Abstract
Value Capitalization Effect of Protected Properties: A Comparison of Conservation Easement with Mixed-Bag Open Spaces A u t h o r Jay Mittal A b s t r a c t In this paper, I examine the impact of open space restrictions on neighboring house prices using hedonic modeling framework and GIS. The comparison is between two groups of parcels in Worcester, Massachusetts: one has a mixture of open space restrictions that limit or prohibit development and the other has a conservation easement. Conservation easement (CE) involves voluntarily restricted conservation worthy private lands from future developments in perpetuity. The sample used is single-family detached houses that were sold in 2005 to 2008. Since future development restrictions lower the property values and tax base for local communities, the findings confirm that spatially targeted CE parcels with proximity of, and visibility to CE parcels drive up the surrounding property values, thereby providing additional tax base and income to the communities. Conservation easement (CE) is a voluntary land protection tool that has been in existence since the 1800s in United States. It is used in preserving conservation worthy private lands that otherwise have potential to change due to surmounting urban growth pressure (Whyte, 1959; Brenneman and Bates, 1984). Typically, CE entails a legal agreement between a landowner and a qualified non-profit or government organization that permanently limits future development of the land in subject. The participating private landowners either can donate, or sell their property development rights and can then claim federal tax credits (Wright, 1994; Gustanski and Squires, 2000). Barring the future development rights, the landowners continue to retain title of the property and right to enter, farm, lease, mortgage, bequeath or even sell (Whyte, 1959; Merenlendar, Huntsinger, Guthey, and Fairfax, 2004; Daniels and Lapping, 2005). CE can be tailored to the needs of each property owner, but usually limit any form of subdivisions, non-farm- based development, and other uses that hinder the conservation objective. Further, the easement agreement could be applied either to the total parcel or its partial acreage and is aimed to protect the environmental amenity. The restrictions are self-imposed and are consistent with the conservation objectives (Boyd, Caballero, and Simpson, 1999; Gustanski and Squires, 2000). A CE protects preservation- worthy lands by encouraging landowners to act in ways that further their own J O S R E V o l . 6 N o . 1 – 2 0 1 4 u u 2 4 M i t t a l self-interest and the public good. Several parks, trails, waterways, and wildlife areas in the past have been protected using CE. CE not only the most commonly used, but also the fastest growing private land protection tool in the U.S. Over 10 million acres of land has been conserved just during 2005 and 2010; over 47 million acres of land has been conserved via local, state, and national land trusts (Land Trust Alliance, 2012), which in terms of land area compares to approximately five times the size of Massachusetts and is even larger than the total land area of Ohio. Of this, 8.83 million acres is privately owned and preserved under CE agreements. The recent gain in the popularity of CE as a land-protection tool is due to its cost effectiveness owing to ‘‘shared’’ ownership in land where landowners and conservation agencies both share partial rights in the land. CE is less expensive to conservation agencies as compared to the fee-simple with full property rights, as no upfront acquisition costs are involved. This form of private land protection involves serious public tax dollars. Billions of dollars’ worth of public money is involved in land conservation efforts; as tax abatements and in acquisition of new lands. In terms of tax abatements, over $10.21 billion in tax deductions were claimed from 2003 to 2008. In 2008 alone, CE landowners claimed about $1.21 billion in tax deductions while in 2007 it was $2.1 billion (Colinvaux, 2012). The environmental concerns and tax savings from CEs seem to be the dominant driving forces behind the importance and growth of CEs, encouraging landowners to act in ways that further their own self- interest, as well as the public good. Brenner, Lavallato, Cherry, and Hileman (2013) surveyed 513 private landowners in Finger Lakes Region of New York and found that owners’ personal characteristics such as gender, education, being part of environmental organizations, and how actively are they engaged with their land and its usage are a few important attributes in predicting owners’ interest in CE. Further, subsistent, passive, and recreational land use activities, if they exist on the subject land, all play important roles in predicting the potential interest from the landowners’ perspective in CE. S t u d y M o t i v a t i o n a n d i t s L o c a t i o n C o n t e x t This form of private land protection involves serious public tax dollars. A few local communities believe that CEs lower property taxes, diminishing local revenues. With billions of public dollars involved in protecting private land in terms of tax abatement and other costs, it is important to understand whether there is a measurable economic benefit to the community in terms of enhancement in the surrounding property values. Additionally, is this benefit similar to the benefit that different types of open spaces externalize or, would CE parcels externalize benefits differently to its surroundings? The City of Worcester was chosen to quantify the measurable benefits from CEs. City of Worcester is the second largest city in Massachusetts and is 60 miles southwest of Boston. It has a population of 181,045 and median family income of $79,700 (Census, 2010). The city is situated near the headwaters of the Blackstone River that forms the John H. Chafee Blackstone River Valley National V a l u e C a p i t a l i z a t i o n E f f e c t o f P r o t e c t e d P r o p e r t i e s 2 5 Heritage (BRVNH) Corridor. This is one of the 49 National Heritage Areas designated by the U.S. Congress for their unique qualities and resources (National Heritage Area, 2013). This BRVNH corridor includes 22 communities and is a repository of historically, environmentally, and ecologically important sites located throughout its watershed from Worcester, MA to Providence, RI (Billington, 2004). Worcester is the largest city in this corridor and has 26 clusters of CE land parcels that are privately owned and voluntarily conserved. These CE land parcels are either in joint-ownership with private landowners, the city, and its conservation commission, or with the private non-profit land trusts such as the Greater Worcester Land Trust (GWLT) and the Massachusetts Audubon Land Trust. The 26 CE parcels range from approximately one acre to over 400 acres in size. Several parcels are scenic in nature, preserving environmental amenities such as farms, urban gardens, historic sites, rivers, streams, waterfalls, flora, and fauna. In this paper, I examine the perceptions of the house buyers and sellers in the City of Worcester regarding privately protected CE parcels. L i t e r a t u r e R e v i e w o n E n v i r o n m e n t a l S t u d i e s Researchers have studied the relationships between various environmental amenities such as farmlands (Geoghegan, 2002; Geoghegan, Lynch, and Bucholtz, 2003), forests (Ham, Loomis, and Reich, 2012), public parks and open spaces (Crompton, 2001, 2007, 2008; National Association of Realtors, 2001; Troy and Grove, 2008), waterfronts (Lansford and Jones, 1995; Benson, Hansen, Schwartz, and Smersh, 1998; Mahan, Polasky, and Adams, 2000; Shultz and Schmitz, 2008; Conroy and Milosch, 2011; Walsh, Milon, and Scrogin, 2011) and other environmental amenities (Simons, 1999; Boyle and Kiel, 2001; Bourassa, Hoesli, and Sun, 2004; McConnell and Walls, 2005; Simons and Saginor, 2006). A few researchers have discussed the impact of various amenities on house prices; however, there are very few who have measured the impact of conserved land. Value Capitalization Effect of Proximity Boyle and Kiel (2001) reviewed 35 hedonic studies in relationship with pollution point sources and their effect on the home values. Most studies in this review focused on various forms of proximity measures (distance) from the pollution source. A few also focused on visibility analyses (e.g., visibility through the high suspended particulate matter (SPM), content in the air quality studies, and visibility of waste dumping sites or undesirable land uses. All studies reported that the polluting sources generate negative externality and as the proximity of homes to the polluting sites increases, home value decreases. McConnell and Walls (2005) reviewed 60 published studies, where 40 studies just focused on the effect of general open space, parks, natural areas, green buffers, greenbelts, wildlife habitats, wetlands, forest preserves, farmlands, and golf courses on home values using hedonic framework. A generic finding was that proximity to golf courses and lakes had a significant positive effect on house prices. In general, proximity to a large natural area or wildlife habitat contributed to a 0.07%–4% increase in house prices. J O S R E V o l . 6 N o . 1 – 2 0 1 4 u u 2 6 M i t t a l The proximity variable came out as an important variable in most studies, with a general consensus that proximity of parks and open spaces increase house prices. Crompton (2001) reviewed 30 impact studies of parks and opens spaces where 25 studies reported a positive impact on surrounding property prices. The impact varied considerably with park attributes, such as the size and typology of the park, but park proximity contributed a 10%–20% increase in the nearby property prices. This effect of proximity extended to at least 500 feet in some cases, while in other cases as far as 2,000 feet into the surrounding neighborhoods. Geogheghan (2002) used a measure of 1,600 meters (5,250 feet) or a 20-minute walking distance from open spaces, while Acharya and Bennett (2001) used a 0.25 mile (2,640 feet) visual zone and a one mile (5,280 feet) increment for walking distance. The proximity of undesirable activities adversely impact property prices (Simons, 1999; Simons and Saginor, 2006). Palmquist, Roka, and Vukina (1997) studied the effect of large-scale hog operations on surrounding property prices and developed an index of hog manure production at different distances from the houses. They concluded that proximity caused statistically significant reductions in house prices of up to 9%, depending on the number of hogs and their distance from the house. Parks increase property prices; however, undesirable and unsafe activities such as crime, heavy traffic, and loud activities reduce it. Troy and Grove (2008) studied the relationship between property prices, parks, and crime in Baltimore, MD. They concluded that crime is a critical factor in how residents perceive parks. When crime rates are relatively low, parks have a positive impact on property prices; however, for each unit increase in the crime score estimated for a given park, there is a 0.017% decrease in the values of the homes associated with that park. Proximity to incompatible land uses reduces property prices. Song and Knaap (2004) measured the effect of six different mixed-land uses on the prices of single- family houses in Portland, OR. They used four key explanatory variables to define the characteristics of mixed-land uses: (1) distance to land uses from each transit area zones (TAZ); (2) proportional areas of each non-residential land uses within a neighborhood; (3) diversity index; and (4) measure of job to population ratio. The control variables included structural, public sector, metro level accessibility characteristics, amenity characteristics, neighborhood socio-economic characteristics, and urban design characteristics. They concluded that prices tend to fall near multi-family houses and rise in neighborhoods dominated by single- family houses, and where non-residential land uses are evenly distributed and more service jobs are available. Proximity of well-planned neighborhoods with amenities, availability of high- quality infrastructure services, and direct access to the amenities and infrastructure to the nearby land owners increases their property value. This land value capture effect is used extensively to finance large-scale public infrastructure projects (Mittal, 2013, 2014). Value Capitalization Effect Varies with Environmental Amenity The ability to place an economic price on ecosystem services is central to formulating sound environmental policy (Krupnick and Siikama ¨ki, 2007), which V a l u e C a p i t a l i z a t i o n E f f e c t o f P r o t e c t e d P r o p e r t i e s 2 7 is true in the case of CEs as well. Researchers have used contingent valuation (a stated preference) and hedonic valuation (a revealed preference) for valuing the contribution of environmental goods (Boyle and Kiel, 2001; Hidano, 2002; Malpezzi, 2002; McConnell and Walls, 2005; Sirmans, Macpherson, and Zietz, 2005; Boyd, 2008). The hedonic studies of various environmental amenities vary in terms of how they incorporate the amenity. For example, for open space variable, Acharya and Bennett (2001) used a single open space variable to represent all the lands with no developments. Others (Bolitzer and Netusil, 2000; Lutzenhiser and Netusil, 2001; Anderson and West, 2006) differentiated between different types such as parks, golf courses, cemeteries, and other open spaces. In most cases, researchers who made this distinction found that houses in close proximity to parks have higher property prices, all other factors held constant. Asabere and Huffman (1996), Do and Grudnitski (1997, 1995), Netusil (2005), and Bark, Osgood, Colby, and Halper (2011) estimated that the effect of golf courses on adjacent house prices range from 4.8% to 8%; however, this impact quickly diminishes as the distance from the golf course increases. Authors of an earlier impact study of four parks in Worcester, MA concluded that a house located 20 feet from a park sold for $2,675 more (1982 price) than a similar house 2,000 feet away [Stevens, More, and Allen (1982) as quoted in NPS (1995)] keeping other factors constant. Standiford and Scott (2001) used regression modeling of land price and found that property prices significantly increase around open spaces. The prices of houses adjacent to a less-developed open space increase by 23%–32%, as compared to a house a block away. Similarly, Correll, Lillydahl, and Singell (1978) and Nelson, Duncan, Mullen, and Bishop (1995) found that average price per acre increases by $1,200 if the home property is within 1,000 feet of open space. In another study of land parcels in Colorado, Loomis, Rameker, and Seidl (2004) found that a property parcel with access to water body commands a $937 higher price per acre than average, while if a similar parcel is adjacent to a park or open space, the price increases by as much as $11,039 an acre. So in theory, proximity to open space and greater accessibility to recreation opportunity and scenic view enhances property prices; however, the effects vary with the nature of the environmental amenity. Permanent protection increases property prices while intense activities reduce prices. Le Goffe (2000) used the hedonic price method to identify and monitor the external effects of agricultural and sylvi-culture activities. The author examined the rental prices of rural self-catering cottages. Intense livestock farming caused the rental prices to decrease whereas permanent grassland had the opposite effect. In general, perpetually preserved land increases desirability; and so, the price of the surrounding property increases as amenity seekers are willing to pay a premium for the perpetual presence of open space (Brewer 2003; Thompson, 2004; Mitchell and Johnson, 2005). Value Capitalization Effect of Views Proximity is important but view of a desirable land use is important too. Researchers have repeatedly used view as an important variable in impact measurement. Appleton (1975) explained the appeal of views by proposing that J O S R E V o l . 6 N o . 1 – 2 0 1 4 u u 2 8 M i t t a l ‘‘humans are biologically programmed to prefer vantage points where it is possible to see a good deal without necessarily being seen.’’ Homes that command scenic views have a premium price (Wolverton, 1997; Benson, Hansen, Schwartz, and Smersh, 1998; Lake, Lovett, Bateman, and Langford, 1998; Lake, Lovett, Bateman, and Day, 2000; Sirmans, Macpherson, and Zietz, 2005; Shultz and Schmitz, 2008). The proximity measure of distance from a water body is an important variable but view is even more important. For example, a one mile increase from the coast reduced the house price by $8,680 (Conroy and Milosch, 2011). Lansford and Jones (1995) estimated the marginal price of water in lake recreational and aesthetic use. A hedonic price equation indicated that lakefront location, distance to lake, and scenic views are significant in house price. Waterfront properties command a premium price for the private access they offer. Beyond the waterfront, the marginal price falls rapidly with increasing distance, becoming asymptotic to minimum. In their sample, 22% of housing prices were found to be attributable to the recreational and aesthetics that lakefronts offer. Bourassa, Hoesli, and Sun (2004) reviewed 30 studies that focus on scenic views and their effect on home values. Several types of view amenities were studied, such as water views, lake and ocean views, mountain or valley views, agriculture and farmlands, forests views, and open spaces with landscapes. Thirteen studies used distance to lake; 28 studies used a binary yes / no dummy view variable or a 1-to-5-scaled dummy view-quality variable; one study used degree of panorama; three studies used GIS or viewshed-based view scores; and three studies used land use diversity as a proxy for the view. It was concluded that the view premium was highest for the wide views of water such as lake views, with a very high premium of 89% for the lakefront abutting homes, while ocean view front homes commanded a 129% premium. The wooded areas, landscaped areas, and forest views had a relatively lower premium of 3%–8%, on average. The positive impact of view was also found to diminish with distance. Uninterrupted quality of views increases property price. Benson, Hansen, Schwartz, and Smersh (1998) studied the estimated price of the view amenity in single family residential real estate markets in Bellingham, Washington, a city with a variety of views, including ocean, lake, and mountain. Results from a hedonic model, estimated for several years, suggested that depending on the particular view, willingness to pay for this amenity is quite high. The highest quality ocean views were found to increase the market price of an otherwise comparable home by almost 60%, while the lowest quality ocean views were found to add only about 8%. For ocean views of all quality levels, the price of a view is found to vary inversely with distance from the water. The diversity of land use surrounding a house is important in value creation. Geoghegan, Wainger, and Bockstael (1997) enquired into the spatial patterns of land use character and how this pattern contributes to the price in Patuxent watershed counties of Washington D.C. suburbs. Developing spatial land use diversity indices for areas within a 0.1-kilometer radius, they concluded that the proportion of open space positively impacts land prices; however, within a 1- kilometer buffer, this variable negatively influences land prices. They interpreted this result to suggest that individuals price open space, like a view from their V a l u e C a p i t a l i z a t i o n E f f e c t o f P r o t e c t e d P r o p e r t i e s 2 9 house (visual zone within 100 meters), and prefer to be able to walk to diverse land uses at the larger scale from their houses. Conclusions from Past Studies Based on the past studies, the key impact variables that were found useful include: home to CE inverse distance, weighted sum of inverse distance (accessibility index), land use diversity index (Pooler, 1987; Geoghegan, Wainger, and Bockstael, 1997; Acharya and Bennett, 2001), and view shed analysis (Wolverton, 1997; Lake, Lovett, Bateman, and Langford, 1998; Lake, Lovett, Bateman, and Day, 2000; Shultz and Schmitz, 2008). The interaction effect of the view and distance was even more important than just the view. In most articles above, researchers used contingent valuation (a stated preference) and hedonic valuation (a revealed preference) to estimate the implicit price of the environmental amenity (Boyle and Kiel, 2001; Hidano, 2002; Malpezzi, 2002; McConnell and Walls, 2005; Sirmans, Macpherson, and Zietz, 2005; Boyd, 2008). However, the hedonic estimation technique was more commonly used. D a t a a n d M e t h o d o l o g y In this study, I used a hedonic framework to estimate the price capitalization effect of CEs. Two types of GIS-based data were used. One included two datasets of Environmental Amenity Generators, which included a mixed bag of open spaces in the city and its subset that only included the CE parcels. The other dataset was Environmental Amenity Absorbers, which included single-family detached (SFD) houses sold between 2005 and 2008 within a close proximity of the Environmental Amenity Generators. Preparing Datasets for Analyses Environmental Amenity Generators Datasets: The two datasets of Environmental Amenity Generators were available in GIS format (*.shp) from the Office of Geographic and Environmental Information (MassGIS). The first set included 54 parcels and was a larger dataset, as shown in Exhibit 1. This set included all types of ‘‘mixed-bag open spaces’’ in the City of Worcester, such as open spaces with active and passive recreation, and was used for various purposes such as public parks, golf courses, ball parks, playgrounds, and trails. This dataset also included CE parcels that city and land trusts owned. The second dataset was a smaller subset of the first one and included only 26 conservation easement parcels in the City of Worcester, as shown in the Exhibit 1 (see left map). The size of these 26 CE parcels varied from as small as one acre to as large as 400 acres (Exhibit 2). These CE parcels have environmental amenities that are scenic in nature such as waterfalls, streams, ponds, large boulders, marsh, wetlands or vernal pools, woods and vegetated lands, including a mature hardwood forest, mountain laurels, and silver beech, as well as various types of environmental amenities. Some larger CE parcels are also habitat for wildlife including deer, birds, and rare species such as spotted turtles and salamanders in the riparian region. The parcels also include J O S R E V o l . 6 N o . 1 – 2 0 1 4 u u 3 0 M i t t a l Exhibit 1 u Environmental Amenity Generators Sources: Greater Worcester Land Trust, Registry of Deeds, Worcester and City Assessor’s land parcel data. habitat for giant pileated woodpeckers, turkeys, and owls, as well as over 80 species of butterflies. There is a conservation center in one of the parcels that offers scientific, educational, and passive recreation environmental amenities to the public. Environmental Amenity Absorbers Dataset (Houses): The effect of the environmental amenity was observed on the sale prices of SFD houses. The sales data for 2005–2008 was used for all SFD transactions within a half-mile from the 54 mixed-bag open space parcels. This data set included a total of 2,406 SFD houses (mean sales price 5 $228,880, std. dev. 5 $71,242) that were sold during this period. A smaller subset of this SFD sales data was extracted for all sales within a half-mile distance from only the 26 CE parcels after eliminating all the sales that may have occurred around other mixed-bag amenities. This data set included a total of 1,244 SFD sales (mean sales price 5 $232,511, std. dev. 5 $71,318). The descriptive statistics of these two SFD sales data sets are given in Exhibit 3. Hedonic Model for Measuring Value Capitalization Effect Two OLS-based hedonic models were prepared using the spatial explanatory variables. One model focused only on the 26 CE-parcels as externality generator and 1,224 SFD houses as the externality absorbers. This model is called the ‘‘CE Model.’’ The other model was for all the 54 mixed-bag open spaces as externality V a l u e C a p i t a l i z a t i o n E f f e c t o f P r o t e c t e d P r o p e r t i e s 3 1 Exhibit 2 u Land Under Conservation Easements Site Name Area (Acres) Type of Easement City Land under Conservation Easements by GWLT, Mass Audubon, & City of Worcester Broad Meadow Brook Savannah 87.00 Conservation Easement Worcester Cascades East 30.86 Conservation Easement Worcester Coal Mine Brook Parcel 7.30 Conservation Easement Worcester Coal Mine Brook II Parcel 4.60 Conservation Easement Worcester Crow Hill 27.90 Conservation Easement Worcester Green Hill Park 487.00 Conservation Easement Worcester Parson’s Cider Mill 43.08 Conservation Easement Worcester Ryan Ornamental 1.94 Conservation Easement Worcester Subtotal (8) 677.00 Mass Audubon Granite Street Conservation Area 14.00 Conservation Easement Worcester Coes Reservoir ,12.00 Conservation Easement Worcester NEPC (Nr. Granite) 108.00 Conservation Easement Worcester Cooks Pond ,32.00 Conservation Easement Worcester Nr. NEPC 14.50 Fee Owned Property Hjeim Road ,12.00 NA Worcester Massasoit Rd ,3.00 Worcester Sprague Ln ,1.80 Worcester GWLT Owned Land (Fee Owner) Bovenzi Conservation Area 120.68 Fee Owned Property Worcester Brigham Road Parcels 2.53 Fee Owned Property Worcester Cascades West 122.99 Fee Owned Property Worcester periphery/ Holden/Paxton Cascading Waters 2.40 Fee Owned Property Worcester Curtis Pond Parcel 50.10 Fee Owned Property Worcester Kettle Brook 14.37 Fee Owned Property Worcester Marois Property 28.20 Fee Owned Property Worcester/Leicester Nick’s Woods 59.76 Fee Owned Property Worcester Sargent’s Brook Property 55.00 Fee Owned Property Worcester/Holden Southwick Pond 113.77 Fee Owned Property Paxton/Leicester Subtotal (10) 470.00 Government Land Preserved with GWLT Assistance Antell Conservation Land 280.00 Partner Massachusetts Spencer/E. Brookfield DEM Turkey Hill Brook Addition to 30.00 Partners Massachusetts Paxton Moore State Park DEM; Paxton Land Trust Subtotal (2) 310.00 Notes: The sources are Greater Worcester Land Trust (2008) and Mittal (2011). J O S R E V o l . 6 N o . 1 – 2 0 1 4 u u 3 2 M i t t a l Exhibit 3 u Descriptive Statistics of SFD House Sales 1,244 Home Sales within 0.5 2,406 Home Sales within 0.5 miles from 26 Parcels miles from 54 Parcels Home Variables Mean Std. Dev. Mean Std. Dev. Sale price $232,511 $71,318 $228,880 $71,242 Time of Sale sl 2005 (sale year 2005) 0.32 0.465 0.32 0.47 sl 2006 (sale year 2006) — — 0.27 0.44 sl 2005 (sale year 2007) 0.26 0.438 0.25 0.43 sl 2007 (sale year 2008) 0.16 0.370 0.17 0.37 Structural Features Tula: Total utilizable area (sq. ft.) 1,401 581.49 1,368 536.66 Lotsf: Lot area (sq. ft.) 10,185 9,387.74 9,584 8,976.47 Qual: Quality of house 40.14 4.15 40.00 4.01 Age: Age of house 67.38 138.87 68.87 129.57 Beds: No. of bedrooms 2.98 0.87 2.95 0.84 Bath: No. of bathrooms 1.30 0.54 1.28 0.52 Hbath: Half bath 0.45 0.52 0.42 0.51 Deck: Deck (dummy) 0.31 0.46 0.29 0.46 Patio: Patio (dummy) 0.04 0.21 0.04 0.20 Neighborhood Variables hous dens: Housing density 3.07 2.49 3.21 2.40 prc black: Percentage of blacks 4.63 5.11 4.58 5.71 md hs val: Median house price $121,794 $25,049.52 $120,611 $26,075.31 Note: The source is the city assessor’s SFD house sales data. generator and 2,406 SFD houses as the externality absorbers and is called the ‘‘OS Model OS.’’ The OS Model is linear and is expressed as: Y 5 b 1 b T 1 b S 1 b N 1 b E 1 « , i 0 1 i 2 i 3 i 4 i i where: Dependent Variable: Y 5 Sale price (in $). Independent Variables: T 5 Vector of time of sale 5 sl 2005, sl 2006, sl 2007, sl 2008; S 5 Vector of structural features 5 Tula, Lotsf, Beds, Bath, Qual, Hbath, Age, Patio, Deck; N 5 Vector of neighborhood features 5 Prc black, Md hs val, Hous dens; E 5 Vector of environmental features: d , V , and A ; and i ij ij ij « 5 Error term. i V a l u e C a p i t a l i z a t i o n E f f e c t o f P r o t e c t e d P r o p e r t i e s 3 3 The three spatial explanatory variables were in a matrix form and included: (1) the visible area of each CE( ) parcel vertices from each HOME( )—V to measure j i ij visibility; (2) the shortest distance to the visible portion of each CE( ) parcel from each HOME( )—d to measure proximity; and (3) a weighted index of view and i ij distance for each HOME( )—A to measure visual accessibility from each house i ij to each CE parcel and mixed-bag of open space parcels. For the proximity variable dij, in ArcGIS, Chasan’s (2003) visual basic tool (vbtool) was used to measure multiple distances between the two data sets: between environmental amenity generators and absorbers. The first distance matrix was between SFD houses (2,406) and all 54 mixed-bag open space parcels; the second distance matrix was from the SFD homes (1,244) to the 26 CE parcels. For visibility variable Vij, first, three-dimensional digital models for the entire city were created using topographic and building height data (Lake, Lovett, Bateman, and Langford, 1998; Lake, Lovett, Bateman, and Day, 2000; Sander and Manson, 2007; Shultz and Schmitz, 2008; Sander and Polasky, 2009). Once this data was ready, viewsheds were created from each home sample. Using the spot elevations and topographic data in ArcGIS Spatial Analyst, first digital elevation model (DEM) for the topography of the City of Worcester was created; then, the 3-D view impending built structures in the City were added to this DEM. Using the building footprints data in GIS, heights were assigned to the building footprints, to generate a 3-D surface of all the built structures in the city. Finally, the two surfaces were combined to form a seamless 3-D model for the entire city where the built structures in the city were draped over the topographic surface (Exhibit 4). The viewsheds were then created in ArcGIS Spatial Analyst, between the two sets of SFD (2,406 houses and 1,244 houses) and the two sets of environment amenities (the 26 CE parcels and the 54 mixed-bag open space parcels). All viewsheds were calculated at human eye level—five feet above the ground. The GIS viewshed analysis generated an output Viewshed( ) raster for visible and non- visible areas from the observation points. This viewshed output had only two possible pixel values: a value of one indicated visible and a value of zero indicated invisible. The viewsheds were shot keeping the observer(s) at the periphery of environmental amenities. The resulting viewsheds were then summed, creating view counts ranging from 0 to 677 for SFDs, the value of zero on the viewshed output raster meant no view, while the higher value meant more view from that point to the subject environmental amenity. After categorizing the final viewshed raster into five view groups, in GIS, the entire set of SFD data parcels was then clipped with this final viewshed raster to assign view counts to the SFD datasets. The descriptive statistics for the two sets of variables are in Exhibit 5. M o d e l S u m m a r y The OS Model included 2,406 SFD house sales (mean sales price 5 $228,880, std. dev. 5 $71,242) within a 0.5-mile distance from all 54 open-space parcels transacted between 2005 and 2008. The house sales within a 0.5-mile buffer area J O S R E V o l . 6 N o . 1 – 2 0 1 4 u u 3 4 M i t t a l Exhibit 4 u Merged Raster: DEM of Building Heights and Topographic Features of the City from the 54 open-space parcels (environmental amenities) were used for this research. This buffer is like local submarket characteristics and is used to control for any other spatial factors (variable not accounted for in the model) that may affect the house prices. This OS Model measured the combined effect of all the 54 open-space parcels in Worcester on SFD prices. This model also included 26 CE-protected parcels. The control variables used in the hedonic model were physical home characteristics, neighborhood characteristics, and the year of house sale. The three test variables were: dij 54, the squared distance from the nearest open space property; ViewInteract54, interaction of view and squared distance; and Vij54, view to open space parcel vertices. First, the test of heteroscedasticity was conducted visually. On plotting the residuals, they were found to be randomly patterned, which indicates homoscedasticity. Any linear form of heteroscedasticity can be detected using the V a l u e C a p i t a l i z a t i o n E f f e c t o f P r o t e c t e d P r o p e r t i e s 3 5 J O S R E V o l . 6 N o . 1 – 2 0 1 4 u u Exhibit 5 u Descriptive Statistics of Two SFD Sales Data Sets 2,406 Home Sales within 0.5 Miles from all Open Spaces: 1,244 Home Sales within 0.5 Miles Open Space, City Parks, Playgrounds, CE-protected 54 from CE-protected 26 Parcels Parcels Home Variables Mean Std. Dev. N Mean Std. Dev. N Test Variables dij 26: Squared distance from the nearest CE property (ft.) 779,386 720,116 1,244 — — — (882 feet) (849 feet) dij 54: Squared distance from the nearest open spaces (ft.) — — — 5,587,790,324 125,663,110,498 2,406 (74,751 feet) (354,489 feet) ViewInteract26: Interaction of view and squared distance 68,286 95,436 1,244 — — — of the nearest CE property ViewInteract54: Interaction of view and squared distance — — — 476,052,893,686 13,693,065,342,153 2,406 of the nearest open spaces (ft.) Vij 26: View to CEs 57.24 65.32 1,244 — — — Vij 54: View to open spaces — — — 53.85 63.98 2,406 3 6 M i t t a l Exhibit 6 u OS Model Summary Change Statistics Std. Error of 2 2 2 Model R R Adj. R the Estimate R Change F Change df1 df2 Sig. F Change OS .725(a) .53 0.522 49,280.23 .53 146.63 18 2,387 .000 Notes: Predictors: (Constant), md hs val, sl 2006, patio, view, age, InvSq154, deck, lotsf, hbath, sl 2008, prc black, beds, hous dens , bath, sl 2007, tula, qual, ViewInteract154. The dependent variable is saleprice. Breusch-Pagan test for heteroscedasticity in SPSS. After running the test, the small chi-square value indicated that heteroscedasticity was absent from the sample. Exhibit 6 provides a summary of the OS Model and Exhibit 7 provides beta weights of the independent variables. As can be seen in the Exhibit 6, the variables have 52.2% explanatory power (adjusted R ). As can be seen in Exhibit 7, none of the environmental externality capturing variables was significant. The open spaces used in the OS Model represented all open spaces in Worcester—both actively used open spaces, such as playgrounds and parks, and passively used open spaces, such as CE parcels. The actively used open spaces include ball parks, playgrounds, and basketball courts, which could be construed as loud, or likely to generate high traffic volume, which may be less desirable to some amenity- seeking homeowners. This effect of greater noise and intense activity levels could result in a negative externality impact on house prices, which is evident in the significance level of the explanatory variables. The OS Model potentially has a mixed effect of positive and negative externality generating various types of open spaces. In the CE Model, only the effect of CE-protected properties was measured. The model included 1,244 home sales (mean sales price 5 $232,511, std. dev. 5 $71,318) within a 0.5-mile distance from CE-protected properties in Worcester, representing sales transacted between 2005 and 2008. The control variables were the same as in the OS Model: physical home characteristics, neighborhood characteristics, time of sale. Three test variables were used to measure the CE amenity effect: dij 26, the squared distance from the nearest CE property; ViewInteract26, the interaction of view and squared distance; and Vij26, view to CEs. Exhibit 8 is a summary of the CE Model and Exhibit 9 provides beta weights for various independent variables. Exhibit 8 shows that the CE Model CE has a greater explanatory power than the OS Model. Nearly 58.4% variance in the SFD house prices can be predicted from the variables used in this model. The model was also tested for multicollinearity. The VIF is , 3, which indicates that the model was stable and there was no multicollinearity detected among the variables. All variables independently contributed to the model’s predictive power. The beta V a l u e C a p i t a l i z a t i o n E f f e c t o f P r o t e c t e d P r o p e r t i e s 3 7 J O S R E V o l . 6 N o . 1 – 2 0 1 4 u u Exhibit 7 u OS Model: OLS Output with 2,406 Houses within a 0.5 Mile Distance from all Open Space Properties for 2005–2008 Collinearity Unstd. Coeff. Std. Coeff. T Sig. 95% Con. Inter. for b Correlations Statistics Model b Std. Error b Lower Bound Upper Bound Zero-order Partial Part Tolerance VIF Constant 297,699.860 15,314.637 26.380 0.000 2127,731.225 267,668.496 Year of Sale (Dummy) sl 2005* (Sale Year 2005) 26,256.805 2,637.057 20.039 22.373 0.018 211,427.963 21,085.646 0.070 20.049 20.033 0.736 1.358 sl 2007* (Sale Year 2007) 221,361.873 2,709.055 20.129 27.885 0.000 226,674.216 216,049.530 20.035 20.159 20.111 0.739 1.353 sl 2008* (Sale Year 2008) 250,174.252 3,056.659 20.262 216.415 0.000 256,168.232 244,180.271 20.202 20.318 20.232 0.779 1.283 Physical Features of Houses Tula*: Total utilizable area 43.099 3.017 0.325 14.285 0.000 37.183 49.015 0.579 0.281 0.201 0.385 2.596 Lotsf *: Lot area sq. ft. 0.813 0.117 0.102 6.944 0.000 0.583 1.042 0.262 0.141 0.098 0.914 1.093 Qual*: Quality of house 5,202.145 419.153 0.293 12.411 0.000 4,380.204 6,024.087 0.389 0.246 0.175 0.358 2.792 Age*: Age of house 159.424 10.932 0.290 14.583 0.000 137.987 180.862 0.015 0.286 0.206 0.503 1.987 Beds: No. of bedrooms 22,612.217 1,632.720 20.031 21.600 0.110 25,813.913 589.480 0.367 20.033 20.023 0.538 1.859 Bath*: No. of bathrooms 24,189.953 2,430.933 0.175 9.951 0.000 19,422.995 28,956.912 0.426 0.200 0.140 0.645 1.551 Hbath*: Half bath 7,908.118 2,157.373 0.057 3.666 0.000 3,677.600 12,138.637 0.225 0.075 0.052 0.822 1.217 Deck*: Deck (dummy) 6,081.108 2,266.192 0.039 2.683 0.007 1,637.199 10,525.016 0.100 0.055 0.038 0.947 1.056 Patio: Patio (dummy) 8,110.219 4,970.953 0.023 1.632 0.103 21,637.613 17,858.050 0.032 0.033 0.023 0.997 1.003 CE Test Variables InvSqd54: Squared distance from 21.27E-008 0.000 20.022 20.945 0.345 0.000 0.000 20.008 20.019 20.013 0.352 2.837 the nearest CE property ViewInteract54: Interaction of 9.61E-011 0.000 0.018 0.778 0.437 0.000 0.000 0.002 0.016 0.011 0.353 2.832 view and squared distance View54:View to CEs 20.181 15.912 0.000 20.011 0.991 231.383 31.021 0.060 0.000 0.000 0.974 1.026 Neighborhood Variables: Census Blkgroup Level hous dens *: Housing density 22,485.362 439.239 20.087 25.658 0.000 23,346.691 21,624.033 20.131 20.115 20.080 0.847 1.181 prc black: Percentage of blacks 2329.676 187.668 20.026 21.757 0.079 2697.686 38.333 20.117 20.036 20.025 0.879 1.137 md hs val*: Median house price 0.305 0.045 0.112 6.857 0.000 0.218 0.393 0.377 0.139 0.097 0.749 1.336 Notes: The dependent variable is saleprice, n 5 2,406, df 5 18. * Significant at p , .05. 3 8 M i t t a l Exhibit 8 u CE Model Summary Change Statistics Std. Error of 2 2 2 Model R R Adj. R the Estimate R Change F Change df1 df2 Sig. F Change 1 .768 .59 0.584 46,009.69 .59 97.87 18 1,225 .000 Notes: Predictors: (Constant), md hs val, sl 2008, view, age, patio, InvSqd45, hbath, deck, lotsf, sl 2007, hous dens , beds, prc black, bath, sl 2005, tula, qual, ViewInteract45. The dependent variable is saleprice. weights, t values, and the significance levels of variables used in the model are provided in Exhibit 9. All variables used to signify the physical features of SFD were statistically significant with the right sign within p , .05. Similarly, all the neighborhood variables had the right signs and were found significant within p , .02. All three CE test variables were found significant, but at p , .09, as presented in Exhibit 9. Two of the three CE test variables were significant within alpha 5 0.01 and the view variables were significant with alpha 5 0.09. The model output summary for these variables is: for dij 26, b 5 20.01, t 5 22.82, and p , .01; for Vij 26, b 5 260.64, t 5 21.69, and p 5 .09; and for the interaction effect variable of view and distance, b 5 0.07, t 5 2.63, and p , .01. Except for the view variable, the beta signs for the three test variables were as expected. The proximity variable dij 26 is highly significant at p , 0.01. The distance variable has a negative sign as expected. So, for every unit change in the house sample, as measured by squared distance from the CE-protected parcel, the average SFD price reduces by 0.01. This means that if the home is abutting the CE-protected property, the price will be the highest; however, if the home is 10 feet away from the CE-protected property, the price will decline by $1, while if it is 100 feet away, the price declines by $100. Similarly, if a SFD house is 500 feet away, the average price reduction will be $2,500, holding all other variables constant. The view variable has a negative beta sign, which means that the greater the value of visible vertices, the house price reduces. This is not very intuitive, but can only be explained that more CE parcel vertices can only be seen if a house sample is located so that it can see most. This situation can only occur if the house sample has a higher elevation and can see several CE properties, but may be from a farther distance from the CE parcels. In the original regression data set with a view (mean 5 57, std. dev. 5 65), meaning that an average home in the house sample set views 57 vertices of the CE-protected properties, where an average CE- protected property had average of 10 vertices (a few properties were irregular in shape, or very small or large in size), the average house in the dataset can see about six protected properties. Note that view as measured using the GIS viewshed V a l u e C a p i t a l i z a t i o n E f f e c t o f P r o t e c t e d P r o p e r t i e s 3 9 J O S R E V o l . 6 N o . 1 – 2 0 1 4 u u Exhibit 9 u CE Model: OLS Output with 1,244 Houses within 0.5 Mile Distance from CE-protected Properties for 2005–2008 Collinearity Unstd. Coeff. Std. Coeff. T Sig. 95% Confidence Interval for b Correlations Statistics Model b Std. Error b Lower Bound Upper Bound Zero-order Partial Part Tolerance VIF Constant 284,536.32 19,984.42 24.23 0.000 2123,743.79 245,328.85 Year of Sale (dummy) sl 2005* (Sale Year 2005) 7,224.75 3,469.35 0.047 2.08 0.038 418.21 14,031.28 0.120 0.059 0.04 0.65 1.53 sl 2007* (Sale Year 2007) 212,758.99 3,629.52 20.078 23.52 0.000 219,879.76 25,638.22 20.026 20.100 20.06 0.67 1.49 sl 2008* (Sale Year 2008) 241,136.84 4,138.80 20.214 29.94 0.000 249,256.75 233,016.92 20.183 20.273 20.18 0.72 1.38 Physical Features of Houses Tula*: Total utilizable area 44.19 3.73 0.360 11.86 0.000 36.88 51.50 0.604 0.321 0.22 0.36 2.76 Lotsf *: Lot area sq. ft. 1.02 0.15 0.134 6.81 0.000 0.73 1.31 0.326 0.191 0.13 0.86 1.16 Qual*: Quality of house 5,244.72 539.65 0.306 9.72 0.000 4,185.98 6,303.46 0.396 0.268 0.19 0.33 2.95 Age*: Age of house 157.07 13.58 0.306 11.57 0.000 130.43 183.72 0.018 0.314 0.21 0.47 2.09 Beds: No. of bedrooms 25,364.50 2,128.67 20.066 22.52 0.012 29,540.74 21,188.25 0.389 20.072 20.05 0.49 2.02 Bath*: No. of bathrooms 26,712.63 3,142.34 0.204 8.50 0.000 20,547.66 32,877.59 0.494 0.236 0.16 0.58 1.72 Hbath*: Half bath 8,438.58 2,783.20 0.062 3.03 0.002 2,978.22 13,898.94 0.225 0.086 0.06 0.81 1.23 Deck*: Deck (dummy) 7,922.04 2,941.35 0.051 2.69 0.007 2,151.40 13,692.68 0.132 0.077 0.05 0.92 1.08 Patio: Patio (dummy) 12,913.64 6,388.96 0.037 2.02 0.043 379.12 25,448.15 0.055 0.058 0.04 0.98 1.01 CE Test Variables InvSqd26*: Squared distance 20.01 0.002 20.060 22.82 0.005 20.01 20.002 20.097 20.080 20.05 0.74 1.33 from the nearest CE property ViewInteract26*: Interaction of 0.07 0.03 0.090 2.63 0.009 0.02 0.12 0.091 0.075 0.05 0.28 3.49 view and squared distance View26***: View to CEs 260.64 35.83 20.056 21.69 0.091 2130.92 9.65 0.082 20.048 20.03 0.31 3.21 Neighborhood Variables Census Blkgroup Level hous dens Housing Density* 22,973.46 579.856 20.104 25.13 0.000 24,111.08 21,835.84 20.148 20.145 20.09 0.81 1.23 prc black*: Percentage of blacks 2892.49 286.053 20.064 23.12 0.002 21,453.69 2331.28 20.131 20.089 20.06 0.79 1.25 md hs val*: Median house price 0.19 0.061 0.065 3.05 0.002 0.07 0.31 0.364 0.087 0.06 0.73 1.36 Notes: The dependent variable is saleprice; n 5 1,244, df 5 18. * Significant at p , .05. **5 Insignificant. ***5 Significant only at p , 0.10. 4 0 M i t t a l technique, which only generates binary view information—view is available or not available—it does not account for the quality of view. It also does not account for how far away the view generating amenity is. The view variable was developed in ArcGIS Spatial Analyst using the DEM, as discussed above. The view sheds were created from the vertices of CE properties (Edge) to capture if sample houses can see those vertices or not. This view shed also accounted for the impeding view effect due to topography and building heights, if any. The combined view shed raster provided ‘‘0’’ value for invisible areas and higher values ranging from 1 to 647 for visible areas. The interaction effect variable ViewInteract26 had a positive sign with b 5 0.07, t 5 2.63, and p , .01. This variable signified the importance of both the distance to CE from SFD houses and the visibility to the CE-protected property from the SFD houses. This means that by increasing this interaction variable by one unit, on average, the home price increases by $60, holding all other variables constant. The ViewInteract26 variable is View x Sqdist26. Even if the property is abutting a CE-protected property or as close as 10 feet away, the impact of 0 view is dramatic, with the price effect of 0. This indicates that distance does not matter, meaning, it is not simply close proximity to protected property that creates value, but being able to see and enjoy the view of that property is important too. C o n c l u s i o n From the OS and CE Models, it is clear that the perpetually conserved CE properties offer passive amenity effects that are unlike the mixed-bag of open spaces in Worcester, which also include some active activities within them. Based on the higher prices obtained for houses sold in areas surrounding the CE- protected properties, it appears that people place an economic price on properties with quieter, everlasting landscapes versus those that support more active recreation, which is likely to generate more noise and traffic. The findings reveal that home prices increase the closer the properties are to CE- protected properties. This price elevation is due to the ‘‘amenity magnet’’ effect that an environmental amenity generates in the CE-protected properties. Further, as the measure of proximity was defined by a squared distance term, it shows that the home price effect reduces more rapidly as the distance from the CE-protected property increases. So, for example, holding all other variables constant, if a home is 10 feet away from a CE-protected property, the price will decline by $1. Similarly, if it is 100 feet away, the home price declines by $100. If it is 500 feet away, the average home price declines by $2,500. If it is 1,000 feet away, the average home price declines by $10,000. The view variable is insignificant at p , .05. The home price increase with the interaction of visibility and distance from CE-protected properties is very important. Even if a property abuts or is within 10 feet of the CE-protected property, the absence of a view means there will be no positive impact of the amenity on the sales price. In Worcester, CE-protected properties had aesthetic, passive recreational, and bio- diversity value. Some of these CEs also provided a buffer to create habitat for V a l u e C a p i t a l i z a t i o n E f f e c t o f P r o t e c t e d P r o p e r t i e s 4 1 wildlife, which included natural landscape features and provided support for other associated ecosystem values such as water purification, reduction in river pollution, and flood control, etc. The preservation of open spaces benefits the people living in the region. However, landowners nearest the preserved parcels receive extra direct benefits, which are capitalized into the prices of their SFD houses. Theoretically, as the size of the open space increases, the range of it externality impact should increase as well. However, the accessibility index used in the model was insignificant. This could be due to the large range of acreage among the protected properties in the study (maximum 487 acres and minimum 1 acre). Also, if the house samples are smaller in size (square footage) and do not have their own private open space (small lot), they would price public open spaces more than the larger size houses. It would be interesting to explore how spatially grounded models could refine our understanding of the impact of CE-restricted properties on home prices. The findings support the notion that house buyers and sellers place a higher price on quieter, everlasting conserved landscapes of CE parcels as compared to more active and relatively louder open recreational spaces around the mixed-bag open spaces. The surrounding houses become desirable because of the protected viewsheds provided by adjacent CEs making some home sites more expensive, which in turn provides additional taxes to the local authority, income for investors, and neighboring landowners (Brewer, 2004; Fairfax et al., 2005; Morris, 2008; Aspen Valley Land Trust). E n d n o t e s Loomis, Rameker, and Seidl (2004) discuss the cost benefits and fiscal advantages of publicly funded protected land. Conservation Easements—Fact vs. Fiction, The Nature Conservancy, accessed March 12, 2011: http: / / www.nature.org / aboutus / privatelandsconservation / conservationeasements / conservation-easements-fact-versus-fiction.xml. Bourassa, Hoesli, and Sun (2004) provide a chronological review of 35 studies that have used view as a variable in measuring externality impact on home values. These studies and their findings are tabulated in a six-page summary (pp. 1431–36). Only homes sold within a 0.5 mile buffer area from the environmental amenities were used for this research. This 0.5 mile buffer was chosen to control for any other spatial factors (variable not accounted for in the model) that may affect the house prices and the buffer area functions as a local housing submarket around the amenity. R e f e r e n c e s Acharya, G. and L.L. Bennett. Valuing Open Space and Land-Use Patterns in Urban Watersheds. Journal of Real Estate Finance and Economics, 2001, 22:2, 221–37. Anderson, S.T. and S.E. West. Open Space, Residential Property Values, and Spatial. Regional Science and Urban Economics, 2006, 36:6, 773–89. Appleton, J. The Experience of Landscape. Chichester, Sussex London: John Wiley, 1975. J O S R E V o l . 6 N o . 1 – 2 0 1 4 u u 4 2 M i t t a l Asabere, P.K. and F.E. Huffman. Negative and Positive Impacts of Golf Course Proximity on Home Prices. Appraisal Journal, 1996, 64:4, 351–55. Aspen Valley Land Trust. ND. Conservation Easement Appraisal Considerations. Bark, R.H., D.E. Osgood, B.G. Colby, and E.B. Halper. How Do Homebuyers Value Different Types of Green Space? Journal of Agricultural and Resource Economics, 2011, 36:2, 395–415. Benson, E.D., J.L. Hansen, A.L. Schwartz Jr., and G.T. Smersh. Pricing Residential Amenities: The Value of a View. Journal of Real Estate Finance and Economics, 1998, 16:1, 55–73. Billington, R. Federal Leverage Attracts Private Investment at U.S. Heritage Sites: A Case Study. International Journal of Heritage Studies, 2004, 10:4, 349–59. Bolitzer, B. and N.R. Netusil. The Impact of Open Spaces on Property Values in Portland, Oregon. Journal of Environmental Management, 2000, 59:3, 185–93. Bourassa, S.C., M. Hoesli, and J. Sun. What’s in a View? Environment and Planning A, 2004, 36, 1427–50. Boyd, J.W. Location, Location, Location: The Geography of Ecosystem Services. Resources, 2008, Fall, 11–8. Boyd, J.W., K. Caballero, and D. Simpson. The Law and Economics of Habitat Conservation: Lessons from an Analysis of Easement Acquisitions. Resources for the Future, Discussion Paper, 1999, 29–32. Boyle, M. and K. Kiel. A Survey of House Price Hedonic Studies of the Impact of Environmental Externalities. Journal of Real Estate Literature, 2001, 9:2, 117–44. Brewer, R. Conservancy: The Land Trust Movement in America. Lebanon, NH: University Press of New London, 2004. Brenneman, R. and S.M. Bates. Land-Saving Action. Island Press, 1984. Brenner, J.C., S. Lavallato, M. Cherry, and E. Hileman. Land Use Determines Interest in Conservation Easements Among Private Landowners. Land Use Policy, 2013, 35, 24–32. Census 2010. Chasan, R. Multiple Minimum Distance. MultipleMinimumDistance.zip ArcScript for ArcGIS. Available at: http: / / arcscripts.esri.com / details.asp?dbid513193, 2003. Colinvaux, R. The Conservation Easement Tax Expenditure: In Search of Conservation Value. Columbia Journal of Environmental Law, 2012, 37:1, 1–61. Conroy, S.J. and J.L. Milosch. An Estimation of the Coastal Premium for Residential Housing Prices in San Diego County. Journal of Real Estate Finance and Economics, 2011, 42, 211–28. Correll, M.R., J.H. Lillydahl, and L.D. Singell. The Effects of Greenbelts on Residential Property Values: Some Findings on the Political Economy of Open Space. Land Economics, 1978, 54:2, 207–17. Crompton, J.L. The Impact of Parks on Property Values: A Review of the Empirical Evidence. Journal of Leisure Research, 2001, 33:1, 1–31. ——. The Role of the Proximate Principle in the Emergence of Urban Parks in the United Kingdom and in the United States. Leisure Studies, 2007, 26:2, 213–34. ——. Empirical Evidence of the Contributions of Park and Conservation Lands to Environmental Sustainability: The Key to Repositioning the Parks Field. World Leisure, 2008, Research Paper 3, 154–72. V a l u e C a p i t a l i z a t i o n E f f e c t o f P r o t e c t e d P r o p e r t i e s 4 3 Daniels, T. and M. Lapping. Land Preservation: An Essential Ingredient in Smart Growth. Journal of Planning Literature, 2005, 19:3, 316–29. Do, Q. and G.A. Grudnitski. Golf Courses and Residential House Prices: An Empirical Examination. Journal of Real Estate Finance and Economics, 1995, 10:3, 261–70. ——. Adjusting the Value of Houses Located on a Golf Course. Appraisal Journal, 1997, 65:3, 261–267. Fairfax, S.K., L. Gwin, M.A. King, L. Raymond, and L.A. WattFlorax. Buying Nature: The Limits of Land Acquisition as a Conservation Strategy, 1780–2004. Cambridge, MA: The MIT Press, 2005. Geoghegan, J. The Value of Open Spaces in Residential Land Use. Land Use Policy, 2002, 19, 91–8. Geoghegan, J., L. Lynch, and S. Bucholtz. Capitalization of Open Spaces into Housing Values and the Residential Property Tax Revenue Impacts of Agricultural Easement Programs. Agricultural and Resource Economics Review, 2003, 32:1, 33–45. Geoghegan, J., L. Wainger, and N.E. Bockstael. Spatial Landscape Indices in a Hedonic Framework: An Ecological Economics Analysis Using GIS. Ecological Economics, 1997, 23:3, 251–64. Gustanski, J.A. and R.H. Squires. Conservation Easements, Voluntary Actions and Private Lands. In: Protecting the Land: Conservation Easements Past, Present and Future. California: Island Press, 2000. Ham, C., P.A. Champ, J.B. Loomis, and R.M. Reich. Accounting for Heterogeneity of Public Lands in Hedonic Property Models. Land Economics, 2012, 88:3, 444–56. Hidano, N. The Environmental Valuation of the Environmental and Public Policy. North Hampton, MA: Edward Elgar Publishing Inc., 2002. Krupnick, A.J. and J.V. Siikama ¨ki. How People Value What Nature Provides. Resources, 2007, 165, 14–16. Lake, I.R., A.A. Lovett, I.J. Bateman, and B. Day. Using GIS and Large-Scale Digital Data to Implement Hedonic Pricing Studies. International Journal of Geographical Information Science, 2000, 14:6, 521–41. Lake, I.R., A.A. Lovett, I.J. Bateman, and I.H. Langford. Modeling Environmental Influences on Property Prices in an Urban Environment. Computers Environment and Urban Systems, 1998, 22:2, 121–36. Land Trust Alliance. 2010 National Land Trust Census Report—A Look at Voluntary Land Conservation in America. Washington, DC: Land Trust Alliance. Accessed at: http: / / www.landtrustalliance.org / land-trusts / land-trust-census / census on November 21, 2012. Land Trust Alliance. 2005 National Land Trust Census Report. Washington, DC: Land Trust Alliance, 2006. Lansford, N.H., Jr. and L.L. Jones. Marginal Price of Lake Recreation and Aesthetics: A Hedonic Approach. Journal of Agricultural and Applied Economics, 1995, 27:1, 212–23. Le Goffe, P. Hedonic Pricing of Agriculture and Forestry Externalities. Environmental & Resource Economics, 2000, 15:4, 397–401. Loomis, J., V. Rameker, and A. Seidl. A Hedonic Model of Public Market Transactions for Open Space Protection. Journal of Environmental Planning and Management, 2004, 47:1, 83–96. Lutzenhiser, M. and N.R. Netusil. The Effect of Open Spaces on a Home’s Sale Price. Contemporary Economic Policy, 2001, 19:3, 291–98. J O S R E V o l . 6 N o . 1 – 2 0 1 4 u u 4 4 M i t t a l Mahan, B.L., S. Polasky, and R.M. Adams. Valuing Urban Wetlands: A Property Price Approach. Land Economics, 2000, 76:1, 100–13. Malpezzi, S. Hedonic Price Models: A Selective and Applied Review. The Center for Urban Land Economic Research Report. University of Wisconsin. Prepared for Housing Economics: Essays in Honor of Duncan Maclennan, 2002. MassGIS: Office of Geographic and Environmental Information. MassGIS, 2008. Accessed at: http: / / www.mass.gov / mgis / laylist.htm. McConnell, V. and M. Walls. The Value of Open Space: Evidence from Studies of Nonmarket Benefits. Lincoln Institute of Land Policy. Working Paper WP04VM1, 2005. Merenlender, A.M., L. Huntsinger, G. Guthey, and S.K. Fairfax. Land Trusts and Conservation Easements: Who is Conserving What for Whom? Conservation Biology, 2004, 18:1, 65–75. Mitchell, S.B. and B.B. Johnson. Valuing Farmland Conservation Easements. Institute of Agriculture and Natural Resources. University of Nebraska–Lincoln. Accessed on July 24, 2007 at: http: / / www.ianrpubs.unl.edu / epublic / live / g1428 / build / g1428.pdf, 2005. Mittal, J. Extending Land Readjustment Schemes to Regional Scale: A Case of Regional Ring Road via Mosaicking Neighborhood Level Plans. Real Estate Finance Journal, 2013, 30:2, 62–73. ——. Self-Financing Land and Urban Development via Land Readjustment and Value Capture. Habitat International, 2014, 44, 314–23. Morris, A.W. Easing Conservation? Conservation Easements, Public Accountability and Neoliberalism. Geoforum, 2008, 39, 1215–27. National Association of Realtors. NAR survey shows public support for open space depends on use and cost. Press release, April 25, 2001. National Heritage Area. What are National Heritage Areas? Accessed on July 22, 2013 at: http: / / www.nps.gov / history / heritageareas / FAQ / . Nelson, A.C., J.B. Duncan, C.J. Mullen, and K.R. Bishop. Growth Management Principles and Practices. Chicago, IL: American Planning Association, 1995. Netusil, N.R. The Effect of Environmental Zoning and Amenities on Property Values: Portland, Oregon. Land Economics, 2005, 81:2, 227–46. Palmquist, R.B., F.M. Roka, and T. Vukina. Hog Operations, Environmental Effects, and Residential Property Values. Land Economics, 1997, 73:1, 114–24. Pooler, J. Measuring Geographical Accessibility: A Review of Current Approaches and Problems in the Use of Population Potentials. Geoforum, 1987, 18:3, 269–89. Sander, H.A. and S.M. Manson. Heights and Locations of Artificial Structures in Viewshed Calculation: How Close is Close Enough? Landscape and Urban Planning, 2007, 257–70. Sander, H.A. and S. Polasky. The Value of Views and Open Space: Estimates from a Hedonic Pricing Model for Ramsey County, Minnesota, USA. Land Use Policy, 2009, 26, 837–45. Shultz, S. and N. Schmitz. View Shed Analyses to Measure the Impact of Lake Views on Urban Residential Properties. The Appraisal Journal, 2008, 76:3, 224–32. Simons, R.A. The Effect of Pipeline Ruptures on Non-contaminated Residential Easement- Holding Property in Fairfax County. The Appraisal Journal, 1999, July, 255–63. Simons, R.A. and J.D. Saginor. A Meta-analysis of the Effect of Environmental Contamination and Positive Amenities on Residential Real Estate Values. Journal of Real Estate Research, 2006, 28:1, 71–104. V a l u e C a p i t a l i z a t i o n E f f e c t o f P r o t e c t e d P r o p e r t i e s 4 5 Sirmans, G.S., D.A. Macpherson, and E.N. Zietz. The Composition of Hedonic Pricing Models. Journal of Real Estate Literature, 2005, 13:1, 3–43. Song, Y. and G-J. Knaap. Measuring the Effects of Mixed Land Uses on Housing Prices. Regional Science and Urban Economics, 2004, 34, 663–80. Standiford, R.B. and T.A. Scott. Value of Oak Woodlands and Open Space on Private Property Values in Southern California. Special issue Towards the New Forestlands Commercial and Environmental Benefits Accounting: Theories and Applications. P.C. Papacin (ed.), 2001. The Trust for Public Land and Land Trust Alliance. Americans Invest in Parks and Conservation. LandVote 2008. Accessed on February 19, 2010 at: http: / / www. landtrustalliance.org / policy / documents / landvote-2008. Thompson, B.H. The Trouble with Time: Influencing the Conservation Choices of Future Generations. Natural Resources Journal, 2004, 44:2, 301–20. Troy, A. and J.M. Grove. Property Values, Parks, and Crime: A Hedonic Analysis in Baltimore, MD. Landscape and Urban Planning, 2008, 87:3, 233–45. Walsh, P.J., J.W. Milon, and D.O. Scrogin. The Spatial Extent of Water Quality Benefits in Urban Housing Markets. Land Economics, 2011, 87:4, 628–44. Whyte, W.H. Securing Open Space for Urban America: Conservation Easements. In: The Essential William H. Whyte. A. LaFarge (ed.). New York: Fordham University Press, 2000. Wolverton, M.L. Empirical Study of the Relationship between Residential Lot Price, Size and View. Journal of Property Valuation and Investment, 1997, 15:1, 48–57. Wright, J.B. Designing and Applying Conservation Easements. Journal of the American Planning Association, 1994, 60:3, 380–88. Jay Mittal, Iowa State University, Ames, IA 50011 or mittalphd@yahoo.com. J O S R E V o l . 6 N o . 1 – 2 0 1 4 u u This page intentionally left blank
Journal
Journal of Sustainable Real Estate
– Taylor & Francis
Published: Jan 1, 2014
