with lower prices. Relative to this latter result, estimates are generated showing evidence of a shrinking noise discount between 1995 and 2002. An important question is whether this result reflects a reduced effect of a given noise level on house prices or declining noise levels in the areas subject to relatively more noise. One explanation is that soundproofing of houses in noisy areas has increased their values. A lack of data, however, prevents definitive statements concerning the importance of soundproofing. A second explanation is that the trend reflects the limitations of the underlying noise contours. Over time, houses in the neighborhoods near the airport have been subjected to less airport-related noise. Evidence is presented consistent with this explanation.

JEL Codes: Q53, Q51, R31, L93 Keywords: noise, airports, housing prices, hedonic pricing, proximity

The authors thank Deborah Roisman for excellent research assistance and the City of Atlanta Department of Aviation for noise contour data. The views expressed are those of the individual authors and do not necessarily reflect official positions of the Federal Reserve Bank of St. Louis, the Federal Reserve System, or the Board of Governors.

1

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 3 of 39

Introduction Due to the flight paths of arriving and departing aircraft, noise is spatially concentrated in areas near airports. A common way to explore the impact of this geographically-concentrated externality is to estimate its effects on housing prices. Our research focuses on the impact of airport-related noise, as well as proximity, on housing prices during 1995-2002 in neighborhoods near Hartsfield-Jackson Atlanta International Airport, the world's busiest passenger airport.1 We estimate a hedonic pricing model. Noise and proximity are simply two of many attributes that affect the value of a housing unit. Such models have been used for many years in housing price studies, with numerous studies attempting to estimate the effect of amenities as well as disamenities on housing prices.2 There have been several hedonic airport noise studies, but only a small number of recent studies have examined noise effects of U.S. airports on housing prices.3 Coinciding with our focus on the connection between noise and housing prices is the simultaneous consideration of the impact of proximity. Brueckner (2003) found that metropolitan area employment was related to the level of air traffic (i.e., passenger enplanements). Thus, it is reasonable to ask how residents might be affected by closer proximity to those airport-related jobs. Access to airport-related jobs and air

1

The airport was renamed to honor a former Atlanta mayor, Maynard H. Jackson, in October 2003. For the period we examine, the airport was generally identified as simply Hartsfield. 2 For example, see Greenbaum et al. (2005) ­ crime, Anstine (2003) ­ environmental disamenities associated with emissions from manufacturing facilities, and Benson et al. (1998) ­ environmental amenities such as the value of an ocean, lake, or mountain view. 3 The hedonic price method, which relies on revealed preference, is the most common approach for assessing the cost of noise. This cost has been examined to a very limited extent via three other methods-- artificial neural networks, contingent valuation, and happiness surveys. See Collins and Evans (1994), Feitelson et al. (1996), and van Praag and Baarsma (2005), respectively. Similar to hedonic pricing models, the use of artificial neural networks relies on revealed preference. The other methods rely to some degree on subjective, survey-based information.

2

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 4 of 39

transportation services can become capitalized into the value of a house. Ignoring the value of accessibility in the present context could bias the estimates of the impact of noise. In a survey of early work on airport-related noise and housing prices, Nelson (1980) noted that most studies had found a reduction in property values between 0.4 and 1.1 percent per decibel of additional noise. In a more recent survey, Nelson (2004) found a smaller range--0.5 to 0.64-- for the reduction in property values per decibel of additional noise. The results of two recent studies of specific airports are especially relevant to our analysis. McMillen (2004a) found that residential property values for homes within a 65 decibel noise contour band of Chicago's O'Hare Airport were about nine percent lower than otherwise similar homes.4 Similarly, Espey and Lopez (2000) identified a significant decrease in the prices of homes subject to greater noise levels. They found a $2400 difference in the price of a home in Reno-Sparks, Nevada, in areas where the noise level reaches at least 65 decibels. To date only one previous study has examined the effect of noise at the Atlanta airport on property values. O'Byrne, Nelson and Seneca (1985) examined the prices of properties near the Atlanta airport for 1970-72 and 1979-80 using hedonic price models. Noise negatively affected price in both sets of regressions. Moreover, despite using prices based on individual house sales in one period and owner-appraised Census block aggregates in the other period, their results revealed similar estimates of the noise discount for the two periods.

4

Nonetheless, because the noise associated with aircraft is diminishing over time, McMillen (2004a) estimated that residential real estate prices for houses near O'Hare would rise if an additional runway were built.

3

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 5 of 39

Our research will update and improve upon the results of O'Byrne et al. concerning the impact of noise associated with the Atlanta airport. Noise levels and the geographic distribution of noise affecting nearby neighborhoods have changed due to the expansion of the airport, specific attempts by the aircraft authorities to mitigate noise, and the development of quieter aircraft. As we will demonstrate, the changing geographic distribution complicates our attempt to generate insights concerning how noise has affected housing prices. A noteworthy advance over prior studies is that we produce results concerning how the impact of noise on housing prices has changed over time at a specific airport. In an early survey of airport-related noise estimates and housing prices, Nelson (1980) concluded that the impact of noise was relatively stable across studies; however, a recent review by Schipper et al. (1998) of 19 studies found much variation among the estimates produced by the studies. Much of this variation could be explained by differences across studies in terms of either the characteristics of the sample population (e.g., mean house price) or the study (e.g., time period, country, and specification). In an even more recent meta-analysis based on an expanded sample of 33 hedonic estimates of the noise discount for 23 airports in the United States and Canada, Nelson (2004) concluded that variability in estimated noise discounts was related to the country in which the airport was located and to the specification of the hedonic model. In contrast to prior studies, we use price data from one source over a moderately long time period to explore how the impact of noise near the Atlanta airport has changed over time. Based on noise levels for 1995, the magnitude of the noise discount appears to have shrunk over time. This empirical result, however, might be misleading. The

4

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 6 of 39

evidence that we are able to generate suggests that a changing distribution of noise is, at least, partially responsible for this result. In other words, the estimated shrinking noise discount reflects the fact that actual noise levels have declined in some areas near the Atlanta airport rather than that a given noise level is affecting housing prices to a smaller degree. Due to a lack of data, we were unable to assess the impact of soundproofing programs on the noise discount over time. Data and Model The standard categories of explanatory variables used in studies of housing prices are the structural features of the housing units, location characteristics, and attributes of the social and natural environment. To estimate the impact of specific determinants on housing prices, we combined data from various sources. The data consist of noise contour maps for the neighborhoods surrounding the airport, demographic data on a block-group basis for average income and percent of housing occupied by blacks, as well as data for single family house sales, including sales prices, the geographic location, and housing characteristics. Noise contour files for 1995 and 2003 have been obtained from the City of Atlanta Department of Aviation, and they are in a format that enabled them to be read into ArcView Geographic Information Systems (GIS) software.5 ArcView GIS software enables the user to read noise contour maps and match up the corresponding noise data with other geo-coded data corresponding to individual properties, such as distance to the airport and other demographic data. The noise contour maps are based on a standard measure of noise used by the Federal Aviation Administration and other federal agencies.
5

Nelson (2004) refers to GIS studies as the "next generation of hedonic studies", with the first generation having looked at Census tract data, and the second generation focusing on individual sale price data. For an overview of applying GIS to various economic issues, see Bateman et al. (2002).

5

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 7 of 39

This measure, the yearly day-night sound level (DNL), is measured in decibels. Because an increase of 10 decibels is equivalent to a ten-fold increase in sound, a ten-unit increase in the DNL can be viewed similarly. Nelson (2004) notes that normal background noise levels in urban areas are approximately 50-60 decibels during daytime hours and 40 decibels during nighttime. A DNL of 65 decibels is the Federal Aviation Administration's lower limit for defining a significant noise impact on people.6 At 65 decibels and above, individuals experience the disruption of normal activities, such as speaking, listening, learning, and sleeping. A DNL of 75 decibels or more is viewed as incompatible with single family housing.7 Our analysis uses two noise contours, one for 65 decibels and one for 70 decibels. Single family dwelling sale price data for the years 1995 through 2002 have been purchased from First American Real Estate Services for the Atlanta neighborhoods that fall in the 65 DNL and 70 DNL boundaries, as well as within a half mile outside of the 65 DNL boundary, which is termed the "buffer zone."8 9 Relative to the buffer zone, we expect to find that, all else equal, houses subjected to more noise will sell for lower prices.

6

Research on noise suggests that roughly 12 percent of people subjected to a DNL of 65 decibels report that they are "highly annoyed" by transportation noise. Annoyance is defined as the adverse psychological response to noise. Meanwhile, about 3 percent are highly annoyed when subjected to a DNL of 55 decibels and nearly 40 percent are highly annoyed at a DNL of 75 decibels. See Federal Register (2000). 7 Nelson (2004) notes that since 1979 federal agencies have regarded land subjected to DNLs ranging from 65 to 74 decibels as "normally" incompatible with residential use, while land exposed to a DNL of less than 65 decibels is regarded as "normally" compatible with residential land use. 8 There were 28 sales in the 75 decibel noise contour that we have chosen to eliminate from the sample. 9 There are 3 reasons why we chose the 0.5 mile buffer zone as opposed to looking at the entire Atlanta metropolitan area. First, other airport noise studies have handled analogous problems in a similar manner. Second, we wanted to examine the impact of additional noise relative to some base that exhibits some noise. Third, other parts of Atlanta may face dramatically different housing price determinants than the area surrounding the airport.

6

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 8 of 39

The geographic area that we examine and the relevant noise contours for 1995 are shown in Figures 1 and 2. The airport lies ten miles south of downtown Atlanta. The area under consideration covers parts of three counties--Fulton, Clayton, and DeKalb. The number of home sales total 2,370; however, none of these sales occurred in DeKalb County. Housing sale prices are deflated by the National Association of Realtors median housing price index for Atlanta, with 1995 median sales price for Atlanta as the base year. Between 1995 and 2002 this index increased 50 percent, substantially larger than the 20 percent increase in the consumer price index. In addition to the sales price, housing characteristics such as the numbers of stories, bedrooms, baths, fireplaces, lot size, and the age of the dwelling were contained in the purchased dataset. Table 1 lists

all the variables used in our analysis and how these variables were measured, while Table 2 contains summary information of our data.10 We expect each of the housing characteristics variables, with the exception of the age of the dwelling, to be related positively to housing prices. An increase in the age of a house, holding all other things constant, should tend to reduce its price. The demographic data is from the Bureau of the Census. Specifically, the block group data for the years 1995-1999 came from the 1990 Summary Tape File 3 ­ Sample data, and the 2000-2002 block group data came from the 2000 Summary Tape File 3 ­ Sample data. With respect to the neighborhood characteristics, we expect that the sign on the median income coefficient should be positive due to neighborhood effects. In other

10

Following Pennington et al. (1990) and Espey and Lopez (2000), dummy variables are used for measuring selected structural housing characteristics, such as the numbers of bedrooms, bathrooms, and fireplaces.

7

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 9 of 39

words, houses in neighborhoods where residents have higher incomes should result in higher housing prices. Despite some research finding a negative relationship, such as O'Byrne et al. (1985), our expectation concerning the sign of the coefficient relating the percentage of housing occupied by blacks to housing prices is somewhat tenuous.11 Possible changes in racial attitudes as well as differences in the composition of the neighborhoods we examine versus O'Byrne et al. (1985) preclude a firm prior concerning this relationship. In addition to noise and neighborhood characteristics, the location of a house is likely to affect its price via some other characteristics. Whether the house is located in Fulton County or Clayton County is potentially important. The benefits and costs of services provided at the county level can differ across counties and can affect housing prices. A priori, we have no expectation as to whether this public sector variable should be relatively more favorable to housing prices in Fulton County or Clayton County. Finally, we use ArcView to calculate another potentially important location characteristic -- the distance between each property address and the airport. As found in some of the previous airport noise studies, after accounting for airport noise, we expect that less distance from the airport should result in higher housing prices, due to more convenient access to jobs at the airport and air transportation service.12 13 Ignoring the

11

Greenbaum et al. (2005) found a positive relationship between housing prices and the percentage of white population for neighborhoods in Columbus, Ohio. 12 Proximity to an airport may have positive (due to accessibility) as well as negative (due to noise) effects on property values. In a study of an airport in Manchester, England, Tomkins et al. (1998) found that the benefits of easy access to the airport outweighed the costs of living near the airport. In contrast, Espey and Lopez (2000) found proximity to an airport in Reno, Nevada to be a disamenity. 13 Numerous studies have related property values to the distance from rail stations. Bowes and Ihlanfeldt (2001, p. 3) conclude that "the majority do find that rail stations have a positive (but relatively modest) impact on nearby property values." In the case of rail stations, Bowes and Ihlanfeldt examine four factors related to proximity that might affect property values. The positive factors are the transportation access

8

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 10 of 39

value of accessibility to the airport could bias the estimated noise effect downward (i.e., to showing a lessened effect). Thus, the model is a hedonic housing price model, with the individual housing characteristics, noise exposure, neighborhood characteristics, the county in which the house is located, and distance to the airport as the explanatory variables. Results ­ A First Cut Using 1995 Noise Contours We estimated numerous hedonic housing price models. Lacking a priori knowledge of the appropriate functional form, we estimated three different specifications--a double log form, a semi-log form, and a linear form.14 Our empirical results can be separated into two parts--one part consists of results of models without noise time trends and the other part with noise time trends. We begin by examining the results of models without a noise time trend. Table 3 contains the results of models using the noise contours for 1995. In terms of the variation in housing prices, the estimated models explain from 44 percent of the variation in the case of the double log form to 37 percent for the linear form. The results indicate that nearly every individual variable performed as expected. We begin by examining the results for the variables not related to the Atlanta airport. Regardless of functional form, variables measuring the structural characteristics of houses exhibited the expected impact on housing prices and, virtually without exception, were statistically significant, often at the 1 percent level. For example, the dummy

advantage provided by rail stations and the increased retail activity stemming from rail stations. The negative factors are crime and environmental disamenities, such as noise, pollution, and the unsightliness of rail stations. In their study of Atlanta's rail system, all four factors affected property values, with the relative importance varying by distance from downtown and the median income of the neighborhood. 14 As Gayer et al. (2002) note, Box-Cox transformations are also frequently used in hedonic pricing studies.

9

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 11 of 39

variable differentiating houses with two or more stories from other houses, dStories, was found to be a positive, statistically significant determinant of housing prices. The dummy variables differentiating houses based on the number of bedrooms, beds_3, beds_4, and beds_5, were all related positively to housing prices and, except for beds_3 in the linear form, were statistically significant. Moreover, the size of the estimated coefficients increased as the number of bedrooms increased, which is what one would expect. The results for the number of bathrooms are similar to the results for the number of bedrooms. The dummy variables differentiating houses based on the number of bathrooms, baths_2 and baths_3, were all related positively to housing prices, increased in size with the number of bathrooms, and were statistically significant. The dummy variable differentiating houses with two or more fireplaces from other houses, dFire_2, was related positively to housing prices and was statistically significant. Indicating that newer houses tend to sell for higher prices than older houses, the variable measuring the age of the house, log_age in the double log specification and age in the other specifications, was related negatively to housing prices; however, this variable was not statistically significant for the semi-log and linear specifications. Finally, indicating that larger lots were associated with higher housing prices, the variable measuring lot size, log_acres in the double log specification and acres in the other specifications, was related positively to housing prices and was statistically significant. Turning to the neighborhood characteristics, the variable measuring the median household income in the neighborhood in which a house was sold, log_medhhinc in the double log specification and med_hhinc in the other specifications, was related positively to housing prices and was statistically significant. Meanwhile, the percentage of homes

10

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 12 of 39

in a neighborhood occupied by blacks, perc_tenureblack, was related positively to housing prices and, except for the semi-log specification, was statistically significant. This positive relationship is contrary to the findings of O'Byrne et al. (1985). Whether a house is located in Fulton County rather than Clayton County appears to affect its price. The dummy variable measuring county location, fulton, was related positively to housing prices and was statistically significant. Identifying the precise reasons for this result is beyond the scope of our analysis; however, two facts about publicly-provided services suggest reasons for this result. First, at least in recent years, public elementary and secondary schools in Fulton County tended to outperform similar schools in Clayton County. For example, the Fulton County system met the 2004-2005 Adequate Yearly Progress standard, one of the cornerstones of the federal No Child Left Behind Act. Meanwhile, the Clayton County system did not meet this standard.15 Second, the Center for Digital Government, a national research and advisory institute on information technology policies and best practices in local and state government, has identified Fulton County as among the top ten leading counties in its use of digital technology.16 Meanwhile, Clayton County was not mentioned. The remaining results concern the variables related to the airport--distance and noise. The distance from a house to the airport affects the price of the house. The variables measuring that distance, log_distance in the double log specification and distance in the other specifications, were related negatively to housing prices and were statistically significant. Our results are consistent with findings by Tomkins et al. (1998) that proximity to the airport in Manchester, England, and by McMillen (2004a) that

15 16

See www.gadoe.org. See www.centerdigitalgov.com.

11

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 13 of 39

proximity to Chicago's O'Hare Airport were amenities, but are inconsistent with a finding by Espey and Lopez (2000) that proximity to the airport in Reno, Nevada, was a disamenity. Turning to the results with respect to airport noise, the results are consistent with expectations, but are far from overwhelming. Regardless of the functional form, airport noise is related negatively to housing prices and the relative magnitudes of the coefficients for db65_95 and db70_95 are as expected with a larger noise discount for the noisier location. However, the relationship is statistically significant for only three of the six possibilities.17 As noted above, we include the distance variable in our final specification because the parameter estimate was statistically significant. Also, we find that excluding proximity tends to shrink the estimated impact of noise on housing prices. For example, based on the model in Table 3, the parameter estimates for the 65 decibel noise contour for the log, semi-log, and linear functional forms when proximity is excluded (included) are: -0.008 (-0.016), -0.011 (-0.023), and -2,741.55 (-3,649.05). The parameter estimates for the 70 decibel noise contour are: -0.024 (-0.060), -0.010 (-0.044), and -1,410.67 (-4,041.68). In addition, excluding proximity tends to reduce the statistical significance of noise for housing prices. These results are similar for the remaining models in this paper, so we have chosen not to repeat this consistent finding when we discuss these other models. Potential Biases Stemming from Changing Noise Contours

17

Note that our assessment of statistical significance is based on a two-sided hypothesis test. In the present case, in light of the theoretical relationship of a negative relationship between noise and the price of a house, a one-sided test can be justified. In that case, four of the six possibilities show statistical significance, with the other two being very close to significance at the ten percent level.

12

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 14 of 39

One question that deserves scrutiny is whether there are potential biases from the use of the 1995 contours for the entire period. A comparison of the 1995 contours with contours estimated for 2003 suggest that the noise contours have changed systematically between 1995 and 2003. Figure 3 shows how the 65 decibel noise contour has changed between 1995 and 2003. Generally speaking, the 65 decibel noise contour has shifted closer to the ends of the runways. This change and a similar change for the 70 decibel noise contour have resulted from a combination of technological changes, regulatory policy, and airport authority efforts to reduce the effects of noise.18 Prior to 1968 aircraft noise was not regulated on a national basis in the United States. Following Congressional authorization, the Federal Aviation Administration promulgated standards requiring that the best available noise reduction technology be used in new designs of civil subsonic turbojet aircraft. Aircraft satisfying this standard were categorized as Stage 2 aircraft, while those not meeting this standard were classified as Stage 1 aircraft. In 1977, the Federal Aviation Administration adopted more stringent noise standards that applied to all newly manufactured aircraft. Aircraft meeting the new standards, which only apply to commercial subsonic jet aircraft over 75,000 pounds, were classified as Stage 3 aircraft. Following passage of the Airport Noise and Capacity Act in 1990, all commercial jets operating in the United States were required to be Stage 3 compliant by December 31, 1999. Thus, one implication to be drawn from phasing-in the Stage 3 requirement is that houses initially identified as located in the 65 decibel 1995 noise contour might well be in the buffer zone near the end of the time period we examine.
Additional information on this topic can be found on the airport's website: www.atlanta-airport.com. After reaching the site, navigate through "Airport Information," "Environmental," and then "Noise and Operations Monitoring System" (NOMS).
18

13

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 15 of 39

The preceding discussion indicates that both the 70 decibel and 65 decibel noise contours shrink over time so that some houses become subject to less noise. Meanwhile, the buffer zone becomes a relatively larger share of the area under consideration and it too is becoming subjected to less noise. Thus, on average, the houses in each of these areas should tend to increase in value. Because the estimation of the noise discount is relative to the buffer zone, it is unclear whether the estimated noise discount using the 1995 contours for 1995-2002 will tend to be biased high, biased low, or even unbiased if the effects on price are identical in each area. The existence of the 2003 contours provides two other ways to reconfigure the dataset for estimation purposes. One approach is to use the 2003 contours for the entire period. Because the 70 and 65 decibel noise contours shrink between 1995 and 2003, the number of house sales in our dataset in these two zones also declines. The number of house sales in the 70 decibel noise contour declines from 249 to 67, while the decline in the 65 decibel noise contour is from 1,113 to 748. Because the overall geographic area is unchanged, the number of house sales in the buffer zone increases from 1,008 to 1,555. What can we say about the noise that houses in the three areas are subject to over time? First, the houses in the 70 decibel noise contour are subject to roughly the same noise level in 1995 as they were in 2002. Second, the houses in the 65 decibel noise contour are subject to at least the same noise level, with some houses subjected to more at some time during the period. The reason for this observation is that some houses that began the period in the 70 decibel noise contour are identified in the 65 decibel noise contour by 2003. Third, the houses in the buffer zone based on the 2003 contours are subject to conflicting effects. The houses that were also in the buffer zone based on the

14

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 16 of 39

1995 contours are subjected to less noise over time. However, some houses that were subjected to noise levels of 65 decibels or higher early in the period are viewed as being subjected to less noise because they are in the buffer zone based on the 2003 definition. It is unclear how these two effects affect the buffer zone and, consequently, whether and how the 70 decibel and 65 decibel noise discount estimates are biased. A second way to use the 2003 noise contours is to combine them with the 1995 noise contours. The resulting sample consists of houses in the 70 decibel contour using both the 1995 and 2003 noise contours, houses in the 65 decibel contour for both noise contours, and houses in the buffer zone for both noise contours. The dummy variables identifying houses in the 65 and 70 decibel noise contours using 1995 and 2003 measures are db6565 and db7070, respectively. Houses classified as switching noise areas between 1995 and 2003 were deleted.19 As a result, the sample size decreased from 2370 to 1643. In addition, as indicated in Table 4, the distribution of houses across the three noise areas changed. The location of houses in this sample is shown in Figure 4. The percentage of houses in the buffer zone increased by nearly 19 percentage points, while the percentage of houses in the 65 and 70 decibel noise contours declined by roughly 12 and 6 percentage points, respectively. Some minor changes can also be observed in the means of the variables. Not surprisingly, with the relative decline of the houses in the 65 and 70 noise contours, the mean of the adjusted sales price rose as did the distance from the airport.

19

McMillen (2004b) follows another approach to analyze whether homes have risen faster in neighborhoods where noise exposure has changed over time. He assigns dummy variables to properties that switched sides of the 65db noise contour over time, and for the most part finds that on average the houses that switch sides of the noise contour over time do not significantly appreciate.

15

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 17 of 39

Neighborhood characteristics also changed. The percentage of houses in the neighborhood occupied by blacks fell, while the median household income rose. What can we say about potential bias using this sample? Using this reduced sample, it is clear that houses in the 70 decibel and 65 decibel noise contours are subject to roughly the same noise levels throughout the period. (More precisely, the houses in the 70 decibel noise contour are subject to at least 70 decibels of noise throughout the period, with some houses subjected to slightly less noise at the end of the period than at the beginning.) Meanwhile, houses in the buffer zone are likely to be subject to less noise over time. Because the houses in the buffer zone are expected to be affected only minimally by noise, an argument can be made that any reduction in noise for these houses is likely to have a negligible effect on their price. This is especially pertinent due to the construction of this sample. Thus, relative to the buffer zone, the noise discount estimates, to the extent that any bias exists, should be biased high. Results Incorporating 2003 Noise Contours First, we focus on the results based on the two samples that use the 2003 noise contours. Table 5 contains the results using the entire dataset with the 2003 noise contours defining the noise levels for the entire sample period. Table 6 contains the results using the sample restricted to housing sales that occur in the same noise contour using both the 1995 and 2003 noise contours. In light of our focus on the noise discount estimates, we focus our discussion on the variables related to the airport, especially the noise discount estimates.20

20

Generally speaking, the results in Tables 3, 5, and 6 for the "non-noise" variables are very similar. The key difference in terms of statistical significance between the results in Tables 3 and 5 is that perc_tenureblack is not statistically significant in the linear form in Table 5, while it is statistically significant in Table 3. The key differences between the results in Table 6 relative to Tables 3 and 5 is that

16

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 18 of 39

The results in Tables 5 and 6 continue to indicate that the distance from a house to the airport affects the price of a house. The variables measuring distance, log_distance and distance, are related negatively to housing prices and are statistically significant. Comparing the parameter estimates, one sees that in Table 6 the marginal value of proximity is much larger than in Tables 3 and 5. The parameter estimates for the noise variables in Table 5 are, as expected, negative. Moreover, the magnitudes of the estimates for the 70 decibel noise contour exceed those for the 65 decibel noise contour. However, in only two of the six cases is the relationship between noise and housing price statistically significant. In Table 6, there are stronger results. Not only are the parameter estimates negatively signed and of the appropriate relative magnitudes, but in five of the six cases the relationship is statistically significant. The estimates for each functional form and for both noise levels are larger (in absolute value) in Table 6 than in Tables 3 and 5. Recall from the prior discussion that a case can be made that the estimates in Table 6 are biased high. Thus, these estimates provide an upper bound on the noise discount. The coefficient for the 65 decibel noise contour in the double log specification has a value of -0.025, which implies that after accounting for other physical and neighborhood characteristics, houses in this noise contour sold for about 2.5 percent less on average than houses in the buffer zone.21 Meanwhile, the coefficient for the 70 decibel noise contour suggests a noise discount of 8.6 percent.

fulton is no longer statistically significant and beds_5 is no longer statistically significant in the double log and semi-log forms. 21 The estimated coefficients for the noise contours using the double log and semi-log functional forms are approximately equal to the noise discount. In percentage terms, the noise discount equals (e ­ 1) x 100, price.

17

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 19 of 39

For the semi-log specification, the noise variables are statistically significant with the coefficients for the 65 and 70 decibel noise contour zones equal to -0.038 and -0.099, respectively. Thus, the noise discount for houses in the 70 decibel noise contour, 9.4 percent, is more than double that of houses in the 65 decibel noise contour. In contrast to our noise discount finding of 3.7 percent, Espey and Lopez (2000) estimated a noise discount of 2 percent for houses in the 65 decibel noise contour for the Reno airport, while McMillen (2004a) estimated a 9 percent noise discount for houses in the 65 decibel noise contour for Chicago's O'Hare Airport. In addition to examining different geographic areas, different time periods of other studies may account for some of these differences between their findings and our results. Finally, for the linear specification, the results presented in Table 6 indicate noise discounts for the 65 and 70 decibel noise contours of $4,725 and $12,491, respectively. With the average adjusted price of a home in the buffer zone equal to $74,414, the sales prices of homes in the 65 and 70 decibel noise contour bands are roughly 6.3 and 16.8 percent, respectively, lower than in the buffer zone.22 In contrast to our finding of 6.3 percent, Espey and Lopez (2000) estimated a noise discount of slightly more than 2 percent for houses in the 65 decibel noise contour for the Reno airport. Results - A Changing Noise Discount? The next question that we explore is whether the noise discount has changed over time.23 The preceding discussion suggests that our empirical finding of a changing noise discount might be due to the changing noise patterns surrounding the airport. However,

22

Consistent with a general finding by Schipper et al. (1998), the estimated impact of noise on housing prices tends to be larger for the linear specification relative to double log and semi-log specifications. 23 We also examined time trends for the other independent variables. Our results revealed no linear time trend between these variables and housing sales prices.

18

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 20 of 39

this factor is not really causing a change in the noise discount, but rather reflects the impact of less noise. There are reasons for why the actual noise discount might change over time. Moreover, it is not clear whether this noise discount is more likely to increase or decrease over time. With increasing incomes and with noise being a disamenity, then it is likely that the noise discount should tend to increase over time. On the other hand, airport authorities were pursuing policies during the time period under consideration to reduce the effects of noise.24 For example, houses within the noise contours were soundproofed. Such efforts should make the purchase of these houses more desirable and, consequently, reduce the noise discount. Tables 7 through 9 contain the results of adding a time trend for the 65 decibel noise contour to the models estimated previously.25 Adding the time trends only affected the results concerning the noise contours. The overall explanatory power of the models remained unchanged and the other individual variables were affected only minimally. Turning to the noise-related results, the statistical evidence suggests that if the measured noise discount within the 65 decibel noise contour has changed at all, it has shrunk over time. In Table 7, which uses the 1995 noise contours throughout the sample period, for the double log specification, the estimated discount in 1995 was 6.2 percent. In other words, houses in the 65 decibel noise contour sold for 6.2 percent less than otherwise comparable houses in the buffer zone. This discount tended to shrink at a rate
24

According to the authorities at the Atlanta Airport (http://www.atlantaairport.com/sublevels/airport_info/noise.htm): "The DOA's Noise Mitigation Program's purpose is to be proactive in addressing eligible noise-impacted properties inside the approved noise contours. This program provides assistance to the communities surrounding HJAIA by continuing to reduce noisesensitive uses inside the 70 DNL through acquisition/relocation, and complete acoustical treatment for the remaining noise-sensitive uses inside the 65 DNL, thereby enhancing their living conditions." 25 We also ran regressions with a time trend for the 70 decibel noise contour. With the exception of the sample using the 1995 noise contours, these models performed poorly. The time trend for the 70 decibel noise contour exhibited no consistent sign and was not statistically significant. In addition, adding this time trend resulted in the 70 decibel noise dummy to become insignificant.

19

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 21 of 39

of roughly 1.2 percentage points per year.26 As a result, our estimates suggest that the noise discount had been eliminated by 2002. A similar result was found using the semilog functional form. For the linear functional form, the time trend in the 65 decibel contour, while tending to shrink the noise discount, was not statistically significant. The results contained in Table 8, which use the 2003 noise contours for the entire 1995-2002 sample period, also provide evidence of a shrinking noise discount. For each functional form the time trend is positive and, in each case, is also statistically significant. Similar to the results in Table 7, the estimates suggest that the noise discount was eliminated by the end of the sample period. Finally, the results contained in Table 9, which use only the houses that are in the same noise contour in both the 1995 and 2003 noise contours, provide, at best, only weak support for a shrinking noise discount. The time trend, albeit positive, does not indicate a statistically significant relationship between the trend of the noise discount and housing prices. The tentative results in Tables 7 through 9 suggest that the noise discount might have shrunk over time. Two explanations can justify our statistical results. First, the soundproofing efforts of the Atlanta Airport Authority might have reduced the irritation associated with airport noise and increased the attractiveness of the properties in the noise contours relative to those in the buffer zone. If so, then the estimated trend is a real phenomenon because the discount shrinks for houses in the 65 decibel noise contour. Second, we cannot rule out measurement error as a cause of our results. The finding of a shrinking noise discount for certain houses is due to the fact that these houses

26

Note that the parameter estimates for 1995 indicate that the noise discount was larger for the 65 decibel noise contour than for the 70 decibel noise contour. This is likely an artifact of a linear time trend. In this case after one year the noise discount for the 65 decibel noise contour is estimated to be less than the noise discount for the 70 decibel noise contour.

20

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 22 of 39

are subject to less noise at the end of the estimation period relative to the beginning of the period. Because houses are assets, the prices of houses reflect various expectations of future conditions, one of which is likely to be the physical environment.27 It is reasonable to think that those involved in the housing market were aware, at least to some extent, of the changing noise patterns surrounding the airport and that this information affected housing prices. A case can be made that the results associated with the reduced sample are less likely to be afflicted with measurement error. If so, then the results in Table 9 might be preferred in terms of identifying the real noise discount. Most likely measurement error has contributed to the finding of shrinking noise discount. While it remains possible that the real noise discount has shrunk, the results in Table 9 do not indicate a statistically significant trend. Conclusion Our research produces results on two dimensions of housing prices near the Atlanta airport. Our findings with respect to proximity can be stated succinctly. Proximity to the airport is related positively to housing prices. Moreover, when this variable is excluded, the estimated impact of noise on housing prices is much less in absolute value than when this variable is included. Thus, our evidence suggests that ignoring the value of accessibility to the Atlanta airport likely biases the estimated noise effect to showing a lessened impact. Consistent with previous airport noise studies, including the prior study of the Atlanta airport by O'Byrne et al. (1985), we find that airport-related noise having a "significant noise impact on people" is associated with lower housing prices. Our point
27

For example, Gayer et al. (2002) found that housing prices responded to environmental risk information.

21

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 23 of 39

estimate of the noise discount is sensitive to both the functional form and the noise contours used to classify houses. For example, using the double-log specification, the noise discount for houses in the 70 decibel noise contour ranges from 5.8 percent to 8.6 percent. Generally speaking, the linear specification yields the largest estimate of the noise discount. Using the 2003 noise contours for the entire sample period, the noise discount for houses in the 70 decibel noise contour ranges from 5.4 percent to 10.6 percent. As expected, using the restricted sample yields the largest noise discounts, with the noise discount ranging from 8.6 percent to 16.8 percent. Turning to the results for houses in the 65 decibel noise contour, houses sell for, at most, 6.3 percent less than comparable houses in the buffer zone. Our estimates for the discount for houses in the 65 decibel noise contour tend to fall between the recent estimates of Espey and Lopez (2000) for the Reno airport and of McMillen (2004a) for Chicago's O'Hare Airport. A novel feature of our research was our attempt to identify whether and how this noise discount has changed over time. Our analysis produced results that warrant additional scrutiny. Using the noise levels for either 1995 or 2003, the noise discount for the 65 decibel noise contour is estimated to have shrunk to virtually zero between 1995 and 2002. However, using what we term the reduced sample, such a statement for the noise discount is much less tenable. Two explanations, which are not mutually exclusive, could account for this time trend. One is that the soundproofing of houses in noisy areas could increase the attractiveness of purchasing houses in these areas relative to less noisy areas. Thus, the noise discount would shrink over time. The second explanation is that the estimated time

22

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 24 of 39

trend reflects the difficulty of measuring noisy areas accurately. It appears that over time the noisy geographic areas have tended to shrink. Thus, some houses have effectively switched from noisy to less noisy areas. Consequently, while the impact of noise on these houses has been reduced, the impact of a given noise level on housing prices has not necessarily changed. We lack sufficient data to rule out the impact of soundproofing. This is an issue that warrants additional research. What appears to be clear, however, is that changing noise contours do play a role in our estimated results. More generally, changing noise contours might well affect the estimates of airport noise studies using data for multiple years.

23

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 25 of 39

REFERENCES Anstine, Jeff (2003) "Property Values in a Low Populated Area when Dual Noxious Facilities are Present," Growth and Change 34(3), 345-358. Bateman, I.J., A.P. Jones, A.A. Lovett, I.R. Lake, and B.H. Day (2002) "Applying Geographical Information Systems (GIS) to Environmental and Resource Economics," Environmental and Resource Economics 22 (1-2), 219-269. Benson, E., J. Hansen, A. Schwartz, Jr., and G. Smersh (1998) "Pricing Residential Amenities: The Value of a View," Journal of Real Estate Finance and Economics 16(1), 55-73. Bowes, David R. and Keith R. Ihlanfeldt (2001) "Identifying the Impacts of Rail Transit Stations on Residential Property Values," Journal of Urban Economics 50(1), 1-25. Brueckner, Jan K. (2003) "Airline Traffic and Urban Economic Development," Urban Studies 40(8), 1455-1469. Collins, Alan and Alec Evans (1994) "Aircraft Noise and Residential Property Values: An Artificial Neural Network Approach," Journal of Transport Economics and Policy 28(2), 175-197. Espey, Molly and Hilary Lopez (2000) "The Impact of Airport Noise and Proximity on Residential Property Values," Growth and Change 31(3), 408-419. Federal Register (July 14, 2000), 65(136), 43802-43824. Feitelson, Eran I., Robert E. Hurd, and Richard R. Mudge (1996) "The Impact of Airport Noise on Willingness to Pay for Residences," Transportation Research: Part D: Transport and Environment 1(1), 1-14. Gayer, Ted, James T. Hamilton, and W. Kip Viscusi (2002) "The Market Value of Reducing Cancer Risk: Hedonic Housing Prices with Changing Information," Southern Economic Journal 69(2), 266-289. Greenbaum, Robert T., Tricia L. Petras, and George E. Tita (2005) "Do Changes in Crime Rates Affect Housing Prices? A Neighborhood Level Analysis," paper presented at the American Real Estate and Urban Economics Association Annual Meetings, Philadelphia, PA, January 7-9. McMillen, Daniel P. (2004a) "Airport Expansions and Property Values: The Case of Chicago O'Hare Airport," Journal of Urban Economics 55(3), 627-640. ___________ (2004b) "House Prices and the Proposed Expansion of Chicago's O'Hare Airport," Federal Reserve Bank of Chicago Economic Perspectives 28(3), 28-39.

24

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 26 of 39

Nelson, Jon P. (1980) "Airports and Property Values: A Survey of Recent Evidence," Journal of Transport Economics and Policy 14(1), 37-52. ___________ (2004) "Meta-Analysis of Airport Noise and Hedonic Property Values: Problems and Prospects," Journal of Transport Economics and Policy 38(1), 1-28. O'Byrne, Patricia Habuda, Jon P. Nelson, and Joseph J. Seneca (1985) "Housing Values, Census Estimates, Disequilibrium, and the Environmental Cost of Airport Noise: a Case Study of Atlanta," Journal of Environmental Economics and Management 12(2), 169178. Pennington, G., N. Topham, and R. Ward (1990) "Aircraft Noise and Residential Property Values Adjacent to Manchester International Airport," Journal of Transport Economics and Policy 24(1), 49-59. Schipper, Youdi, Peter Nijkamp, and Piet Rietveld (1998) "Why do aircraft noise value estimates differ? A meta-analysis," Journal of Air Transport Management 4(2), 117-124. Tomkins, J., N. Topham, J. Twomey and Robert Ward (1998) "Noise versus Access: The Impact of an Airport in an Urban Property Market," Urban Studies 35(2), 243-258. Van Praag, Bernard M.S. and Barbara E. Baarsma (2005) "Using Happiness Surveys to Value Intangibles: The Case of Airport Noise," Economic Journal 115(500), 224-246.

25

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 27 of 39

Table 1 Variables in Hedonic Regressions Name adj_price db65_95 (db65_03) db70_95 (db70_03) time_65_95 (time_65_03) dStories beds_3 beds_4 beds_5 baths_2 baths_3 dFire_2 age* acres* distance* perc_tenureblack

Definition

House sale price deflated by median housing price index for Atlanta. Dummy variable equal to one for houses within the 65 decibel day-night sound level 1995 (2003) noise contour; zero otherwise. Dummy variable equal to one for houses within the 70 decibel day-night sound level 1995 (2003) noise contour; zero otherwise. Interaction term: db65_95 (db95_03) x time, with db65_95 (db95_03) as defined above and with time = 0 for 1995 sales, time = 1 for 1996 sales, ..., time = 7 for 2002 sales. Dummy variable equal to one for houses with more than one story; zero otherwise. Dummy variable equal to one for houses with three bedrooms; zero otherwise. Dummy variable equal to one for houses with four bedrooms; zero otherwise. Dummy variable equal to one for houses with five or more bedrooms; zero otherwise. Dummy variable equal to one for houses with two bathrooms; zero otherwise. Dummy variable equal to one for houses with three or more bathrooms; zero otherwise. Dummy variable equal to one for house with two or more fireplaces; zero otherwise. Age of house in years at the time of its sale. Lot size in acres.

Distance in miles from house to airport. Percentage of homes in the neighborhood in which a house was sold occupied by blacks; only two years of data are available: 1990 Census data is used for sales in 1995-1999 and 2000 Census data is used for sales in 2000-2002. med_hhinc* Median household income in the neighborhood in which a house was sold; only two years of data are available: 1990 Census data is used for sales in 1995-1999 and 2000 Census data is used for sales in 2000-2002. fulton Dummy variable equal to one for houses located in Fulton County; zero for houses located in Clayton County. db6565 Dummy variable equal to one for houses with the 65 decibel day-night sound level noise contour for both 1995 and 2003; zero otherwise. db7070 Dummy variable equal to one for houses with the 70 decibel day-night sound level noise contour for both 1995 and 2003; zero otherwise. time_6565 Interaction term: db6565 x time, with db6565 as defined above and with time = 0 for 1995 sales, time = 1 for 1996 sales, ..., time = 7 for 2002 sales. * These variables are also expressed in natural logarithmic form, which is indicated by "log_" prefix.

26

Case 1:01-cv-00201-VJW Document 219-10 Filed 10/05/2006 Table 2: Summary Statistics, Full Sample -- 2370 Observations Count Frequency House Sales in the buffer zone -- 1995 contours 1008 42.53 House Sales in 65 db zone -- 1995 contours 1113 46.96 House Sales in 70 db zone -- 1995 contours 249 10.51 House Sales in the buffer zone -- 2003 contours 1555 65.61 House Sales in 65 db zone -- 2003 contours 748 31.56 House Sales in 70 db zone -- 2003 contours 67 2.83
House Sales in Fulton County House Sales in Clayton County 1 story 2 or more stories 3 bedrooms 4 bedrooms 5+ bedrooms 1 bathroom 2 bathrooms 3+ bathrooms 0 or 1 fireplace 2+ fireplaces 1117 1253 2143 216 1589 261 52 1055 950 365 2297 73 47.13 52.87 90.42 9.11 67.05 11.01 2.19 44.51 40.08 15.40 96.92 3.08

Page 28 of 39

dist age acres adj_price perc_tenureblack med_hhinc

Mean 3.61 37.85 0.40 $71,692.37 46.72 $29,828.44

Std. Dev. 1.00 16.04 0.37 $37,785.67 30.66 $ 8,593.14

27

Table 3: Models Without Noise Time Trend -- Full Sample -- 1995 Contours Log db65_95 db70_95 log_distance distance dStories beds_3 beds_4 beds_5 baths_2 baths_3 dFire_2 log_age age log_acres acres log_med_hhinc med_hhinc perc_tenureblack fulton Constant Observations R-squared 0.001** [2.39] 0.075*** [2.77] 9.328*** [29.37] 2370 0.44 Robust t statistics in brackets * significant at 10%; ** significant at 5%; *** significant at 1% 0.199*** [6.45] 0.000*** [6.79] 0.001 [1.43] 0.113*** [4.32] [155.66] 2370 0.42 0.35*** [4.99] 57.46* [1.86] 8675.42*** [3.77] [7.84] 2370 0.37 0.140*** [7.90] 0.104*** [3.32] 9266.05*** [2.73] 0.115*** [3.41] 0.081*** [3.92] 0.147*** [4.77] 0.192*** [2.97] 0.147*** [8.64] 0.316*** [9.88] 0.212*** [4.88] -0.046*** [4.93] -0.001 [1.65] -56.7 [0.69] -0.016 [1.21] -0.060** [2.55] -0.189*** [5.16] -0.041*** [3.42] 0.132*** [3.83] 0.081*** [3.92] 0.162*** [5.13] 0.199*** [3.01] 0.166*** [9.45] 0.356*** [10.66] 0.220*** [5.03] Semi-log -0.023 [1.63] -0.044* [1.75] Linear

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 29 of 39

-3649.05*** [2.84] -4041.68 [1.27]

-3229.06*** [2.79] 16575.36*** [4.93] 2846.47 [1.41] 12954.03*** [3.79] 23272.52*** [2.59] 9417.34*** [6.59] 27736.13*** [7.45] 24266.76*** [3.47]

10.786*** 52685.30***

28

Case 1:01-cv-00201-VJW Document 219-10 Filed 10/05/2006 Table 4: Summary Statistics, Reduced Sample -- 1643 Observations Count Frequency House Sales in the buffer zone 1008 61.35 House Sales in 65 db zone 568 34.57 House Sales in 70 db zone 67 4.08 House Sales in Fulton County 856 52.10 House Sales in Clayton County 787 47.90 1 story 1475 89.77 2 or more stories 158 9.62 3 bedrooms 1057 64.33 4 bedrooms 190 11.56 5+ bedrooms 40 2.43 1 bathroom 1326 80.71 2 bathrooms 667 40.60 3+ bathrooms 270 16.43 0 or 1 fireplace 1579 96.10 2+ fireplaces 64 3.90

Page 30 of 39

dist age acres adj_price perc_tenureblack med_hhinc

Mean 3.54 39.05 0.41 $ 73,624.65 44.23 $ 30,383.43

Std. Dev. 1.10 17.52 0.38 $38,315.04 30.82 $ 9,391.25

29

Table 5: Models Without Noise Time Trend -- Full Sample -- 2003 Contours Log db65_03 db70_03 log_distance distance dStories beds_3 beds_4 beds_5 baths_2 baths_3 dFire_2 log_age age log_acres acres log_med_hhinc med_hhinc perc_tenureblack fulton Constant Observations R-squared 0.001** [1.96] [3.25] [29.13] 2370 0.44 Robust t statistics in brackets * significant at 10%; ** significant at 5%; *** significant at 1% 0.199*** [6.42] 0.000*** [6.89] 0 [1.05] [4.84] [159.94] 2370 0.42 0.38*** [5.59] 40.65 [1.30] [4.51] [7.68] 2370 0.37 0.139*** [7.92] -0.017 [1.19] -0.065** [2.01] -0.179*** [4.99] Semi-log -0.015 [0.98] -0.055 [1.64] Linear -1821.73 [1.14]

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 31 of 39

-7854.81*** [2.96]

-0.038*** -3052.57*** [3.27] [3.48] [3.83] [4.71] [2.92] [8.59] [9.91] [4.96] -0.046*** [4.86] -0.001 [1.53] -47.89 [0.59] [3.87] [3.86] [5.11] [2.98] [9.50] [10.76] [5.05] [2.73] [4.89] 2724.6 [1.37] [3.84] [2.57] [6.76] [7.56] [3.48] 0.117*** 0.132*** 16289.67*** 0.078*** 0.080***

0.144*** 0.161*** 12957.19*** 0.189*** 0.198*** 23361.62** 0.146*** 0.167*** 9737.64*** 0.316*** 0.358*** 27974.62*** 0.214*** 0.221*** 24342.13***

0.103*** 9333.77*** [3.30] [2.76]

0.087*** 0.124*** 10026.68*** 9.302*** 10.763*** 49497.87***

30

Table 6: Model Without Noise Time Trend -- Reduced Sample -- 95/03 contours Log db6565 db7070 log_distance distance dStories beds_3 beds_4 beds_5 baths_2 baths_3 dFire_2 log_age age log_acres acres log_med_hhinc med_hhinc perc_tenureblack fulton Constant Observations R-squared 0.001*** [3.34] 0.016 [0.50] [26.43] 1643 0.45 Robust t statistics in brackets * significant at 10%; ** significant at 5%; *** significant at 1% 0.219*** [6.52] 0.000*** [7.71] 0.001** [2.55] 0.035 [1.13] [141.79] 1643 0.44 0.52*** [7.53] 91.38** [2.55] 4289.35 [1.57] [9.67] 1643 0.39 0.121*** [5.95] -0.025 [1.50] [2.63] -0.288*** [6.77] Semi-log [2.16] [2.82] Linear [2.80] [4.33]

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 32 of 39

-0.038** -4725.48***

-0.090*** -0.099*** -12490.77***

-0.085*** -6486.35*** [6.01] [4.98] [3.70] [3.38] 0.122 [1.61] [8.44] [10.07] [4.38] -0.037*** [3.96] -0.001* [1.71] -59.27 [0.88] [5.48] [3.71] [3.52] 0.116 [1.54] [9.11] [10.60] [4.48] [4.91] [5.53] 3444.13 [1.35] [2.74] 12910.31* [1.87] [6.86] [9.79] [4.10] 0.164*** 0.182*** 18666.84*** 0.094*** 0.096***

0.128*** 0.137*** 11492.31***

0.174*** 0.192*** 11570.18*** 0.351*** 0.381*** 29223.73*** 0.196*** 0.200*** 18574.81***

0.110*** 10418.43*** [3.38] [2.74]

9.172*** 10.905*** 58897.29***

31

Table 7: Model With Noise Time Trend -- Full Sample -- 1995 Contours Log db65_95 db70_95 time_65_95 log_distance distance dStories beds_3 beds_4 beds_5 baths_2 baths_3 dFire_2 log_age age log_acres acres log_med_hhinc med_hhinc perc_tenureblack fulton Constant Observations R-squared 0.001 [1.63] 0.088*** [3.16] 9.509*** [29.46] 2370 0.44 Robust t statistics in brackets * significant at 10%; ** significant at 5%; *** significant at 1% 0.180*** [5.75] 0.000*** [6.04] 0 [0.74] 0.126*** [4.68] [156.67] 2370 0.42 0.32*** [4.42] 46.61 [1.42] 9240.91*** [4.02] [7.89] 2370 0.37 0.145*** [8.00] 0.109*** [3.47] 9524.65*** [2.78] 0.108*** [3.19] 0.080*** [3.89] 0.148*** [4.82] 0.195*** [2.98] 0.149*** [8.77] 0.319*** [9.98] 0.211*** [4.87] -0.045*** [4.85] -0.001* [1.76] -60.5 [0.73] -0.064*** [2.82] -0.054** [2.30] 0.012*** [2.77] -0.174*** [4.66] -0.037*** -3041.32*** [3.01] [3.46] 0.080*** [3.87] [5.18] [3.05] 0.167*** [9.55] [10.68] [5.11] [2.62] [4.78] 2784.23 [1.38] [3.81] [2.61] 9480.72*** [6.62] [7.45] [3.50] 0.120*** 16034.74*** Semi-log [3.20] -0.038 [1.51] 0.013*** [2.89] Linear [3.06] -3775.93 [1.18] 603.25 [1.58] -0.075*** -6056.79***

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 33 of 39

0.164*** 13043.61*** 0.205*** 23518.70***

0.356*** 27729.91*** 0.223*** 24403.01***

10.796*** 53113.32***

32

Table 8: Model With Noise Time Trend -- Full Sample -- 2003 Contours Log db65_03 db70_03 time_65_03 log_distance distance dStories beds_3 beds_4 beds_5 baths_2 baths_3 dFire_2 log_age age log_acres acres log_med_hhinc med_hhinc perc_tenureblack fulton Constant Observations R-squared 0.001 [1.44] 0.095*** [3.54] 9.446*** [28.43] 2370 0.44 Robust t statistics in brackets * significant at 10%; ** significant at 5%; *** significant at 1% 0.184*** [5.68] 0.000*** [5.97] 0 [0.52] [5.17] [160.99] 2370 0.42 0.33*** [4.48] 25.02 [0.79] [4.95] [7.89] 2370 0.37 0.141*** [8.02] 0.107*** [3.39] 9639.88*** [2.84] 0.113*** [3.34] 0.078*** [3.80] 0.145*** [4.74] 0.191*** [2.94] 0.148*** [8.64] 0.319*** [9.96] 0.213*** [4.95] -0.045*** [4.76] -0.001* [1.65] -54.07 [0.67] -0.068*** [2.63] -0.059* [1.82] 0.012** [2.27] -0.170*** [4.68] -0.036*** [3.02] [3.60] 0.078*** [3.80] [5.14] [3.03] 0.168*** [9.55] [10.80] [5.12] -2829.78** [2.54] [4.63] 2604.43 [1.31] [3.86] [2.60] 9866.31*** [6.77] [7.59] [3.51] Semi-log [2.89] -0.048 [1.42] 0.015*** [2.71] Linear [2.60] -7254.97*** [2.71] 1240.15** [2.15] -0.076*** -6984.28***

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 34 of 39

0.124*** 15589.52***

0.162*** 13050.08*** 0.203*** 23778.58***

0.358*** 27975.68*** 0.223*** 24525.04***

0.132*** 10699.49*** 10.776*** 50652.91***

33

Table 9: Model With Noise Time Trend -- Reduced Sample -- 95/03 contours Log db6565 db7070 time_6565 log_distance distance dStories beds_3 beds_4 beds_5 baths_2 baths_3 dFire_2 log_age age log_acres acres log_med_hhinc med_hhinc perc_tenureblack fulton Constant Observations R-squared 0.001*** [2.99] 0.021 [0.65] 9.250*** [25.64] 1643 0.45 Robust t statistics in brackets * significant at 10%; ** significant at 5%; *** significant at 1% 0.211*** [6.04] 0.000*** [7.15] 0.001** [2.22] 0.04 [1.30] [142.00] 1643 0.44 0.51*** [7.13] 88.58** [2.35] 4423.90* [1.68] [9.60] 1643 0.39 0.124*** [5.92] 0.112*** [3.42] 0.160*** [4.82] 0.093*** [3.68] 0.130*** [3.41] 0.124 [1.62] 0.175*** [8.46] 0.352*** [10.07] 0.196*** [4.38] -0.037*** [3.91] -0.001* [1.74] -59.85 [0.89] -0.053* [1.71] -0.086** [2.52] 0.007 [1.10] -0.282*** [6.60] -0.083*** [5.81] 0.175*** [5.22] 0.095*** [3.68] 0.139*** [3.55] 0.12 [1.57] 0.193*** [9.12] 0.381*** [10.59] 0.201*** [4.53] Semi-log -0.072** [2.22] [2.68] 0.008 [1.28] Linear -5598.95* [1.69] [4.26] 212.75 [0.34]

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 35 of 39

-0.095*** -12381.14***

-6435.48*** [4.99] 18494.43*** [5.42] 3416.85 [1.35] 11545.11*** [2.73] 13004.78* [1.87] 11589.62*** [6.82] 29226.26*** [9.78] 18600.97*** [4.12]

10482.35*** [2.73]

10.909*** 58993.94***

34

Case 1:01-cv-00201-VJW

Figure 1 The Location of Hartsfield-Jackson Atlanta International Airport

Document 219-10

Filed 10/05/2006

Page 36 of 39

.

!
Fulton

DeKalb

[
Clayton

Clayton County DeKalb County Fulton County
!

Downtown Airport

[

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 37 of 39

Figure 2 The Location of Houses in the Full Sample

Runways Hartsfield Atlanta Intl Airport 1995 75 db noise contour 1995 70 db noise contour 1995 65 db noise contour 0.5 mile buffer zone Houses in the full sample

Case 1:01-cv-00201-VJW

Document 219-10

Filed 10/05/2006

Page 3