Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Non-spatial Contagion in Real Estate Markets: The Case of Brookland Greens

Non-spatial Contagion in Real Estate Markets: The Case of Brookland Greens Non-spatial Contagion in Real Estate Markets: The Case of Brookland Greens A u t h o r s Kimberly Winson-Geideman, Andy Krause, Hao Wu, and Georgia Warren-Myers A b s t r a c t We investigate contagion in real estate markets by evaluating the effects of a widely publicized landfill contamination event in one local market on the price of homes near landfills in non-impacted markets within the same metropolitan region. The impact of proximity to open, closed, and redeveloped landfills in the directly affected and contagion neighborhoods is tested at distances varying from ,500 meters to 2,500 meters using the traditional hedonic pricing model. The results are mixed and relative to the current use of the landfill as closed, capped, and redeveloped landfills show no impact. Sites that are capped yet undeveloped and sites with open fills appear to show some impact, although further research is needed to support any contagion effects. There is a substantial amount of research quantifying the relation between proximity to environmental (dis)amenities and value. Amenities such as golf courses (Do and Grudnitski, 1995; Shultz and Schmitz, 2009), scenic views (Benson, Hansen, Schwartz, and Smersh, 1998; Hindsley, Hamilton, and Morgan, 2013), and green space (Bolitzer and Netusil, 2000; Cho, Bowker, and Park, 2006) positively affect value while negative effects are shown in proximity to landfills (Reichert, Small, and Mohanty, 1992; Hite, Chern, Hitzhusen, and Randall, 2001), high traffic noise (Nelson, 2008), and environmental contamination (Simons and Saginor, 2006) such as oil spills. These relations are based on a two-tier spatial component, as well as the principle of substitution (i.e., the value of one property is directly related to the value of adjacent or similar properties, all else held equal). The first spatial component is a direct effect wherein a property is immediately adjacent to or directly affected by the (dis)amenity. The other spatial component links the value of unaffected properties to the change in value of nearby affected properties. Therefore, when a property abuts another that is directly affected by an environmental (dis)amenity, its value is impacted, although likely not to the same degree as the directly impacted property. In this paper, we provide further evidence of these phenomena by investigating the case of a housing estate, Brookland Greens Estate, and an adjacent landfill located near Melbourne, Australia. We test the spatial relations between properties directly and indirectly affected by the issues associated with the landfill. What we seek to address is not the spatial relation between properties per se, but rather how and to what extent the diminution in property values spreads to non-spatially N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 2 3 related, similar markets. The framework for this is based in contagion theory where an event in one market spills over to another that has little or no direct spatial relation to the first. We hypothesize that a detrimental event associated with a particular type of land use (landfill in this case) and evidenced in a diminution in the value of nearby properties may have a ‘‘contagion effect’’ in which real estate proximate to other land uses of this type, although not proximate to the initial event, will experience similar phenomena, even though they are not directly affected by the detrimental event (i.e., non-spatial market contagion). The key mechanism tested is the ability of a distant (spatially) event to raise awareness of the potential negative externalities of an existing disamenity and how this contagion may affect observed market outcomes. While the results are inconclusive, they do suggest a mild contagion effect in some of the submarkets, particularly where the landfill has not been redeveloped. We next discuss the literature on contagion theory, as well as the spatial and non- spatial effect of social (dis) amenities on observed market behavior. A description and analysis of the landfill case and its associated regulatory environment follows. After explaining the methodology and data, the main findings are presented on the direct price impact and possible contagion effects on capped and redeveloped landfills. We conclude with further interpretation and a discussion on future research. L i t e r a t u r e This research is framed in contagion theory—unexposed markets are subject to impacts similar to an affected market by virtue of that similarity—leading us to make a few assumptions about the cause of those effects. First, we assume information transparency and transmission in residential markets (e.g., time lag and market reaction) is measured by price change. We also assume there are at least two types of ‘‘spatial reactions’’: (1) impact at the landfill area, the continuous spatial contagion effect; and (2) landfill areas in other regions, the non- spatial (discrete) contagion effect. A third reaction, not addressed in this research, is (3) a contagion effect on contaminated sites other than landfills, reflecting human ability to relate events (e.g., by imagination). If information transmission efficiency is the dominant force, there will be the negative price relation of (1) . (2) . (3), or else other market or non-market forces or behavioral theories may exist to explain the contagion effect. We also assume that there is likely a diminution in value associated with the stigma of living near a landfill that is already capitalized into the market value of the properties. Stigma refers to the real or perceived, readily identifiable issues associated with landfills or other contamination sources that may affect property value. Consequently, it is important to understand the level at which a community is aware of the disamenity and its implication for occupants and future home purchasers. We posit that when a contagion event occurs, stigma effects, whether J O S R E V o l . 9 2 0 1 7 u u 2 4 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s present or not, will be magnified or will emerge relative to proximity alone, thus we do not control for directional heterogeneity as suggested in Cameron (2006). The literature review that follows is rooted in these assumptions. We begin with an overview of contagion theory, followed by studies that investigate the links between stigma, location, and house prices from spatial, non-spatial, and social justice perspectives. Contagion Theory According to contagion theory, groups of people, by virtue of communication with the social networks they engage in, are likely to adopt the attitudes and behavioral norms of the group. Just as a virus spreads through personal contact, so do attitudes, behaviors, and actions. Efforts to influence do not have to be present; the mere existence of communication portals is enough. These links are well established in the literature beginning with the Ryan and Gross (1943) study of hybrid seed corn and continuing through today with a vast number of studies conducted in the financial markets (e.g., Ghosh, Guttery, and Sirmans, 1998; Mink and de Haan, 2013; Caccioli, Shrestha, Moore, and Farmer, 2014; Chen, Tsai, Sing, and Yang, 2015). Contagion may be transmitted through social or public (e.g., media) channels. Contagion studies in the real estate and other financial literature typically follow a predictable pattern and often focus on negative economic or other shocks and the resulting behaviors, actions or attitudes of individuals, organizations or communities. When an event occurs, information flows through communication networks, elevating levels of perceived risk. As perceived risks increase and information flows from person to person, the ‘‘group belief’’ reinforces or intensifies. The group may then engage in illogical, emotionally-driven activities they would not necessarily engage in as individuals, primarily due to the anonymity the group provides. The fluidity of communications within existing social networks helps create and refine new behaviors, attitudes, and actions that become socially acceptable, if only for limited time. The rapid pace at which movements, ideas, and / or activities can spread has the potential to impact markets on a global scale, particularly given recent technological improvements in communication. Research by Seiler, Collins, and Fefferman (2013) shows that the number of strategic mortgage defaults during the Global Financial Crisis (GFC) were likely exacerbated by the suggestion that underwater homeowners exercise their put option, further depressing home values and predicating the market collapse. Prior to the GFC, most defaults were economic, the result of job loss or divorce, for example. Strategic defaults, conversely, were based on the mortgagor’s willingness to ‘‘turn over the keys’’ to the lending institution because the value of the mortgage was greater than that of the collateral. Strategic defaults were comparatively rare before that time. Freybote and Fruit’s (2015) study of media coverage in the case of the construction of a natural gas pipeline provides additional evidence of how information networks N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 2 5 contribute to market contagion. The authors investigate how residential properties are impacted by proximity to the pipeline and how unrelated pipeline explosions in different markets affect value during the subject pipeline construction process. Consistent with prior literature, they find that proximity to the pipeline negatively affected value. Further, they also determined that the perceived riskiness of proximity to natural gas pipelines increases in months when a fatality associated with an explosion is widely reported in the media, and prices near the pipeline under construction moderate. Stigma, Location, and House Prices The literature on the effects of environmental disamenities on value is lengthy and well established with studies focused on landfills (Nelson, Genereux, and Genereux, 1992; Gunterman, 1995; Hite, Chern, Hitzhusen, and Randall, 2001), Superfund sites (Michaels and Smith, 1990; Kohlhase, 1991; Kiel, 1995; Kiel and McClain, 1995; Kiel and Zabel, 2001), and leaking underground storage tanks (Sementelli and Simons, 1997; Simons, Bowen, and Sementelli, 1999; Simons, Levin, and Sementelli, 2009), for example. In general, these studies show a significant negative effect associated with proximity that decreases as distance from the source of the contamination increases. Studies showing little to no effect are typically site specific or address less intense uses, as shown in Ready’s (2010) research on low-volume landfills. Others show the effects are unequally distributed based on direction (Cameron, 2006) or property type (Ihlanfeldt and Taylor, 2004). Empirical inquiry into the relation between environmental hazard, stigma (e.g., landfill), and land prices fall under (1) a spatial perspective or (2) a non-spatial perspective. The former links to mainstream location (urban economic) theory, while the latter links more closely to information economics and behavioral ideas. A third stream of analysis concerns normative strategy and decisions such as location choice, social justice and compensation determination, data, methodology, and the cross-context applicability of results. Spatial Perspective Brasington and Hite (2005) estimate spatial effects using the demand price elasticity of homes in relation to the source of pollution. They find that ‘‘point- source pollutants’’ and nearby house prices are negatively correlated, while income, education, and concern for children are sensitive to the location of polluted site. Eshet, Baron, Shechter, and Ayalon (2007) estimate the economic value of disamenities by measuring the spatial impact of waste transfer stations on local property prices. In addition to the expected negative relation, they also raise issues on the measurement of externalities and the hedonic pricing model. Hanna (2007) focused on the dynamic relation of a polluting industrial site, house price, and household mobility as determined by income, finding that income and pollution are strongly correlated. Cameron (2006) argues the existence and significance of directional heterogeneity and its role in determining the actual spatial price effect of environmental hazard. J O S R E V o l . 9 2 0 1 7 u u 2 6 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s Non-spatial Perspective McCluskey and Rausser (2003a) suggest the possibility of short-term and long- term stigma due to discovery and clean-up associated with the hazardous site, emphasizing the relation between household income, discounted property values, and social segregation. Messer et al. (2006) discuss the significance of stigmatization in the case of clean-up action on environmentally damaged sites, specifically focusing on the impact of people’s memory in forming persistence perceptions over a long period of time. Messer’s study exposes an interesting hypothesis: If cleanup is treated as an investment to counter-act existing social cost (e.g., land pollution) and the build-up of negative perception (e.g., stigma), the actual level of damage and price behavior is likely dependent on the supply elasticity of suitable replacement sites. Reichert (1997) finds that polluted landfills tend to have liquidity, distance, and permanent devaluing effects on associated property markets. McCluskey and Rausser (2003b) suggest that clean-up actions on hazardous sites may not be as effective as new information about an environmental issue in triggering change in behavior and expectation in consumer valuation. Their findings support earlier research by Kiel and McClain (1995). Social Justice and Distribution Perspective Palmquist and Smith (2001) evaluate the use and effectiveness of the hedonic technique to evaluate environmental policy and litigation. They suggest market price based models are less directly relevant to public policy but have direct relevance to litigation that is empirical and fact-based. Simons and Winson- Geideman (2005) applied the contingent valuation method to understand buyer perceptions of hazardous sites on residential property. Caplan, Grijalva, and Jackson-Smith (2007) investigate the extent that a non-host community compensates a host community for the actual and perceived impacts of a landfill. They test and argue the value of using the resource-resource format for landfill site evaluation and decision-making, because it allows the parties to be aware of and compare alternative resource trade-offs and suitable forms of compensation for a host community. In the Australian context, Bond (2001) and Chan (2001, 2002) investigate the role of stigma and associated risk perceptions in the valuation of contaminated land in Australia’s capital cities. They identify risk factors and decision models to approach landfill site evaluation. Wu and Chen (2012) use a structural-agent approach to investigate the legal framework and process that Melbourne has implemented for its inner city brownfield sites, leaving specific investigation of the relation between environmental hazard and house price (e.g., from spatial or non-spatial perspective) unattended. In this study, we fill this gap by paying special attention to the contagion effect within a specific legal policy framework in Melbourne, Australia, a major city that is frequently regarded as having high- quality living standards and property environments. N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 2 7 T h e B r o o k l a n d G r e e n s C a s e a n d C o r r e s p o n d i n g R e g u l a t o r y E n v i r o n m e n t The Brookland Greens case began in 1992 when the Shire of Cranbourne (eventually becoming the City of Casey in 1994) approved the development of a landfill located on Stevenson’s Road, about 30 kilometers from the Melbourne CBD. The area was generally rural and undeveloped and the landfill approved as a Type II by the Environmental Protection Agency (EPA) to accept putrescible waste, which is biodegradable and produces landfill (methane) gas (see Exhibit A1 in the Appendix for details regarding landfill types). Under the terms of the Environmental Protection Act of 1970, state governments are vested with the right to govern the construction and operation of landfills, and it is the EPA’s responsibility to monitor and regulate them. The EPA grants permits for new sites, examines them for compliance, and then issues a license to operate when the work is complete. The license includes conditions for operation, limits on discharge, maintenance, reporting, and monitoring with the objective of minimizing any negative impact on the environment (Brouwer, 2009). Councils (local government) generally have the responsibility of issuing management contracts, thus they also hold a level of responsibility for the operation of the landfills. Records show that, in order to contain the methane gas at the Stevenson’s Road site, the EPA intended for the landfill to be lined with compacted clay, but through a series of missed deadlines and poor negotiations, the liner was not installed. Construction was completed and a license to operate was granted regardless. Operations began on the site in June 1996 (Brouwer, 2009). In 1999, Peet & Co., a residential development company, began construction of a housing development known as Brookland Greens Estate on land adjacent to the landfill. In early 2000, the developer petitioned the city to have parts of the site rezoned to Residential Zone 1 to facilitate the construction of additional housing units. At the same time, the developer was engaged in discussions with the EPA regarding the appropriate buffer distance required to minimize the impacts of the fill to nearby home owners. After a number of panel hearings debating 200-meter and 500-meter distances, the EPA panel sided with the developer and supported the 200-meter buffer for Brookland Greens. Incidentally, in 2001, the EPA published the Best Practices Environmental Management, Siting, Design, Operation and Rehabilitation of Landfills, recommending a 500-meter distance for all other fills. The City of Casey complied with the 200-meter request, noting that no homes were to be built within that distance. Not to be deterred, the developer approached the Victorian Civil and Administrative Tribunal (VCAT) for permission to construct stage 10 on land that was partially located within the 200-meter landfill buffer. The proposal was approved by VCAT in May 2004, who ruled that the 200-meter buffer distance was to be measured from the ‘‘active tipping area,’’ J O S R E V o l . 9 2 0 1 7 u u 2 8 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s being the ‘‘active tipping face’’ of the landfill. This ruling effectively allowed the developer to build up to the perimeter of the fill (Brouwer, 2009). However, when the required disclosures were made for the subsequent sale of Brookland Greens homes, the site was referred to as ‘‘sand extraction operations’’ and not as a landfill. In June 2005, the landfill ceased operations and less than a year later the EPA received an anonymous complaint about ‘‘bubbling storm water puddles’’ on the Brookland Greens Estate development site. Initial tests found that methane gas and carbon dioxide had entered the aquifer and was migrating from the landfill into the estate. A post-closure pollution abatement notice was issued, taking effect on February 2, 2007, and the waste discharge license was revoked later that month (Brouwer, 2009). Testing began at houses within the estate, and on August 31, 2008, methane gas at 63% volume for volume of air was found in one of the homes. On September 9 the residents were declared to be in imminent danger and the County Fire Authority was made responsible for leading the emergency response. Twenty-nine families were instructed to evacuate their homes immediately for fear they would explode. As word spread and the media began reporting the story, the EPA refused to release any information on which of the 35 other landfills in the metropolitan region were lined and officially put all of them on notice to ‘‘clean up their act.’’ (The Age, September 21, 2008). While publicly stating that they were not aware of similar incidents affecting residential neighborhoods, the EPA began conducting reviews at all other Victorian landfills to see if they posed a significant risk to any neighboring properties (Cooper, 2008). Approximately 300 homes in the Brookland Greens Estate were tested for the gas, and the danger level reduced on October 31, 2008 when the risk to residents was deemed to be at tolerable levels. Unsurprisingly, a class action lawsuit was filed against the City of Casey, as well as the EPA. A settlement was reached in March 2011 for AU$23.5 million; $13.5 million paid by the city and $10 million by the EPA and $6 million of the total covered legal fees. The 771 residents received payments ranging from $7,000 to $132,000 depending on the severity of the impact to their property, an average of $22,373 each (Schulz and Barber, 2011). A schedule of the incidents is shown in Exhibit 1. M e t h o d s We test for two potential causes of price impacts. Firstly, the analysis of direct impacts to property values surrounding Brookland Greens that resulted from the methane gas leak in August 2008. We then test to see if any indirect or contagion effects are felt at the other landfills that were publicly put on notice regarding potential impacts from methane. As homes surrounding the other landfills were not directly impacted by the initial methane leak in Brookland Greens, any impact on prices surrounding these sites may be attributable to non- spatial contagion. N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 2 9 Exhibit 1 u The History of the Brookland Greens Estate Year Incidents and Actions 1996 Landfill at Stevenson’s Road operating as a landfill. 1999 Peet & Co. Casey Land Syndicate Limited (PEET) began construction of Brookland Greens Estate. 2000 Amendment to planning scheme to rezone land to Residential 1 to facilitate development of the estate; and the Section 173 agreement between Council and PEET stipulated no homes to be built within a 200-meter buffer of the landfill. 2004 Victorian Civil and Administrative Tribunal (VCAT) hearing that PEET seeking to develop Stage 10 of the estate, and this land was situated within the 200m landfill buffer. VCAT approved the reduction of the buffer along the western side of the landfill boundary. 2005 Landfill ceased operation. March 2006 EPA notified of bubbling stormwater and testing confirmed landfill gas containing methane and carbon dioxide, which was migrating from the landfill into the estate. EPA issued a pollution abatement notice. August 2008 Home in Brookland Greens was detected as having Methane gas detected at 63% volume for volume of air. September 2008 Residents in the Brookland Greens Estate were evacuated by the Country Fire Authority as the EPA had advised that the methane gas levels posed an imminent danger to residents. October 31, 2008 The risk levels were downgraded and residents allowed to move back into their homes. September 2008 Complete cessation of demand for rental accommodation in the Brookland Greens Estate was noted; future residents who were building on their land either ceased building activity or suspended construction contracts; businesses operating in homes were unable to continue; residents were unable to obtain or renew policies of insurance for their homes or property in Brookland Greens Estate. 277 homes across the Brookland Greens Estate had landfill gas/methane gas-in house monitoring installed; physical works to 252 homes in the estate were undertaken to mitigate the migration of land fill gas; and physical works comprising the construction of a bentonite diaphragm wall on the perimeter of the landfill (completed September 2009). November 2008 Slater and Gordon Lawyers lodged class action on behalf of residents. May 2009 Mediation was conducted at the County Court and also the Federal Court. February 2011 Date of hearing 1 February 2011 and Date of Ruling, February 3, 2011. March 2011 Residents win $23.5 million AUD compensation for the gas leak fiasco, $10 million to be paid by the EPA and $13.5 million to be paid by Casey Council. Residents share $17.25 million after $6 million payment for court costs and legal fees. This equates to approximately payouts from $7,000 to $132,000 for the worst affected households. Note: The sources are Brouwer (2009) and Wheelan v. City Council of Casey & Ors (No. 3). 2011. J O S R E V o l . 9 2 0 1 7 u u 3 0 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s Further, we test two different types of landfills: (1) closed and capped and (2) open and active. The sites are distributed throughout the Melbourne area, some in gentrifying neighborhoods, others near industrial areas, and some in more rural parts. While closed, capped, and redeveloped make up the bulk of our analysis, one of the capped, Narre Warren, consisted of undeveloped vacant land at the time the notice was issued. Further, it is important to note that the Brookland Greens case is something of an anomaly in the Melbourne area—the landfill had ceased operations when the contamination was discovered but still possessed a license to operate. The short time frame between those events (about nine months) coupled with homes being within 200 meters of an active, operating landfill for several years prior to the cessation of operations contributes to the unique qualities of the case. Because we recognize that comparing Brookland Greens to closed and capped fills, some of which provide amenities such as parkland to local communities, may not sufficiently address the pertinent questions, evaluation of active landfills is also required. The problem with active landfills is that few are positioned within 500 meters of a residential neighborhood, the minimum distance recommended by the EPA in their 2001 report. This is to be expected; most developers and homebuyers are wary of investing in properties that may be subject to environmental disamenities, as previously discussed. That said, one site on the list of 35, the Clayton South landfill, is positioned appropriately for analysis. Clayton South is about 32 kilometers from Brookland Greens and is also located in the southeast region of Melbourne. The area is generally less rural than the City of Casey, but would not be considered densely developed. We use a hedonic modeling framework to test for both the direct and contagion effects. The hedonic model is the method traditionally used to determine the impact of one or more characteristics on the price of housing, thus it is the preferred method for this research. It has been used widely to establish the value of a variety of (dis)amenities since Rosen’s (1974) seminal work on the subject. We use it to compare the prices of homes within 500 meters—the notification distance and the distance recommended by the EPA to minimize the risk of exposure to ‘‘dust, noise, windblown litter, and other emissions in relation to the landfill site (EPA, 2001)—to those between 500 meters to 2,500 meters of the landfills. As there are differences in prices between the areas near to the landfills and those farther away prior to the leak and the subsequent announcement, we approach the analysis through a before–after comparison. First, differences in prices between the affected (,500 m) and the control (500 m , 2,500 m) for the period before the leak are estimated. This pre-existing difference is then subtracted from any observed difference post-leak to determine the change in price differentials (if any) due to the leak and / or non-spatial contagion. As an example, if the homes around landfill X sold for 3% less prior to the leak and 12% less after the leak, we would conclude a change in value, or an impact due to the leak of 29%. Both the pre-leak (before) and post-leak (after) periods cover a 32-month span. The before period is from January 1, 2006 to August 31, 2008 and the after period is from September 1, 2008 to April 20, 2011, the date of the Brookland Greens legal settlement. N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 3 1 We use a traditional semi-log hedonic pricing model approach: ln( price) 5 a 1 bX 1 g t 1 lL 1 «, (1) where price is the observed sales price, X is a vector of physical home characteristics such as number of bedrooms and lot size, t is a vector of time dummy variables (quarter), and L is a dummy variable indicating whether or not the home is located within the 500 meter buffer zone of the landfill. One model is estimated before the leak date (August 31, 2008) and one after. The coefficients for the L dummy variables, l, from the before and after models are then compared to determine the change in relative price impacts of proximity to the landfill after the leak. This two model, before–after approach is executed first on the Brookland Greens (direct) case and then on the remaining potential contagion landfills. Due to data limitations and landfill locations, only 10 of the 35 notified landfills had enough surrounding residential property sales to estimate potential contagion effects from the initial leak and the subsequent announcement. We began our analysis by attempting to locate the exact locations of the landfills in the original EPA list. Of the 35 listed sites, 30 were located to an exact parcel(s) level. Of these 30, 9 have been closed and capped and have residential neighborhoods or estates located within 500 meters of the exterior boundary of the landfill and had at least 50 sales of residential properties during the study period. Only one (Clayton South) met the open landfill criteria with the requisite amount of property sales. Exhibit 2 shows the location of each landfill in the Melbourne Metropolitan Area. The majority of the landfills examined in this study had been, in 2008, redeveloped by the city councils into public amenities such as parks, gardens, sporting facilities, and the like. Consequently, public awareness of landfills in the area might be minimal, because the current uses are considered amenities. Of the sites we investigated, two were vacant and had not been redeveloped and one was partially redeveloped and partially vacant. The others were parks of varying sorts. Details regarding the redevelopment of these sites are found in Exhibit A2 in the Appendix. D a t a Our data includes all sales of single-family residential properties from 2006 to 2011. The source of the data is the Australian Property Monitors (APM), obtained through the Australian Urban Research Infrastructure Network (AURIN). The data includes a limited number of characteristics about the physical attributes of the homes: number of bedrooms, number of bathrooms, size of the lot (m ), and the presence of a study. Living area of the home as well as judgments regarding the quality or condition of the home are not commonly found in property data in Australia; this dataset is not an exception. J O S R E V o l . 9 2 0 1 7 u u 3 2 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s Exhibit 2 u Landfill Location Map Unfortunately, the exact latitude and longitude for each home are not available. Home location is available only at the mesh block level. Home sales within mesh blocks that are located within 2,500 meters of the outer boundary of each landfill make up our study area. The lack of exact geolocations does impact our choice of model functions as we are not able to estimate more advanced spatial autoregressive specifications. Exhibit 3 is an example of the size of mesh blocks, which generally contain 30 to 60 homes. As Exhibit 3 shows, the landfills are widely spread over the Melbourne Metropolitan Area. Home prices in Melbourne also vary considerably based on location, with the outer suburban areas offering markedly cheaper housing than inner city locations. These differences in prices are highlighted in Exhibit 4. Additionally, inner city locations such as Northcote and Farnsworth are composed of smaller homes on smaller lots. R e s u l t s The investigation begins with a univariate analysis of prices surrounding each landfill both before and after the announcement date. In most cases, exceptions being Altona, Brookland Greens, and Scotch, the prices in the affected (,500 m) and control (500 m to 2,500 m) areas during the before time period are relatively N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 3 3 Exhibit 3 u Mesh Blocks in the Brookland Greens Area Exhibit 4 u Summary Statistics: Mean Values Sales Price Bedrooms Bathrooms Lot Size (sq. m) Brookland Greens $319,799 3.48 1.80 628 Altona $431,107 3.17 1.56 549 Box Hill $784,710 3.18 1.57 610 Clayton South $448,387 3.21 1.47 579 Craigieburn $341,105 3.45 1.89 538 Farnsworth $547,721 2.77 1.34 370 Green Gully $392,302 3.35 1.63 649 Llewellyn $491,688 3.71 1.90 705 Narre Warren $374,988 3.44 1.92 632 Northcote $670,923 2.70 1.27 393 Scotch $492,946 3.06 1.62 523 J O S R E V o l . 9 2 0 1 7 u u 3 4 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s Exhibit 5 u Univariate Price Analysis Before After Change Affected Control Affected Control Affected Control Brooklyn Greens $377,675 $296,475 $420,307 $313,407 11% 6% Altona $440,500 $383,725 $499,166 $466,532 13% 22% Box Hill $652,611 $690,900 $1,049,898 $903,102 61% 31% Clayton South $343,390 $391,401 $440,907 $502,948 28% 28% Craigieburn $280,364 $305,082 $330,950 $362,321 18% 19% Farnsworth $519,305 $475,474 $697,734 $606,782 34% 28% Green Gully $329,878 $365,251 $399,998 $412,463 21% 13% Llewellyn $427,761 $431,432 $530,506 $546,955 24% 27% Narre Warren $387,202 $336,460 $428,026 $399,458 11% 19% Northcote $574,909 $594,570 $745,312 $748,206 30% 26% Scotch $364,037 $472,786 $430,789 $544,863 18% 15% similar. The existence of price dissimilarities, however, means that we will have to properly control for pre-existing differences in the multi-variate regression analysis. As seen in Exhibit 5, most areas (both control and affected) saw average prices move around 15% to 30% from the before and after periods. Brookland Greens offers a noticeable departure from this trend as both the affected and control area mean prices only changed 11% and 6% between these two periods. Other notable deviations from the norm include the high price increases in Box Hill and Farnsworth, two areas undergoing rapid gentrification. We also examine the ‘‘stickiness’’ of house prices in the affected areas. Potential sellers may not simply reduce price to a level at which a buyer can be found, instead—especially if they are owner occupiers—they may choose to forego sale and stay in the home until the market improves (Case and Shiller, 2003; Miles, 2008). As a result, changes in the volume of sales may also signify impacts on markets due to contagion and proximity to a perceived disamenity. Exhibit 6 shows the sales volumes in the affected (,500 m) and control (500 m , 2,500 m) area around each landfill both before and after the leak announcement. Overall sales volumes within 500 meters of the Brookland Greens site declined 18% in the period after the announcement, suggesting a softening of the market. Sales volumes also declined in 7 of the 10 other contagion sites following the leak. While the univariate analysis may provide interesting context to the situation, it does not account for potential changes in the types of houses that sold in the area or the specific times during the 32-month before and after period that homes sold. To better account for these potential differences, an estimate of a set of hedonic N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 3 5 J O S R E V o l . 9 2 0 1 7 u u Exhibit 6 u Sales Volume by Case Jan. 2006 to August 2008 Sept. 2008 to April 2011 ,500 m 500 m , 2,500 m % in 500 m ,500 m 500 m , 2,500 m % in 500 m Change Brooklyn Greens 113 74 60.43% 184 253 42.11% 218.32% Altona 110 344 24.23% 131 461 22.13% 22.10% Box Hill 311 703 30.67% 210 516 28.93% 21.74% Clayton South 23 878 2.55% 31 976 3.08% 0.53% Craigieburn 156 394 28.36% 413 805 33.91% 5.54% Farnsworth 498 864 36.56% 503 955 34.50% 22.06% Green Gully 204 287 41.55% 278 560 33.17% 28.37% Llewellyn 217 271 44.47% 215 359 37.46% 27.01% Narre Warren 289 468 38.18% 315 554 36.25% 21.93% Northcote 595 950 38.51% 627 962 39.46% 0.95% Scotch 407 306 57.08% 466 400 53.81% 23.27% 3 6 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s Exhibit 7 u Brookland Greens Model Results Before After Coeff. Std. Error Coeff. Std. Error Intercept 10.90* 0.2659* 11.0678* 0.1191* Affected Area (,500 m) 0.1236* 0.0268* 0.1064* 0.0191* Bathrooms 0.0594* 0.0254* 0.1098* 0.0113* Bedrooms 0.0412* 0.0159* 0.1731* 0.0178* Lot Size (log) 0.2186* 0.0356* 0.0435* 0.0098* Study 0.0254* 0.0212* 0.059* 0.0119* Monthly Time Dummies Not shown Not shown R 0.5396 0.5952 s 0.1291 0.1207 Note: *Significant at 0.05 level. price models is used to measure the relative changes in price in the affected area after the announcement. Hedonic Price Models In the hedonic price modeling exercise, we analyze the impact of the landfill contamination on the Brookland Greens housing estate. The results of both the before and the after models are shown in Exhibit 7. Overall, the models offer a relatively low R-squared; however, the standard errors are between 12.9% and 12.1%. These low standard errors suggest that the models do fit the data well. The coefficients all have the expected sign, with the exception of a few insignificant coefficients on some of the structural characteristics. Note that the presence of a study in the house acts primarily as a proxy variable indicating a higher quality home or one that has been remodeled (as a direct quality variable is not available in our data). Brookland Greens Looking specifically at Brookland Greens, it is apparent that in the before time period, the area within 500 meters of the landfill was worth, on average, 13.2% more than the control area. After the announcement of the methane leak, this difference dropped to 11.2%, a decrease in relative prices of 1.9%. This decrease represents the average difference in relative value between the two periods, before and after the announcement date. Capped and Redeveloped Landfills. Next, the same methodology is employed on the other ten notified landfills. The results of this analysis are shown in Exhibit 8. For five out of the nine landfills, there was no statistically significant difference N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 3 7 Exhibit 8 u Results Change in Relative Value Landfill Before After after Announcement Capped Sites Altona None None None Box Hill None None None Craigieburn 28.11% 25.62% 2.49% Farnsworth None 3.50% 3.50% Green Gully None None None Llewellyn None None None Narre Warren 5.33% None 25.33% Northcote None None None Scotch 215.67% 212.96% 22.71% Open Sites Clayton South None 28.26% 28.26% in prices (once home characteristics were controlled) between affected and control areas in either the before or after time periods. In the remaining five cases, three sites—Craigieburn, Farnsworth, and Scotch—showed moderate price increases while prices near Narre Warren decreased 5.3%, relatively speaking, in the time period after the announcement date. Open Fill: Clayton South. When the analysis is replicated for the Clayton South site, the results show that the homes within the 500-meter buffer are selling at an 8% discount, post notification. When the affected distance is increased to ,1,000 meters, a decrease in value of 0.79% is observed. C o n c l u s i o n Our original research question is based on an expectation that when information regarding an unexpected, detrimental event, in this case the discovery of methane gas at a housing development adjacent to a landfill, is released, a diminution in value will occur. We further posit that effects may also be present at similarly situated sites that are not directly affected by the event (i.e., non-spatial contagion). The analysis of the direct event involving the methane leak in the Brookland Greens Estate shows a 1.9% relative decline in prices in the period after information regarding the leak was made public. This diminution in value is expected as the methane leak represents an identified disamenity for homeowners in the area. The test for non-spatial contagion in the nine capped landfills and one open site located in heavily residential areas throughout the metropolitan region identified no systematic reduction in residential house prices during the period immediately after the event. Two of the sites, Narre Warren and Clayton South, show declines J O S R E V o l . 9 2 0 1 7 u u 3 8 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s in prices, while three others show increases. Five of the sites exhibited no change in relative values between homes near the closed landfills and those farther away. In other words, there appears to have been no measurable, sustained contagion effect in the closed and capped market around landfills. There are a number of reasons for this null finding. First, many of the covered landfills are currently operating as environmental amenities—parks, sporting facilities, and golf courses—and as such their prior use as landfills may not be evident to potential homebuyers. Even if potential homebuyers have been made aware of recent issues with capped landfills generally, they may not realize a fill is near the property they are buying. Sellers, agents, and local officials are under no obligation to disclose the issue to buyers, therefore information diffusion is likely low in these cases (provided the property is not directly affected by negotiations between the city council and developer as would be noted in a Section 173 Agreement). Limitations to the data may also reduce the reliability of the results as the inability to accurately locate home sales to exact geographic locations could be obscuring actual diminutive effects. In other words, homes located within a few hundred meters maybe be negatively affected. However, this impact is overwhelmed by the lack of impact from homes farther away, all of which are aggregated to the mesh block group level. One of the basic assumptions flowing through this study is that local residents’ knowledge about contaminated land is similar to that of professionals—they react to news of pollutants systematically. This implies that communities react to landfills in the same way (i.e., the reactions are homogeneous and directly comparable). This study is based in a number of diverse communities with a variety of (dis)amenities present. These local factors may be overriding any evidence of non-spatial contagion from the Brookland Greens site, if the contagion even exists. Additionally, because landfill sites are normally developed post- closure and not prior, there may be less of a social impetus and fewer communication networks to motivate collective fear and price drops. Even though the sites were all put on notice by the EPA and a fair bit of information was made available via formal networks (e.g., local newspapers, television newscasts), it is unclear how much of this information effectively reached the affected population. It is possible that the areas that sustained a negative impact (Clayton South and Narre Warren) had stronger and more viable social networks to transfer information, thus contagion effects may be evident there. It is also possible that the buyers interested in these two suburbs were more risk adverse with more alternatives available to them. The evaluation of the closed and capped sites provides interesting insight into ‘‘who knew what when.’’ Local councils in Melbourne are arguably skillful and motivated to convert and cover landfill site surfaces. Many of the closed and capped sites have been redeveloped into parkland, golf courses or other amenities that the casual observer would never recognize as a former landfill. It is rational for local councils, developers, and the EPA to control circulation of or passively slow down information to counteract social networks and whether intended or not, N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 3 9 capping and covering a landfill does just that. The removal of visual evidence has the potential to impair social memory, negatively impacting any associated communication networks, or to simply create new and different ones. While we recognize that two cases do not provide conclusive evidence of a systematic pattern of contagion, the results for Clayton South and Narre Warren are interesting. As an open landfill, Clayton South possesses the characteristics that suggest it is most like Brookland Greens. When analyzed, there is evidence, albeit limited, that the contagion sparked by the Brookland Greens case may be present. While sales volume increased modestly (0.53%), property values were depressed by over 8%. It is entirely plausible that the increase in volume was motivated by the relative decrease in price. Narre Warren was the only capped site that consisted of vacant land that had not been redeveloped and the surrounding properties show a diminution in value, along with a reduction in sales volume. Neither Clayton South nor Narre Warren possessed the amenity effect of the capped and redeveloped fills that were parks or sporting facilities at the time of notification. This suggests that the effects of a single event, regardless of how detrimental they are perceived to be at one site, are not distributed equally across markets. A p p e n d i x Exhibit A1 u EPA Designated Landfill Types Landfill Type Landfill Description Buffer in Operation Buffer Post-closure Type 1 Prescribed industrial Not disclosed as part of different guidelines wastes. Type 2 Putrescible (municipal) 500 meters from 500 meters from waste, solid inert waste buildings and structures building or structures and fill material. Can include category C industrial waste if licensed. Type 3 Solid inert waste, fill 200 meters from 200 meters from material. buildings and structures buildings and structures Note: In the VCAT hearing in regard to the Stevenson’s Road Landfill site and the Brookland Greens Estate, VCAT ruled that the 200-meter buffer distance was to be measured from the ‘‘active tipping area,’’ being the ‘‘active tipping face’’ of the landfill. J O S R E V o l . 9 2 0 1 7 u u 4 0 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s Exhibit A2 u Landfills Date Ceased Landfill Site Status Operations 2008 Use Comments Queen Street Closed 1998 Park 75 hectares of parks and wetlands. Altona 14 Federation Closed 1999 Park Parkland, sporting reserves and vacant St., Box Hill land, 2006 permit granted for development of residential to part of the site. Craigieburn Closed Unknown Park In 2007 the site was already developed, as Gardens, a park, partial leisure center and community Craigieburn facilities. BBQ picnic area, shelter, not well maintained, mostly unstructured open space. There are no known records of the type, depth, compaction or other treatment of fill in the landfill (Hume City Council, 2007). Quarry Park, Closed Park Parkland, nearby city college, and other Farnsworth sporting amenities. Ave., Footscray Clarke Rd., Closed 2002 East of High levels of methane detected in bored Springvale Clarke Rd. holes adjacent to community building and South is Park, residences on southern boundary. This was west of a capped (or poorly capped landfill at this Clarke Rd., time) EPA issued a new notice in Dec 2008. vacant land It was noted that although permissions for use as a rubbish tip expired in 1998, this site was used up to 2009 to stockpile scrap concrete (EPA, 2009). 151 Green Closed Unknown Park The Green Gully Reserve is a 69 hectare Gully Road, approximately site and has been established for some time, Keillor Downs 30 years ago noted as the Council’s second largest (Brimbank City recreational sites with six sports grounds. Council, 2013) Landfill gas issues are not considered to be a problem given the age of the landfill (Brimbank City Council, 2013). Llewellyn Park, Closed 1980s Park and The parklands and reserve were developed Wantirna sporting after the closure; however, exact dates South reserve cannot be located. Residents were notified due to high levels of gas post 2008 Stevenson’s Road report. Quarry Road, Closed 1996 Vacant land This site was previously a Type 2 landfill Narre Warren operating since 1982, it was used as a North landfill gas power station from 1992 until 1996 when it was closed. Clifton Road, Closed Early 2000s Park Former quarry used as a local landfill, Northcote classified as a Type 2 landfill. Redeveloped into major 13 hectare regional park in 2002 known as ‘‘All Nations Park.’’ N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 4 1 Exhibit A2 u (continued) Landfills Date Ceased Landfill Site Status Operations 2008 Use Comments Scotch Pde, Closed 1970s Park The park was developed in the 1980’s and Chelsea was completed in 1988. This site has several redevelopments with other facilities included over time with sporting fields, community center and clubs. Stevenson’s Closed 2005 Vacant land Residents are unaware of landfill in contract Road, of sale and S173. Cranbourne Clayton South Open N/A Landfill Type Residents are aware of landfill and make 2 regular complaints about the odor, with regular updates and community newsletters sent via the EPA and City of Kingston. Furthermore the local ‘‘Leader’’ newspaper has regular articles on the landfill sites in the area. E n d n o t e s The EPA provides a number of guideline documents in regard to landfill sites in Victoria. There are two main documents: Best Practice Environmental Management—Siting, Design, Operation and Rehabilitation of Landfills and Industrial Waste Resource Guidelines. These guidelines classify the type of landfill, and what type of landfill can be disposed of, consequently the guidelines regarding siting, design, operation, and rehabilitation change depending on the classification. The purpose of the buffers is to manage the odors from the landfill (a key concern during the operational phase of a landfill) and landfill gas, which may be an explosion risk and / or asphyxiation risk up to 30 years post closure of the landfill) (EPA, 2015). In the 1991 State Environment Protection Policy (Siting and Management of Landfills Receiving Municipal Wastes) document produced by the EPA, the required distance was 200 meters but due to the number of complaints regarding odors from various fills, the EPA was motivated to change that distance to 500 meters. Residential land subdivisions in Australia are commonly set up to be completed in stages, particularly when they are large subdivisions or estates. This allows efficient site development and sale of lots in a process that provides a feeding of lots into the market without flooding it and creating an oversupply. Most importantly, the staged process assists the funding of the development, as when a stage is complete the land lots can be settled, providing the developer with income and cash flow to continue and fund the following stages. In Victoria, the sale of land provides a contract in which all salient details of a property are to be disclosed. The Section 173 Agreement includes any agreements between a developer and the city council, which are then incorporated into any further contracts for J O S R E V o l . 9 2 0 1 7 u u 4 2 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s the sale of land. The S173 Agreement for the Brookland Greens development failed to disclose the site’s vicinity to the landfill, referring to it instead as ‘‘sand extraction operations.’’ After properly adjusting for dummy variables derived from a semi-log model; exp(c) 2 1 (Halvorsen and Palmquist, 1980). www.apmpropertydata.com.au. www.aurin.org.au. Mesh blocks are small geographic aggregations containing about 50 homes used by the Australian Bureau of Statistics for census purposes. Individual regression results are available upon request. R e f e r e n c e s Benson, E.D., J.L. Hansen, A.L. Schwartz, and G.T. Smersh. Pricing Residential Amenities: The Value of a View. The Journal of Real Estate Finance and Economics, 1998, 16:1, 55– Bolitzer, B. and N. Netusil. The Impact of Open Spaces on Property Values in Portland, Oregon. Journal of Environmental Management, 2000, 59, 185–93. Bond, S. The Use of Conjoint Analysis to Assess the Impact of Environmental Stigma. Pacific Rim Property Research Journal, 2001, 7:3, 182–94. Brasington, D.M. and D. Hite. Demand for Environmental Quality: A Spatial Hedonic Analysis. Regional Science and Urban Economics, 2003, 35:1, 57–82. Brimbank City Council. City of Brimbank Alternative Water Project, Final Report. Prepared for Department of Sustainability, Environment, Water, Population and Communities. August 1, 2013. Brouwer, G. Brookland Greens Estate—Investigation into Methane Gas Leaks. Ombudsman Victoria, October 2009. Caccioli, F.M. Shrestha, C. Moore, and J.D. Farmer. Stability of Financial Contagion Due to Overlapping Portfolios. Journal of Banking and Finance, 2014, 46, 233–45. Cameron, T.A. Directional Heterogeneity in Distance Profiles in Hedonic Property Value Models. Journal of Environmental Economics and Management, 2006, 51:1, 26–45. Caplan, A., T. Grijalva, and D. Jackson-Smith. Using Choice Question Formats to Determine Compensable Values: The Case of a Landfill-siting Process. Ecological Economics, 2007, 60:7, 834–46. Case, K.E. and R.J. Shiller. Is There a Bubble in the Housing Market? An Analysis. Brookings Papers on Economic Activity, September 2003. Chan, N. Stigma and its Assessment Methods. Pacific Rim Property Research Journal, 2001, 7:2, 126–40. ——. Stigma Assessment: A Multi-Criteria Decision-Making Approach. Pacific Rim Property Research Journal, 2002, 8:1, 29–47. Chen, M.C., H.J. Tsai, T.F. Sing, and C.Y. Yang. Contagion and Downside Risk in the REIT Market during the Subprime Mortgage Crisis. International Journal of Strategic Property Management, 2015, 19:1, 42–57. Cho, S., J.M Bowker, and W.M. Park. Measuring the Contribution of Water and Green Space Amenities to Housing Values: An Application and Comparison of Spatially Weighted Hedonic Models. Journal of Agricultural and Resource Economics, 2006, 485–507. N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 4 3 Cooper, M. Gas Leak Triggers Landfill Inquiry. Sydney Morning Herald, September 18, Do, Q. and G. Grudnitski. Golf Courses and Residential House Prices: An Empirical Examination. Journal of Real Estate Finance and Economics, 1995, 10:3, 261–70. Environmental Protection Agency. Siting, Design, Operation and Rehabilitation of Landfills. Best Practice Environmental Management Series. October, 2001. ——. Assessment of the Potential for Methane Gas Movement from Victorian Landfills. Environment report, Publication 1270. February, 2009. ——. Siting, Design, Operation and Rehabilitation of Landfills. Best Practice Environmental Management Series. Accessed 4 / 29 / 2016 from: http: / / www.epa.vic.gov.au/ our-work / publications / publication / 2015 / august / 788-3. August 12, 2015. Eshet, T., M. Baron, M. Shechter and O. Ayalon. Measuring Externalities of Waste Transfer Stations in Israel using Hedonic Pricing. Waste Management, 2007, 27:4, 614–25. Freybote, J. and E. Fruits. Perceived Environmental Risk, Media, and Residential Sales Prices. Journal of Real Estate Research, 2015, 37:2, 217–44. Ghosh, C., R. Guttery, and C.F. Sirmans. Contagion and REIT Stock Prices. Journal of Real Estate Research, 1998, 16:3, 389–400. Guntermann, K. Sanitary Landfills, Stigma and Industrial Land Values. Journal of Real Estate Research, 1995, 10:5, 531–42. Halvorsen, R. and R. Palmquist. The Interpretation of Dummy Variables in Semilogarithmic Equations. American Economic Review, 1980, 70, 474–75. Hanna, B.G. House Values, Incomes, and Industrial Pollution. Journal of Environmental Economics and Management, 2007, 54:1, 100–12. Hindsley, P., S.E. Hamilton, and O.A. Morgan. Gulf Views: Toward a Better Understanding of Viewshed Scope in Hedonic Property Models. The Journal of Real Estate Finance and Economics, 2013, 47:3, 489–505. Hite, D., W. Chern, F. Hitzhusen, and A. Randall. Property-value Impacts of an Environmental Disamenity: The Case of Landfills. The Journal of Real Estate Finance and Economics, 2001, 22, 2–3, 185–202. Hume City Council. Hume City Council Craigieburn Leisure Centre Master plan, December 2007. Accessed 4 / 29 / 2016 from: www.hume.vic.gov.au / files / bd48f2c0-267f... / Attachment2 CBMP.pdf. Ihlanfeldt, K.R. and L.O. Taylor. Externality Effects of Small-scale Hazardous Waste Sites: Evidence from Urban Commercial Property Markets. Journal of Environmental Economics and Management, 2004, 47:1, 117–39. Kiel, K.A. Environmental Contamination and House Values. College of The Holy Cross, Department of Economics, Faculty Research Series, Paper #06-01, 2006. Kiel, K.A. and K.T. McClain. The Effect of An Incinerator Siting on Housing Appreciation Rates. Journal of Urban Economics, 1995, 37:2, 311–23. Kiel, K.A. and J. Zabel. Estimating the Economic Benefits of Cleaning Up Superfund Sites: The Case of Woburn, Massachusetts. Journal of Real Estate Finance and Economics, 2001, 22:2, 163–84. Kohlhase, J.E. The Impact of Toxic Waste Sites on Housing Values. Journal of Urban Economics, 1991, 30:1, 1–26. McCluskey, J.J. and G.C. Rausser. Stigmatized Asset Value: Is It Temporary or Long-term. Review of Economics and Statistics, 2003a, 85:2, 276–85. J O S R E V o l . 9 2 0 1 7 u u 4 4 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s ——. Hazardous Waste Sites and Housing Appreciation Rates. Journal of Environmental Economics and Management, 2003b, 45:1, 166–76. Messer, K.D., W.D. Schulze, K.F. Hackett, T.A. Cameron, and G.H. McClelland. Can Stigma Explain Large Property Value Losses? The Psychology and Economics of Superfund. Environmental & Resource Economics, 2006, 33:2, 299–324. Michaels, R.G. and V.K. Smith. Market Segmentation and Valuing Amenities with Hedonic Models: The Case of Hazardous Waste Sites. Journal of Urban Economics, 1990, 28:2, 223–42. Miles, W. Boom-Bust Cycles and Forecasting Performance of Linear and Nonlinear Models of House Prices. Journal of Real Estate Finance and Economics, 2008, 36:3, 249–64. Mink, M. and J. de Haan. Contagion during the Greek Sovereign Debt Crisis. Journal of International Money and Finance, 2013, 34, 102–13. Nelson, J.P. Hedonic Property Value Studies of Transportation Noise: Aircraft and Road Traffic. In: Hedonic Methods in Housing Markets. Springer, 2008. Nelson, A.C., J. Genereux, and M. Genereux. Price Effects of Landfills on House Values. Land Economics, 1992, 68:4, 359–65. Palmquist, R.B. and V.K. Smith. The Use of Hedonic Property Value Techniques for Policy and Litigation. In: International Yearbook of Environmental and Resource Economics 2002/ 3. Chapter 3, Edward Elgar, 2002. Reichert, A.K. Impact of a Toxic Waste Superfund Site on Property Values. Appraisal Journal, October, 1997, 381–92. Reichert, A.K., M. Small, and S. Mohanty. The Impact of Landfills on Residential Property Values. Journal of Real Estate Research, 1992, 7:3, 297–314. Rosen, S. Hedonic Prices and Implicit Markets: Product Differentiation in Pure Competition. Journal of Political Economy, 1974, 82:1, 34–55. Ryan, B. and B. Gross. The Diffusion of Hybrid Seed Corn in Two Iowa Communities. Rural Sociology, 1943, 8:1, 15. Schulz, M. and D. Barber. Residents Win Tens of Millions in Compensation after Cranbourne Methane Gas Leak Fiasco. Herald Sun, March 25, 2011. Seiler, M., A. Collins, and N. Fefferman. Strategic Mortgage Default in the Context of a Social Network: An Epidemiological Approach. Journal of Real Estate Research, 2013, 35:4, 445–75. Semintelli, A. and R. Simons. Regulation of Leaking Underground Storage Tanks: Policy Enforcement and Unintended Consequences. Economic Development Quarterly, 1997, 11:3, 236–48. Shultz, S.D. and N.J. Schmitz. Augmenting Housing Sales Data to Improve Hedonic Estimates of Golf Course Frontage. Journal of Real Estate Research, 2009, 31:1, 63–79. Simons, R., W. Bowen, and A. Sementelli. The Price and Liquidity Effects of UST Leaks from Gas Stations on Adjacent Contaminated Property. Appraisal Journal, 1999, 67, 186– Simons, R., W. Levin, and A. Sementelli. The Effect of Underground Storage Tanks on Residential Property Values in Cuyahoga County, Ohio. Journal of Real Estate Research, 1997, 14:1, 29–42. Simons, R. and J. 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. N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 4 5 Simons, R. and K. Winson-Geideman. Determining Market Perceptions of Contamination of Residential Property Buyers using Contingent Valuation Surveys. Journal of Real Estate Research, 2005, 27:2, 193–220. th Wu, H. and C. Chen. Urban ‘‘Brownfields’’: An Australian Perspective. 18 Annual Pacific Rim Real Estate Society Conference, Adelaide, January, 15–18, 2012. All articles published in JOSRE are distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Kimberly Winson-Geideman, University of Melbourne, Parkville Victoria 3010, Australia or kimberly.winson@unimelb.edu.au. Andy Krause, Greenfield Advisors, Seattle, Washington or AKrause@ GreenfieldAdvisors. Hao Wu, University of Melbourne, Parkville Victoria 3010, Australia or haow@ unimelb.edu.au. Georgia Warren-Myers, University of Melbourne, Parkville Victoria 3010, Australia or g.warrenmyers@unimelb.edu.au. J O S R E V o l . 9 2 0 1 7 u u http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Sustainable Real Estate Taylor & Francis

Non-spatial Contagion in Real Estate Markets: The Case of Brookland Greens

Download PDF
24 pages

Loading next page...
 
/lp/taylor-francis/non-spatial-contagion-in-real-estate-markets-the-case-of-brookland-2EphgRSk7A

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Taylor & Francis
Copyright
© 2017 American Real Estate Society
ISSN
1949-8284
DOI
10.1080/10835547.2017.12091898
Publisher site
See Article on Publisher Site

Abstract

Non-spatial Contagion in Real Estate Markets: The Case of Brookland Greens A u t h o r s Kimberly Winson-Geideman, Andy Krause, Hao Wu, and Georgia Warren-Myers A b s t r a c t We investigate contagion in real estate markets by evaluating the effects of a widely publicized landfill contamination event in one local market on the price of homes near landfills in non-impacted markets within the same metropolitan region. The impact of proximity to open, closed, and redeveloped landfills in the directly affected and contagion neighborhoods is tested at distances varying from ,500 meters to 2,500 meters using the traditional hedonic pricing model. The results are mixed and relative to the current use of the landfill as closed, capped, and redeveloped landfills show no impact. Sites that are capped yet undeveloped and sites with open fills appear to show some impact, although further research is needed to support any contagion effects. There is a substantial amount of research quantifying the relation between proximity to environmental (dis)amenities and value. Amenities such as golf courses (Do and Grudnitski, 1995; Shultz and Schmitz, 2009), scenic views (Benson, Hansen, Schwartz, and Smersh, 1998; Hindsley, Hamilton, and Morgan, 2013), and green space (Bolitzer and Netusil, 2000; Cho, Bowker, and Park, 2006) positively affect value while negative effects are shown in proximity to landfills (Reichert, Small, and Mohanty, 1992; Hite, Chern, Hitzhusen, and Randall, 2001), high traffic noise (Nelson, 2008), and environmental contamination (Simons and Saginor, 2006) such as oil spills. These relations are based on a two-tier spatial component, as well as the principle of substitution (i.e., the value of one property is directly related to the value of adjacent or similar properties, all else held equal). The first spatial component is a direct effect wherein a property is immediately adjacent to or directly affected by the (dis)amenity. The other spatial component links the value of unaffected properties to the change in value of nearby affected properties. Therefore, when a property abuts another that is directly affected by an environmental (dis)amenity, its value is impacted, although likely not to the same degree as the directly impacted property. In this paper, we provide further evidence of these phenomena by investigating the case of a housing estate, Brookland Greens Estate, and an adjacent landfill located near Melbourne, Australia. We test the spatial relations between properties directly and indirectly affected by the issues associated with the landfill. What we seek to address is not the spatial relation between properties per se, but rather how and to what extent the diminution in property values spreads to non-spatially N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 2 3 related, similar markets. The framework for this is based in contagion theory where an event in one market spills over to another that has little or no direct spatial relation to the first. We hypothesize that a detrimental event associated with a particular type of land use (landfill in this case) and evidenced in a diminution in the value of nearby properties may have a ‘‘contagion effect’’ in which real estate proximate to other land uses of this type, although not proximate to the initial event, will experience similar phenomena, even though they are not directly affected by the detrimental event (i.e., non-spatial market contagion). The key mechanism tested is the ability of a distant (spatially) event to raise awareness of the potential negative externalities of an existing disamenity and how this contagion may affect observed market outcomes. While the results are inconclusive, they do suggest a mild contagion effect in some of the submarkets, particularly where the landfill has not been redeveloped. We next discuss the literature on contagion theory, as well as the spatial and non- spatial effect of social (dis) amenities on observed market behavior. A description and analysis of the landfill case and its associated regulatory environment follows. After explaining the methodology and data, the main findings are presented on the direct price impact and possible contagion effects on capped and redeveloped landfills. We conclude with further interpretation and a discussion on future research. L i t e r a t u r e This research is framed in contagion theory—unexposed markets are subject to impacts similar to an affected market by virtue of that similarity—leading us to make a few assumptions about the cause of those effects. First, we assume information transparency and transmission in residential markets (e.g., time lag and market reaction) is measured by price change. We also assume there are at least two types of ‘‘spatial reactions’’: (1) impact at the landfill area, the continuous spatial contagion effect; and (2) landfill areas in other regions, the non- spatial (discrete) contagion effect. A third reaction, not addressed in this research, is (3) a contagion effect on contaminated sites other than landfills, reflecting human ability to relate events (e.g., by imagination). If information transmission efficiency is the dominant force, there will be the negative price relation of (1) . (2) . (3), or else other market or non-market forces or behavioral theories may exist to explain the contagion effect. We also assume that there is likely a diminution in value associated with the stigma of living near a landfill that is already capitalized into the market value of the properties. Stigma refers to the real or perceived, readily identifiable issues associated with landfills or other contamination sources that may affect property value. Consequently, it is important to understand the level at which a community is aware of the disamenity and its implication for occupants and future home purchasers. We posit that when a contagion event occurs, stigma effects, whether J O S R E V o l . 9 2 0 1 7 u u 2 4 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s present or not, will be magnified or will emerge relative to proximity alone, thus we do not control for directional heterogeneity as suggested in Cameron (2006). The literature review that follows is rooted in these assumptions. We begin with an overview of contagion theory, followed by studies that investigate the links between stigma, location, and house prices from spatial, non-spatial, and social justice perspectives. Contagion Theory According to contagion theory, groups of people, by virtue of communication with the social networks they engage in, are likely to adopt the attitudes and behavioral norms of the group. Just as a virus spreads through personal contact, so do attitudes, behaviors, and actions. Efforts to influence do not have to be present; the mere existence of communication portals is enough. These links are well established in the literature beginning with the Ryan and Gross (1943) study of hybrid seed corn and continuing through today with a vast number of studies conducted in the financial markets (e.g., Ghosh, Guttery, and Sirmans, 1998; Mink and de Haan, 2013; Caccioli, Shrestha, Moore, and Farmer, 2014; Chen, Tsai, Sing, and Yang, 2015). Contagion may be transmitted through social or public (e.g., media) channels. Contagion studies in the real estate and other financial literature typically follow a predictable pattern and often focus on negative economic or other shocks and the resulting behaviors, actions or attitudes of individuals, organizations or communities. When an event occurs, information flows through communication networks, elevating levels of perceived risk. As perceived risks increase and information flows from person to person, the ‘‘group belief’’ reinforces or intensifies. The group may then engage in illogical, emotionally-driven activities they would not necessarily engage in as individuals, primarily due to the anonymity the group provides. The fluidity of communications within existing social networks helps create and refine new behaviors, attitudes, and actions that become socially acceptable, if only for limited time. The rapid pace at which movements, ideas, and / or activities can spread has the potential to impact markets on a global scale, particularly given recent technological improvements in communication. Research by Seiler, Collins, and Fefferman (2013) shows that the number of strategic mortgage defaults during the Global Financial Crisis (GFC) were likely exacerbated by the suggestion that underwater homeowners exercise their put option, further depressing home values and predicating the market collapse. Prior to the GFC, most defaults were economic, the result of job loss or divorce, for example. Strategic defaults, conversely, were based on the mortgagor’s willingness to ‘‘turn over the keys’’ to the lending institution because the value of the mortgage was greater than that of the collateral. Strategic defaults were comparatively rare before that time. Freybote and Fruit’s (2015) study of media coverage in the case of the construction of a natural gas pipeline provides additional evidence of how information networks N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 2 5 contribute to market contagion. The authors investigate how residential properties are impacted by proximity to the pipeline and how unrelated pipeline explosions in different markets affect value during the subject pipeline construction process. Consistent with prior literature, they find that proximity to the pipeline negatively affected value. Further, they also determined that the perceived riskiness of proximity to natural gas pipelines increases in months when a fatality associated with an explosion is widely reported in the media, and prices near the pipeline under construction moderate. Stigma, Location, and House Prices The literature on the effects of environmental disamenities on value is lengthy and well established with studies focused on landfills (Nelson, Genereux, and Genereux, 1992; Gunterman, 1995; Hite, Chern, Hitzhusen, and Randall, 2001), Superfund sites (Michaels and Smith, 1990; Kohlhase, 1991; Kiel, 1995; Kiel and McClain, 1995; Kiel and Zabel, 2001), and leaking underground storage tanks (Sementelli and Simons, 1997; Simons, Bowen, and Sementelli, 1999; Simons, Levin, and Sementelli, 2009), for example. In general, these studies show a significant negative effect associated with proximity that decreases as distance from the source of the contamination increases. Studies showing little to no effect are typically site specific or address less intense uses, as shown in Ready’s (2010) research on low-volume landfills. Others show the effects are unequally distributed based on direction (Cameron, 2006) or property type (Ihlanfeldt and Taylor, 2004). Empirical inquiry into the relation between environmental hazard, stigma (e.g., landfill), and land prices fall under (1) a spatial perspective or (2) a non-spatial perspective. The former links to mainstream location (urban economic) theory, while the latter links more closely to information economics and behavioral ideas. A third stream of analysis concerns normative strategy and decisions such as location choice, social justice and compensation determination, data, methodology, and the cross-context applicability of results. Spatial Perspective Brasington and Hite (2005) estimate spatial effects using the demand price elasticity of homes in relation to the source of pollution. They find that ‘‘point- source pollutants’’ and nearby house prices are negatively correlated, while income, education, and concern for children are sensitive to the location of polluted site. Eshet, Baron, Shechter, and Ayalon (2007) estimate the economic value of disamenities by measuring the spatial impact of waste transfer stations on local property prices. In addition to the expected negative relation, they also raise issues on the measurement of externalities and the hedonic pricing model. Hanna (2007) focused on the dynamic relation of a polluting industrial site, house price, and household mobility as determined by income, finding that income and pollution are strongly correlated. Cameron (2006) argues the existence and significance of directional heterogeneity and its role in determining the actual spatial price effect of environmental hazard. J O S R E V o l . 9 2 0 1 7 u u 2 6 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s Non-spatial Perspective McCluskey and Rausser (2003a) suggest the possibility of short-term and long- term stigma due to discovery and clean-up associated with the hazardous site, emphasizing the relation between household income, discounted property values, and social segregation. Messer et al. (2006) discuss the significance of stigmatization in the case of clean-up action on environmentally damaged sites, specifically focusing on the impact of people’s memory in forming persistence perceptions over a long period of time. Messer’s study exposes an interesting hypothesis: If cleanup is treated as an investment to counter-act existing social cost (e.g., land pollution) and the build-up of negative perception (e.g., stigma), the actual level of damage and price behavior is likely dependent on the supply elasticity of suitable replacement sites. Reichert (1997) finds that polluted landfills tend to have liquidity, distance, and permanent devaluing effects on associated property markets. McCluskey and Rausser (2003b) suggest that clean-up actions on hazardous sites may not be as effective as new information about an environmental issue in triggering change in behavior and expectation in consumer valuation. Their findings support earlier research by Kiel and McClain (1995). Social Justice and Distribution Perspective Palmquist and Smith (2001) evaluate the use and effectiveness of the hedonic technique to evaluate environmental policy and litigation. They suggest market price based models are less directly relevant to public policy but have direct relevance to litigation that is empirical and fact-based. Simons and Winson- Geideman (2005) applied the contingent valuation method to understand buyer perceptions of hazardous sites on residential property. Caplan, Grijalva, and Jackson-Smith (2007) investigate the extent that a non-host community compensates a host community for the actual and perceived impacts of a landfill. They test and argue the value of using the resource-resource format for landfill site evaluation and decision-making, because it allows the parties to be aware of and compare alternative resource trade-offs and suitable forms of compensation for a host community. In the Australian context, Bond (2001) and Chan (2001, 2002) investigate the role of stigma and associated risk perceptions in the valuation of contaminated land in Australia’s capital cities. They identify risk factors and decision models to approach landfill site evaluation. Wu and Chen (2012) use a structural-agent approach to investigate the legal framework and process that Melbourne has implemented for its inner city brownfield sites, leaving specific investigation of the relation between environmental hazard and house price (e.g., from spatial or non-spatial perspective) unattended. In this study, we fill this gap by paying special attention to the contagion effect within a specific legal policy framework in Melbourne, Australia, a major city that is frequently regarded as having high- quality living standards and property environments. N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 2 7 T h e B r o o k l a n d G r e e n s C a s e a n d C o r r e s p o n d i n g R e g u l a t o r y E n v i r o n m e n t The Brookland Greens case began in 1992 when the Shire of Cranbourne (eventually becoming the City of Casey in 1994) approved the development of a landfill located on Stevenson’s Road, about 30 kilometers from the Melbourne CBD. The area was generally rural and undeveloped and the landfill approved as a Type II by the Environmental Protection Agency (EPA) to accept putrescible waste, which is biodegradable and produces landfill (methane) gas (see Exhibit A1 in the Appendix for details regarding landfill types). Under the terms of the Environmental Protection Act of 1970, state governments are vested with the right to govern the construction and operation of landfills, and it is the EPA’s responsibility to monitor and regulate them. The EPA grants permits for new sites, examines them for compliance, and then issues a license to operate when the work is complete. The license includes conditions for operation, limits on discharge, maintenance, reporting, and monitoring with the objective of minimizing any negative impact on the environment (Brouwer, 2009). Councils (local government) generally have the responsibility of issuing management contracts, thus they also hold a level of responsibility for the operation of the landfills. Records show that, in order to contain the methane gas at the Stevenson’s Road site, the EPA intended for the landfill to be lined with compacted clay, but through a series of missed deadlines and poor negotiations, the liner was not installed. Construction was completed and a license to operate was granted regardless. Operations began on the site in June 1996 (Brouwer, 2009). In 1999, Peet & Co., a residential development company, began construction of a housing development known as Brookland Greens Estate on land adjacent to the landfill. In early 2000, the developer petitioned the city to have parts of the site rezoned to Residential Zone 1 to facilitate the construction of additional housing units. At the same time, the developer was engaged in discussions with the EPA regarding the appropriate buffer distance required to minimize the impacts of the fill to nearby home owners. After a number of panel hearings debating 200-meter and 500-meter distances, the EPA panel sided with the developer and supported the 200-meter buffer for Brookland Greens. Incidentally, in 2001, the EPA published the Best Practices Environmental Management, Siting, Design, Operation and Rehabilitation of Landfills, recommending a 500-meter distance for all other fills. The City of Casey complied with the 200-meter request, noting that no homes were to be built within that distance. Not to be deterred, the developer approached the Victorian Civil and Administrative Tribunal (VCAT) for permission to construct stage 10 on land that was partially located within the 200-meter landfill buffer. The proposal was approved by VCAT in May 2004, who ruled that the 200-meter buffer distance was to be measured from the ‘‘active tipping area,’’ J O S R E V o l . 9 2 0 1 7 u u 2 8 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s being the ‘‘active tipping face’’ of the landfill. This ruling effectively allowed the developer to build up to the perimeter of the fill (Brouwer, 2009). However, when the required disclosures were made for the subsequent sale of Brookland Greens homes, the site was referred to as ‘‘sand extraction operations’’ and not as a landfill. In June 2005, the landfill ceased operations and less than a year later the EPA received an anonymous complaint about ‘‘bubbling storm water puddles’’ on the Brookland Greens Estate development site. Initial tests found that methane gas and carbon dioxide had entered the aquifer and was migrating from the landfill into the estate. A post-closure pollution abatement notice was issued, taking effect on February 2, 2007, and the waste discharge license was revoked later that month (Brouwer, 2009). Testing began at houses within the estate, and on August 31, 2008, methane gas at 63% volume for volume of air was found in one of the homes. On September 9 the residents were declared to be in imminent danger and the County Fire Authority was made responsible for leading the emergency response. Twenty-nine families were instructed to evacuate their homes immediately for fear they would explode. As word spread and the media began reporting the story, the EPA refused to release any information on which of the 35 other landfills in the metropolitan region were lined and officially put all of them on notice to ‘‘clean up their act.’’ (The Age, September 21, 2008). While publicly stating that they were not aware of similar incidents affecting residential neighborhoods, the EPA began conducting reviews at all other Victorian landfills to see if they posed a significant risk to any neighboring properties (Cooper, 2008). Approximately 300 homes in the Brookland Greens Estate were tested for the gas, and the danger level reduced on October 31, 2008 when the risk to residents was deemed to be at tolerable levels. Unsurprisingly, a class action lawsuit was filed against the City of Casey, as well as the EPA. A settlement was reached in March 2011 for AU$23.5 million; $13.5 million paid by the city and $10 million by the EPA and $6 million of the total covered legal fees. The 771 residents received payments ranging from $7,000 to $132,000 depending on the severity of the impact to their property, an average of $22,373 each (Schulz and Barber, 2011). A schedule of the incidents is shown in Exhibit 1. M e t h o d s We test for two potential causes of price impacts. Firstly, the analysis of direct impacts to property values surrounding Brookland Greens that resulted from the methane gas leak in August 2008. We then test to see if any indirect or contagion effects are felt at the other landfills that were publicly put on notice regarding potential impacts from methane. As homes surrounding the other landfills were not directly impacted by the initial methane leak in Brookland Greens, any impact on prices surrounding these sites may be attributable to non- spatial contagion. N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 2 9 Exhibit 1 u The History of the Brookland Greens Estate Year Incidents and Actions 1996 Landfill at Stevenson’s Road operating as a landfill. 1999 Peet & Co. Casey Land Syndicate Limited (PEET) began construction of Brookland Greens Estate. 2000 Amendment to planning scheme to rezone land to Residential 1 to facilitate development of the estate; and the Section 173 agreement between Council and PEET stipulated no homes to be built within a 200-meter buffer of the landfill. 2004 Victorian Civil and Administrative Tribunal (VCAT) hearing that PEET seeking to develop Stage 10 of the estate, and this land was situated within the 200m landfill buffer. VCAT approved the reduction of the buffer along the western side of the landfill boundary. 2005 Landfill ceased operation. March 2006 EPA notified of bubbling stormwater and testing confirmed landfill gas containing methane and carbon dioxide, which was migrating from the landfill into the estate. EPA issued a pollution abatement notice. August 2008 Home in Brookland Greens was detected as having Methane gas detected at 63% volume for volume of air. September 2008 Residents in the Brookland Greens Estate were evacuated by the Country Fire Authority as the EPA had advised that the methane gas levels posed an imminent danger to residents. October 31, 2008 The risk levels were downgraded and residents allowed to move back into their homes. September 2008 Complete cessation of demand for rental accommodation in the Brookland Greens Estate was noted; future residents who were building on their land either ceased building activity or suspended construction contracts; businesses operating in homes were unable to continue; residents were unable to obtain or renew policies of insurance for their homes or property in Brookland Greens Estate. 277 homes across the Brookland Greens Estate had landfill gas/methane gas-in house monitoring installed; physical works to 252 homes in the estate were undertaken to mitigate the migration of land fill gas; and physical works comprising the construction of a bentonite diaphragm wall on the perimeter of the landfill (completed September 2009). November 2008 Slater and Gordon Lawyers lodged class action on behalf of residents. May 2009 Mediation was conducted at the County Court and also the Federal Court. February 2011 Date of hearing 1 February 2011 and Date of Ruling, February 3, 2011. March 2011 Residents win $23.5 million AUD compensation for the gas leak fiasco, $10 million to be paid by the EPA and $13.5 million to be paid by Casey Council. Residents share $17.25 million after $6 million payment for court costs and legal fees. This equates to approximately payouts from $7,000 to $132,000 for the worst affected households. Note: The sources are Brouwer (2009) and Wheelan v. City Council of Casey & Ors (No. 3). 2011. J O S R E V o l . 9 2 0 1 7 u u 3 0 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s Further, we test two different types of landfills: (1) closed and capped and (2) open and active. The sites are distributed throughout the Melbourne area, some in gentrifying neighborhoods, others near industrial areas, and some in more rural parts. While closed, capped, and redeveloped make up the bulk of our analysis, one of the capped, Narre Warren, consisted of undeveloped vacant land at the time the notice was issued. Further, it is important to note that the Brookland Greens case is something of an anomaly in the Melbourne area—the landfill had ceased operations when the contamination was discovered but still possessed a license to operate. The short time frame between those events (about nine months) coupled with homes being within 200 meters of an active, operating landfill for several years prior to the cessation of operations contributes to the unique qualities of the case. Because we recognize that comparing Brookland Greens to closed and capped fills, some of which provide amenities such as parkland to local communities, may not sufficiently address the pertinent questions, evaluation of active landfills is also required. The problem with active landfills is that few are positioned within 500 meters of a residential neighborhood, the minimum distance recommended by the EPA in their 2001 report. This is to be expected; most developers and homebuyers are wary of investing in properties that may be subject to environmental disamenities, as previously discussed. That said, one site on the list of 35, the Clayton South landfill, is positioned appropriately for analysis. Clayton South is about 32 kilometers from Brookland Greens and is also located in the southeast region of Melbourne. The area is generally less rural than the City of Casey, but would not be considered densely developed. We use a hedonic modeling framework to test for both the direct and contagion effects. The hedonic model is the method traditionally used to determine the impact of one or more characteristics on the price of housing, thus it is the preferred method for this research. It has been used widely to establish the value of a variety of (dis)amenities since Rosen’s (1974) seminal work on the subject. We use it to compare the prices of homes within 500 meters—the notification distance and the distance recommended by the EPA to minimize the risk of exposure to ‘‘dust, noise, windblown litter, and other emissions in relation to the landfill site (EPA, 2001)—to those between 500 meters to 2,500 meters of the landfills. As there are differences in prices between the areas near to the landfills and those farther away prior to the leak and the subsequent announcement, we approach the analysis through a before–after comparison. First, differences in prices between the affected (,500 m) and the control (500 m , 2,500 m) for the period before the leak are estimated. This pre-existing difference is then subtracted from any observed difference post-leak to determine the change in price differentials (if any) due to the leak and / or non-spatial contagion. As an example, if the homes around landfill X sold for 3% less prior to the leak and 12% less after the leak, we would conclude a change in value, or an impact due to the leak of 29%. Both the pre-leak (before) and post-leak (after) periods cover a 32-month span. The before period is from January 1, 2006 to August 31, 2008 and the after period is from September 1, 2008 to April 20, 2011, the date of the Brookland Greens legal settlement. N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 3 1 We use a traditional semi-log hedonic pricing model approach: ln( price) 5 a 1 bX 1 g t 1 lL 1 «, (1) where price is the observed sales price, X is a vector of physical home characteristics such as number of bedrooms and lot size, t is a vector of time dummy variables (quarter), and L is a dummy variable indicating whether or not the home is located within the 500 meter buffer zone of the landfill. One model is estimated before the leak date (August 31, 2008) and one after. The coefficients for the L dummy variables, l, from the before and after models are then compared to determine the change in relative price impacts of proximity to the landfill after the leak. This two model, before–after approach is executed first on the Brookland Greens (direct) case and then on the remaining potential contagion landfills. Due to data limitations and landfill locations, only 10 of the 35 notified landfills had enough surrounding residential property sales to estimate potential contagion effects from the initial leak and the subsequent announcement. We began our analysis by attempting to locate the exact locations of the landfills in the original EPA list. Of the 35 listed sites, 30 were located to an exact parcel(s) level. Of these 30, 9 have been closed and capped and have residential neighborhoods or estates located within 500 meters of the exterior boundary of the landfill and had at least 50 sales of residential properties during the study period. Only one (Clayton South) met the open landfill criteria with the requisite amount of property sales. Exhibit 2 shows the location of each landfill in the Melbourne Metropolitan Area. The majority of the landfills examined in this study had been, in 2008, redeveloped by the city councils into public amenities such as parks, gardens, sporting facilities, and the like. Consequently, public awareness of landfills in the area might be minimal, because the current uses are considered amenities. Of the sites we investigated, two were vacant and had not been redeveloped and one was partially redeveloped and partially vacant. The others were parks of varying sorts. Details regarding the redevelopment of these sites are found in Exhibit A2 in the Appendix. D a t a Our data includes all sales of single-family residential properties from 2006 to 2011. The source of the data is the Australian Property Monitors (APM), obtained through the Australian Urban Research Infrastructure Network (AURIN). The data includes a limited number of characteristics about the physical attributes of the homes: number of bedrooms, number of bathrooms, size of the lot (m ), and the presence of a study. Living area of the home as well as judgments regarding the quality or condition of the home are not commonly found in property data in Australia; this dataset is not an exception. J O S R E V o l . 9 2 0 1 7 u u 3 2 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s Exhibit 2 u Landfill Location Map Unfortunately, the exact latitude and longitude for each home are not available. Home location is available only at the mesh block level. Home sales within mesh blocks that are located within 2,500 meters of the outer boundary of each landfill make up our study area. The lack of exact geolocations does impact our choice of model functions as we are not able to estimate more advanced spatial autoregressive specifications. Exhibit 3 is an example of the size of mesh blocks, which generally contain 30 to 60 homes. As Exhibit 3 shows, the landfills are widely spread over the Melbourne Metropolitan Area. Home prices in Melbourne also vary considerably based on location, with the outer suburban areas offering markedly cheaper housing than inner city locations. These differences in prices are highlighted in Exhibit 4. Additionally, inner city locations such as Northcote and Farnsworth are composed of smaller homes on smaller lots. R e s u l t s The investigation begins with a univariate analysis of prices surrounding each landfill both before and after the announcement date. In most cases, exceptions being Altona, Brookland Greens, and Scotch, the prices in the affected (,500 m) and control (500 m to 2,500 m) areas during the before time period are relatively N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 3 3 Exhibit 3 u Mesh Blocks in the Brookland Greens Area Exhibit 4 u Summary Statistics: Mean Values Sales Price Bedrooms Bathrooms Lot Size (sq. m) Brookland Greens $319,799 3.48 1.80 628 Altona $431,107 3.17 1.56 549 Box Hill $784,710 3.18 1.57 610 Clayton South $448,387 3.21 1.47 579 Craigieburn $341,105 3.45 1.89 538 Farnsworth $547,721 2.77 1.34 370 Green Gully $392,302 3.35 1.63 649 Llewellyn $491,688 3.71 1.90 705 Narre Warren $374,988 3.44 1.92 632 Northcote $670,923 2.70 1.27 393 Scotch $492,946 3.06 1.62 523 J O S R E V o l . 9 2 0 1 7 u u 3 4 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s Exhibit 5 u Univariate Price Analysis Before After Change Affected Control Affected Control Affected Control Brooklyn Greens $377,675 $296,475 $420,307 $313,407 11% 6% Altona $440,500 $383,725 $499,166 $466,532 13% 22% Box Hill $652,611 $690,900 $1,049,898 $903,102 61% 31% Clayton South $343,390 $391,401 $440,907 $502,948 28% 28% Craigieburn $280,364 $305,082 $330,950 $362,321 18% 19% Farnsworth $519,305 $475,474 $697,734 $606,782 34% 28% Green Gully $329,878 $365,251 $399,998 $412,463 21% 13% Llewellyn $427,761 $431,432 $530,506 $546,955 24% 27% Narre Warren $387,202 $336,460 $428,026 $399,458 11% 19% Northcote $574,909 $594,570 $745,312 $748,206 30% 26% Scotch $364,037 $472,786 $430,789 $544,863 18% 15% similar. The existence of price dissimilarities, however, means that we will have to properly control for pre-existing differences in the multi-variate regression analysis. As seen in Exhibit 5, most areas (both control and affected) saw average prices move around 15% to 30% from the before and after periods. Brookland Greens offers a noticeable departure from this trend as both the affected and control area mean prices only changed 11% and 6% between these two periods. Other notable deviations from the norm include the high price increases in Box Hill and Farnsworth, two areas undergoing rapid gentrification. We also examine the ‘‘stickiness’’ of house prices in the affected areas. Potential sellers may not simply reduce price to a level at which a buyer can be found, instead—especially if they are owner occupiers—they may choose to forego sale and stay in the home until the market improves (Case and Shiller, 2003; Miles, 2008). As a result, changes in the volume of sales may also signify impacts on markets due to contagion and proximity to a perceived disamenity. Exhibit 6 shows the sales volumes in the affected (,500 m) and control (500 m , 2,500 m) area around each landfill both before and after the leak announcement. Overall sales volumes within 500 meters of the Brookland Greens site declined 18% in the period after the announcement, suggesting a softening of the market. Sales volumes also declined in 7 of the 10 other contagion sites following the leak. While the univariate analysis may provide interesting context to the situation, it does not account for potential changes in the types of houses that sold in the area or the specific times during the 32-month before and after period that homes sold. To better account for these potential differences, an estimate of a set of hedonic N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 3 5 J O S R E V o l . 9 2 0 1 7 u u Exhibit 6 u Sales Volume by Case Jan. 2006 to August 2008 Sept. 2008 to April 2011 ,500 m 500 m , 2,500 m % in 500 m ,500 m 500 m , 2,500 m % in 500 m Change Brooklyn Greens 113 74 60.43% 184 253 42.11% 218.32% Altona 110 344 24.23% 131 461 22.13% 22.10% Box Hill 311 703 30.67% 210 516 28.93% 21.74% Clayton South 23 878 2.55% 31 976 3.08% 0.53% Craigieburn 156 394 28.36% 413 805 33.91% 5.54% Farnsworth 498 864 36.56% 503 955 34.50% 22.06% Green Gully 204 287 41.55% 278 560 33.17% 28.37% Llewellyn 217 271 44.47% 215 359 37.46% 27.01% Narre Warren 289 468 38.18% 315 554 36.25% 21.93% Northcote 595 950 38.51% 627 962 39.46% 0.95% Scotch 407 306 57.08% 466 400 53.81% 23.27% 3 6 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s Exhibit 7 u Brookland Greens Model Results Before After Coeff. Std. Error Coeff. Std. Error Intercept 10.90* 0.2659* 11.0678* 0.1191* Affected Area (,500 m) 0.1236* 0.0268* 0.1064* 0.0191* Bathrooms 0.0594* 0.0254* 0.1098* 0.0113* Bedrooms 0.0412* 0.0159* 0.1731* 0.0178* Lot Size (log) 0.2186* 0.0356* 0.0435* 0.0098* Study 0.0254* 0.0212* 0.059* 0.0119* Monthly Time Dummies Not shown Not shown R 0.5396 0.5952 s 0.1291 0.1207 Note: *Significant at 0.05 level. price models is used to measure the relative changes in price in the affected area after the announcement. Hedonic Price Models In the hedonic price modeling exercise, we analyze the impact of the landfill contamination on the Brookland Greens housing estate. The results of both the before and the after models are shown in Exhibit 7. Overall, the models offer a relatively low R-squared; however, the standard errors are between 12.9% and 12.1%. These low standard errors suggest that the models do fit the data well. The coefficients all have the expected sign, with the exception of a few insignificant coefficients on some of the structural characteristics. Note that the presence of a study in the house acts primarily as a proxy variable indicating a higher quality home or one that has been remodeled (as a direct quality variable is not available in our data). Brookland Greens Looking specifically at Brookland Greens, it is apparent that in the before time period, the area within 500 meters of the landfill was worth, on average, 13.2% more than the control area. After the announcement of the methane leak, this difference dropped to 11.2%, a decrease in relative prices of 1.9%. This decrease represents the average difference in relative value between the two periods, before and after the announcement date. Capped and Redeveloped Landfills. Next, the same methodology is employed on the other ten notified landfills. The results of this analysis are shown in Exhibit 8. For five out of the nine landfills, there was no statistically significant difference N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 3 7 Exhibit 8 u Results Change in Relative Value Landfill Before After after Announcement Capped Sites Altona None None None Box Hill None None None Craigieburn 28.11% 25.62% 2.49% Farnsworth None 3.50% 3.50% Green Gully None None None Llewellyn None None None Narre Warren 5.33% None 25.33% Northcote None None None Scotch 215.67% 212.96% 22.71% Open Sites Clayton South None 28.26% 28.26% in prices (once home characteristics were controlled) between affected and control areas in either the before or after time periods. In the remaining five cases, three sites—Craigieburn, Farnsworth, and Scotch—showed moderate price increases while prices near Narre Warren decreased 5.3%, relatively speaking, in the time period after the announcement date. Open Fill: Clayton South. When the analysis is replicated for the Clayton South site, the results show that the homes within the 500-meter buffer are selling at an 8% discount, post notification. When the affected distance is increased to ,1,000 meters, a decrease in value of 0.79% is observed. C o n c l u s i o n Our original research question is based on an expectation that when information regarding an unexpected, detrimental event, in this case the discovery of methane gas at a housing development adjacent to a landfill, is released, a diminution in value will occur. We further posit that effects may also be present at similarly situated sites that are not directly affected by the event (i.e., non-spatial contagion). The analysis of the direct event involving the methane leak in the Brookland Greens Estate shows a 1.9% relative decline in prices in the period after information regarding the leak was made public. This diminution in value is expected as the methane leak represents an identified disamenity for homeowners in the area. The test for non-spatial contagion in the nine capped landfills and one open site located in heavily residential areas throughout the metropolitan region identified no systematic reduction in residential house prices during the period immediately after the event. Two of the sites, Narre Warren and Clayton South, show declines J O S R E V o l . 9 2 0 1 7 u u 3 8 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s in prices, while three others show increases. Five of the sites exhibited no change in relative values between homes near the closed landfills and those farther away. In other words, there appears to have been no measurable, sustained contagion effect in the closed and capped market around landfills. There are a number of reasons for this null finding. First, many of the covered landfills are currently operating as environmental amenities—parks, sporting facilities, and golf courses—and as such their prior use as landfills may not be evident to potential homebuyers. Even if potential homebuyers have been made aware of recent issues with capped landfills generally, they may not realize a fill is near the property they are buying. Sellers, agents, and local officials are under no obligation to disclose the issue to buyers, therefore information diffusion is likely low in these cases (provided the property is not directly affected by negotiations between the city council and developer as would be noted in a Section 173 Agreement). Limitations to the data may also reduce the reliability of the results as the inability to accurately locate home sales to exact geographic locations could be obscuring actual diminutive effects. In other words, homes located within a few hundred meters maybe be negatively affected. However, this impact is overwhelmed by the lack of impact from homes farther away, all of which are aggregated to the mesh block group level. One of the basic assumptions flowing through this study is that local residents’ knowledge about contaminated land is similar to that of professionals—they react to news of pollutants systematically. This implies that communities react to landfills in the same way (i.e., the reactions are homogeneous and directly comparable). This study is based in a number of diverse communities with a variety of (dis)amenities present. These local factors may be overriding any evidence of non-spatial contagion from the Brookland Greens site, if the contagion even exists. Additionally, because landfill sites are normally developed post- closure and not prior, there may be less of a social impetus and fewer communication networks to motivate collective fear and price drops. Even though the sites were all put on notice by the EPA and a fair bit of information was made available via formal networks (e.g., local newspapers, television newscasts), it is unclear how much of this information effectively reached the affected population. It is possible that the areas that sustained a negative impact (Clayton South and Narre Warren) had stronger and more viable social networks to transfer information, thus contagion effects may be evident there. It is also possible that the buyers interested in these two suburbs were more risk adverse with more alternatives available to them. The evaluation of the closed and capped sites provides interesting insight into ‘‘who knew what when.’’ Local councils in Melbourne are arguably skillful and motivated to convert and cover landfill site surfaces. Many of the closed and capped sites have been redeveloped into parkland, golf courses or other amenities that the casual observer would never recognize as a former landfill. It is rational for local councils, developers, and the EPA to control circulation of or passively slow down information to counteract social networks and whether intended or not, N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 3 9 capping and covering a landfill does just that. The removal of visual evidence has the potential to impair social memory, negatively impacting any associated communication networks, or to simply create new and different ones. While we recognize that two cases do not provide conclusive evidence of a systematic pattern of contagion, the results for Clayton South and Narre Warren are interesting. As an open landfill, Clayton South possesses the characteristics that suggest it is most like Brookland Greens. When analyzed, there is evidence, albeit limited, that the contagion sparked by the Brookland Greens case may be present. While sales volume increased modestly (0.53%), property values were depressed by over 8%. It is entirely plausible that the increase in volume was motivated by the relative decrease in price. Narre Warren was the only capped site that consisted of vacant land that had not been redeveloped and the surrounding properties show a diminution in value, along with a reduction in sales volume. Neither Clayton South nor Narre Warren possessed the amenity effect of the capped and redeveloped fills that were parks or sporting facilities at the time of notification. This suggests that the effects of a single event, regardless of how detrimental they are perceived to be at one site, are not distributed equally across markets. A p p e n d i x Exhibit A1 u EPA Designated Landfill Types Landfill Type Landfill Description Buffer in Operation Buffer Post-closure Type 1 Prescribed industrial Not disclosed as part of different guidelines wastes. Type 2 Putrescible (municipal) 500 meters from 500 meters from waste, solid inert waste buildings and structures building or structures and fill material. Can include category C industrial waste if licensed. Type 3 Solid inert waste, fill 200 meters from 200 meters from material. buildings and structures buildings and structures Note: In the VCAT hearing in regard to the Stevenson’s Road Landfill site and the Brookland Greens Estate, VCAT ruled that the 200-meter buffer distance was to be measured from the ‘‘active tipping area,’’ being the ‘‘active tipping face’’ of the landfill. J O S R E V o l . 9 2 0 1 7 u u 4 0 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s Exhibit A2 u Landfills Date Ceased Landfill Site Status Operations 2008 Use Comments Queen Street Closed 1998 Park 75 hectares of parks and wetlands. Altona 14 Federation Closed 1999 Park Parkland, sporting reserves and vacant St., Box Hill land, 2006 permit granted for development of residential to part of the site. Craigieburn Closed Unknown Park In 2007 the site was already developed, as Gardens, a park, partial leisure center and community Craigieburn facilities. BBQ picnic area, shelter, not well maintained, mostly unstructured open space. There are no known records of the type, depth, compaction or other treatment of fill in the landfill (Hume City Council, 2007). Quarry Park, Closed Park Parkland, nearby city college, and other Farnsworth sporting amenities. Ave., Footscray Clarke Rd., Closed 2002 East of High levels of methane detected in bored Springvale Clarke Rd. holes adjacent to community building and South is Park, residences on southern boundary. This was west of a capped (or poorly capped landfill at this Clarke Rd., time) EPA issued a new notice in Dec 2008. vacant land It was noted that although permissions for use as a rubbish tip expired in 1998, this site was used up to 2009 to stockpile scrap concrete (EPA, 2009). 151 Green Closed Unknown Park The Green Gully Reserve is a 69 hectare Gully Road, approximately site and has been established for some time, Keillor Downs 30 years ago noted as the Council’s second largest (Brimbank City recreational sites with six sports grounds. Council, 2013) Landfill gas issues are not considered to be a problem given the age of the landfill (Brimbank City Council, 2013). Llewellyn Park, Closed 1980s Park and The parklands and reserve were developed Wantirna sporting after the closure; however, exact dates South reserve cannot be located. Residents were notified due to high levels of gas post 2008 Stevenson’s Road report. Quarry Road, Closed 1996 Vacant land This site was previously a Type 2 landfill Narre Warren operating since 1982, it was used as a North landfill gas power station from 1992 until 1996 when it was closed. Clifton Road, Closed Early 2000s Park Former quarry used as a local landfill, Northcote classified as a Type 2 landfill. Redeveloped into major 13 hectare regional park in 2002 known as ‘‘All Nations Park.’’ N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 4 1 Exhibit A2 u (continued) Landfills Date Ceased Landfill Site Status Operations 2008 Use Comments Scotch Pde, Closed 1970s Park The park was developed in the 1980’s and Chelsea was completed in 1988. This site has several redevelopments with other facilities included over time with sporting fields, community center and clubs. Stevenson’s Closed 2005 Vacant land Residents are unaware of landfill in contract Road, of sale and S173. Cranbourne Clayton South Open N/A Landfill Type Residents are aware of landfill and make 2 regular complaints about the odor, with regular updates and community newsletters sent via the EPA and City of Kingston. Furthermore the local ‘‘Leader’’ newspaper has regular articles on the landfill sites in the area. E n d n o t e s The EPA provides a number of guideline documents in regard to landfill sites in Victoria. There are two main documents: Best Practice Environmental Management—Siting, Design, Operation and Rehabilitation of Landfills and Industrial Waste Resource Guidelines. These guidelines classify the type of landfill, and what type of landfill can be disposed of, consequently the guidelines regarding siting, design, operation, and rehabilitation change depending on the classification. The purpose of the buffers is to manage the odors from the landfill (a key concern during the operational phase of a landfill) and landfill gas, which may be an explosion risk and / or asphyxiation risk up to 30 years post closure of the landfill) (EPA, 2015). In the 1991 State Environment Protection Policy (Siting and Management of Landfills Receiving Municipal Wastes) document produced by the EPA, the required distance was 200 meters but due to the number of complaints regarding odors from various fills, the EPA was motivated to change that distance to 500 meters. Residential land subdivisions in Australia are commonly set up to be completed in stages, particularly when they are large subdivisions or estates. This allows efficient site development and sale of lots in a process that provides a feeding of lots into the market without flooding it and creating an oversupply. Most importantly, the staged process assists the funding of the development, as when a stage is complete the land lots can be settled, providing the developer with income and cash flow to continue and fund the following stages. In Victoria, the sale of land provides a contract in which all salient details of a property are to be disclosed. The Section 173 Agreement includes any agreements between a developer and the city council, which are then incorporated into any further contracts for J O S R E V o l . 9 2 0 1 7 u u 4 2 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s the sale of land. The S173 Agreement for the Brookland Greens development failed to disclose the site’s vicinity to the landfill, referring to it instead as ‘‘sand extraction operations.’’ After properly adjusting for dummy variables derived from a semi-log model; exp(c) 2 1 (Halvorsen and Palmquist, 1980). www.apmpropertydata.com.au. www.aurin.org.au. Mesh blocks are small geographic aggregations containing about 50 homes used by the Australian Bureau of Statistics for census purposes. Individual regression results are available upon request. R e f e r e n c e s Benson, E.D., J.L. Hansen, A.L. Schwartz, and G.T. Smersh. Pricing Residential Amenities: The Value of a View. The Journal of Real Estate Finance and Economics, 1998, 16:1, 55– Bolitzer, B. and N. Netusil. The Impact of Open Spaces on Property Values in Portland, Oregon. Journal of Environmental Management, 2000, 59, 185–93. Bond, S. The Use of Conjoint Analysis to Assess the Impact of Environmental Stigma. Pacific Rim Property Research Journal, 2001, 7:3, 182–94. Brasington, D.M. and D. Hite. Demand for Environmental Quality: A Spatial Hedonic Analysis. Regional Science and Urban Economics, 2003, 35:1, 57–82. Brimbank City Council. City of Brimbank Alternative Water Project, Final Report. Prepared for Department of Sustainability, Environment, Water, Population and Communities. August 1, 2013. Brouwer, G. Brookland Greens Estate—Investigation into Methane Gas Leaks. Ombudsman Victoria, October 2009. Caccioli, F.M. Shrestha, C. Moore, and J.D. Farmer. Stability of Financial Contagion Due to Overlapping Portfolios. Journal of Banking and Finance, 2014, 46, 233–45. Cameron, T.A. Directional Heterogeneity in Distance Profiles in Hedonic Property Value Models. Journal of Environmental Economics and Management, 2006, 51:1, 26–45. Caplan, A., T. Grijalva, and D. Jackson-Smith. Using Choice Question Formats to Determine Compensable Values: The Case of a Landfill-siting Process. Ecological Economics, 2007, 60:7, 834–46. Case, K.E. and R.J. Shiller. Is There a Bubble in the Housing Market? An Analysis. Brookings Papers on Economic Activity, September 2003. Chan, N. Stigma and its Assessment Methods. Pacific Rim Property Research Journal, 2001, 7:2, 126–40. ——. Stigma Assessment: A Multi-Criteria Decision-Making Approach. Pacific Rim Property Research Journal, 2002, 8:1, 29–47. Chen, M.C., H.J. Tsai, T.F. Sing, and C.Y. Yang. Contagion and Downside Risk in the REIT Market during the Subprime Mortgage Crisis. International Journal of Strategic Property Management, 2015, 19:1, 42–57. Cho, S., J.M Bowker, and W.M. Park. Measuring the Contribution of Water and Green Space Amenities to Housing Values: An Application and Comparison of Spatially Weighted Hedonic Models. Journal of Agricultural and Resource Economics, 2006, 485–507. N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 4 3 Cooper, M. Gas Leak Triggers Landfill Inquiry. Sydney Morning Herald, September 18, Do, Q. and G. Grudnitski. Golf Courses and Residential House Prices: An Empirical Examination. Journal of Real Estate Finance and Economics, 1995, 10:3, 261–70. Environmental Protection Agency. Siting, Design, Operation and Rehabilitation of Landfills. Best Practice Environmental Management Series. October, 2001. ——. Assessment of the Potential for Methane Gas Movement from Victorian Landfills. Environment report, Publication 1270. February, 2009. ——. Siting, Design, Operation and Rehabilitation of Landfills. Best Practice Environmental Management Series. Accessed 4 / 29 / 2016 from: http: / / www.epa.vic.gov.au/ our-work / publications / publication / 2015 / august / 788-3. August 12, 2015. Eshet, T., M. Baron, M. Shechter and O. Ayalon. Measuring Externalities of Waste Transfer Stations in Israel using Hedonic Pricing. Waste Management, 2007, 27:4, 614–25. Freybote, J. and E. Fruits. Perceived Environmental Risk, Media, and Residential Sales Prices. Journal of Real Estate Research, 2015, 37:2, 217–44. Ghosh, C., R. Guttery, and C.F. Sirmans. Contagion and REIT Stock Prices. Journal of Real Estate Research, 1998, 16:3, 389–400. Guntermann, K. Sanitary Landfills, Stigma and Industrial Land Values. Journal of Real Estate Research, 1995, 10:5, 531–42. Halvorsen, R. and R. Palmquist. The Interpretation of Dummy Variables in Semilogarithmic Equations. American Economic Review, 1980, 70, 474–75. Hanna, B.G. House Values, Incomes, and Industrial Pollution. Journal of Environmental Economics and Management, 2007, 54:1, 100–12. Hindsley, P., S.E. Hamilton, and O.A. Morgan. Gulf Views: Toward a Better Understanding of Viewshed Scope in Hedonic Property Models. The Journal of Real Estate Finance and Economics, 2013, 47:3, 489–505. Hite, D., W. Chern, F. Hitzhusen, and A. Randall. Property-value Impacts of an Environmental Disamenity: The Case of Landfills. The Journal of Real Estate Finance and Economics, 2001, 22, 2–3, 185–202. Hume City Council. Hume City Council Craigieburn Leisure Centre Master plan, December 2007. Accessed 4 / 29 / 2016 from: www.hume.vic.gov.au / files / bd48f2c0-267f... / Attachment2 CBMP.pdf. Ihlanfeldt, K.R. and L.O. Taylor. Externality Effects of Small-scale Hazardous Waste Sites: Evidence from Urban Commercial Property Markets. Journal of Environmental Economics and Management, 2004, 47:1, 117–39. Kiel, K.A. Environmental Contamination and House Values. College of The Holy Cross, Department of Economics, Faculty Research Series, Paper #06-01, 2006. Kiel, K.A. and K.T. McClain. The Effect of An Incinerator Siting on Housing Appreciation Rates. Journal of Urban Economics, 1995, 37:2, 311–23. Kiel, K.A. and J. Zabel. Estimating the Economic Benefits of Cleaning Up Superfund Sites: The Case of Woburn, Massachusetts. Journal of Real Estate Finance and Economics, 2001, 22:2, 163–84. Kohlhase, J.E. The Impact of Toxic Waste Sites on Housing Values. Journal of Urban Economics, 1991, 30:1, 1–26. McCluskey, J.J. and G.C. Rausser. Stigmatized Asset Value: Is It Temporary or Long-term. Review of Economics and Statistics, 2003a, 85:2, 276–85. J O S R E V o l . 9 2 0 1 7 u u 4 4 W i n s o n - G e i d e m a n , K r a u s e , W u , a n d W a r r e n - M y e r s ——. Hazardous Waste Sites and Housing Appreciation Rates. Journal of Environmental Economics and Management, 2003b, 45:1, 166–76. Messer, K.D., W.D. Schulze, K.F. Hackett, T.A. Cameron, and G.H. McClelland. Can Stigma Explain Large Property Value Losses? The Psychology and Economics of Superfund. Environmental & Resource Economics, 2006, 33:2, 299–324. Michaels, R.G. and V.K. Smith. Market Segmentation and Valuing Amenities with Hedonic Models: The Case of Hazardous Waste Sites. Journal of Urban Economics, 1990, 28:2, 223–42. Miles, W. Boom-Bust Cycles and Forecasting Performance of Linear and Nonlinear Models of House Prices. Journal of Real Estate Finance and Economics, 2008, 36:3, 249–64. Mink, M. and J. de Haan. Contagion during the Greek Sovereign Debt Crisis. Journal of International Money and Finance, 2013, 34, 102–13. Nelson, J.P. Hedonic Property Value Studies of Transportation Noise: Aircraft and Road Traffic. In: Hedonic Methods in Housing Markets. Springer, 2008. Nelson, A.C., J. Genereux, and M. Genereux. Price Effects of Landfills on House Values. Land Economics, 1992, 68:4, 359–65. Palmquist, R.B. and V.K. Smith. The Use of Hedonic Property Value Techniques for Policy and Litigation. In: International Yearbook of Environmental and Resource Economics 2002/ 3. Chapter 3, Edward Elgar, 2002. Reichert, A.K. Impact of a Toxic Waste Superfund Site on Property Values. Appraisal Journal, October, 1997, 381–92. Reichert, A.K., M. Small, and S. Mohanty. The Impact of Landfills on Residential Property Values. Journal of Real Estate Research, 1992, 7:3, 297–314. Rosen, S. Hedonic Prices and Implicit Markets: Product Differentiation in Pure Competition. Journal of Political Economy, 1974, 82:1, 34–55. Ryan, B. and B. Gross. The Diffusion of Hybrid Seed Corn in Two Iowa Communities. Rural Sociology, 1943, 8:1, 15. Schulz, M. and D. Barber. Residents Win Tens of Millions in Compensation after Cranbourne Methane Gas Leak Fiasco. Herald Sun, March 25, 2011. Seiler, M., A. Collins, and N. Fefferman. Strategic Mortgage Default in the Context of a Social Network: An Epidemiological Approach. Journal of Real Estate Research, 2013, 35:4, 445–75. Semintelli, A. and R. Simons. Regulation of Leaking Underground Storage Tanks: Policy Enforcement and Unintended Consequences. Economic Development Quarterly, 1997, 11:3, 236–48. Shultz, S.D. and N.J. Schmitz. Augmenting Housing Sales Data to Improve Hedonic Estimates of Golf Course Frontage. Journal of Real Estate Research, 2009, 31:1, 63–79. Simons, R., W. Bowen, and A. Sementelli. The Price and Liquidity Effects of UST Leaks from Gas Stations on Adjacent Contaminated Property. Appraisal Journal, 1999, 67, 186– Simons, R., W. Levin, and A. Sementelli. The Effect of Underground Storage Tanks on Residential Property Values in Cuyahoga County, Ohio. Journal of Real Estate Research, 1997, 14:1, 29–42. Simons, R. and J. 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. N o n - s p a t i a l C o n t a g i o n i n R e a l E s t a t e M a r k e t s 4 5 Simons, R. and K. Winson-Geideman. Determining Market Perceptions of Contamination of Residential Property Buyers using Contingent Valuation Surveys. Journal of Real Estate Research, 2005, 27:2, 193–220. th Wu, H. and C. Chen. Urban ‘‘Brownfields’’: An Australian Perspective. 18 Annual Pacific Rim Real Estate Society Conference, Adelaide, January, 15–18, 2012. All articles published in JOSRE are distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Kimberly Winson-Geideman, University of Melbourne, Parkville Victoria 3010, Australia or kimberly.winson@unimelb.edu.au. Andy Krause, Greenfield Advisors, Seattle, Washington or AKrause@ GreenfieldAdvisors. Hao Wu, University of Melbourne, Parkville Victoria 3010, Australia or haow@ unimelb.edu.au. Georgia Warren-Myers, University of Melbourne, Parkville Victoria 3010, Australia or g.warrenmyers@unimelb.edu.au. J O S R E V o l . 9 2 0 1 7 u u

Journal

Journal of Sustainable Real Estate – Taylor & Francis

Published: Nov 1, 2017

There are no references for this article.