Valuing urban trees: A hedonic investigation into tree canopy influence on property values across environmental and social contexts in Baltimore, Maryland
Introduction
Urban trees provide a multitude of well-documented benefits including stormwater regulation (Berland et al., 2017, Ow and Chow, 2021), urban heat island mitigation (Kim and Brown, 2021; Wang et al., 2019), air quality improvements (Li et al., 2013), wildlife habitat provision (Stagoll et al., 2012, Wood and Esaian, 2020), improved mental health outcomes (Wolf et al., 2020) and crime reduction (Troy et al., 2012). One commonly studied urban tree canopy influence is its beneficial impact on property values (Kovacs et al., 2022, Netusil et al., 2010, Pandit et al., 2013a), which acts as a proxy for homeowners’ willingness to pay for proximity to urban tree canopy. Urban tree canopy is not always associated with increased property values (Kovacs et al., 2022, Siriwardena et al., 2016). This study investigates the nature of that variation.
Researchers have used a variety of methods over time to measure the association of urban environmental amenities with home value, one of the most common being hedonic analysis (Anderson and Cordell, 1988, Mei et al., 2017; H Sander et al., 2010; Siriwardena et al., 2016). Hedonic models are widely used to statistically disaggregate the marginal value for characteristics of a home and its location, by regressing sale price against quantifiable attributes (Rosen, 1974). The resulting variable coefficients represent the marginal implicit of those variables, or the amount a given variable contributes to a home’s value holding all other variables constant. Hedonic models can also identify important interactions between variables (Hill, 2013).
Among the studies that have looked specifically at the price effects of urban tree canopy, Anderson and Cordell (1988) reviewed home sale listing photographs to demonstrate that trees were associated with a 3.5–4.5 % increase in homes sales price in Athens, Georgia. Focusing on mature trees, Dombrow et al. (2000) relied on realtor descriptions of mature trees in home sale listings to link mature trees to an approximately 2 % increase to home sale prices in Louisiana. Mansfield et al. (2005) used Landsat satellite images to calculate urban tree canopy coverage, using a hedonic model to assess the impact of various types of urban tree canopy on land value. Pandit et al., 2013a, Pandit et al., 2013b assessed the impact of both broad-leaved and palm trees in Australia using spatial hedonic models with the emphasis on the types of trees and their locations.
Few studies have addressed spatial or contextual variability in the price effects of environmental amenities. In one of the few examples, a meta-analysis by Kovacs et al. (2022) showed that the property value of homes rises more if tree canopy is not located on land that homeowners are responsible for maintaining. In another, for Los Angeles, authors Saphores and Li (2012) integrated block group level demographic data with canopy and property value analysis to estimate the value of urban trees, irrigated grass, and non-irrigated grass areas. They found that additional trees and grassy areas had varying impacts on the value of single family detached houses at the parcel or neighborhood levels. Other urban environmental amenities beyond trees have also been studied for their variability in implicit price. In Helsinki, researchers explored the marginal willingness to pay for urban green spaces across urban-core-to-fringe gradients using a hedonic pricing model with interactions, finding that different types of green space had varying impact on apartment pricing (Votsis, 2017). In Illinois, USA, authors Chen et al. (2022) assessed the spatial impacts of multimodal accessibility to green spaces on housing price. In Oslo, Norway, Łaszkiewicz et al. (2022) explored non-linearity of the marginal willingness to pay for urban green spaces, focusing on walking distance to green spaces and demonstrating how it varies with spatial context. This study captured the concept of “distance decay,” the reduction in willingness to pay for ecosystem services delivered at a distance from a residential location (Łaszkiewicz et al., 2022).
Clearly, more research is needed on the question of how urban canopy’s marginal price effects vary based on neighborhood and spatial context, as there are many potential factors that may moderate the marginal value of urban trees. Furthermore, little appears to be known about the value of urban trees in a humid maritime climate that characterizes Baltimore and many developing mid-sized cities. This paper is the first to explore urban tree canopy in Baltimore, MD in a hedonic property price model study. In this study we propose to assess how the marginal implicit price of urban tree canopy varies with four contextual factors: proximity to major roads and downtown, lot size, and crime rate. We selected moderator variables that we felt represented conditions of higher relative value for urban trees in terms of functions like privacy and mitigation of pollution, heat, and noise, among others. Several of these moderators had been previously established in the hedonic literature (Donovan and Prestemon, 2012a, François et al., 2002, Saphores and Li, 2012; Troy and Grove, 2008a, Troy and Grove, 2008b).
We hypothesize that proximity to major roads is associated with higher marginal implicit prices for canopy because road-adjacent trees buffer the effects of heat and pollution. There is evidence that living near a major road is associated with an increase in risk for low term birth weight (Dadvand et al., 2014) premature death from cardiopulmonary causes (Yang and Omaye, 2009) and headache, fatigue, and allergic rhinitis (Norbäck et al., 2019). With this evidence along with the understanding that living near a major road often considered a nuisance (H Sander et al., 2010, SA et al., 2010), there may be a premium on tree canopy near major roads.
We also investigate distance to downtown as a contextual factor. Research illustrates that downtowns are characterized by disproportionate levels of urban heat and pollution, which may interact to reduce quality of urban life (Li et al., 2018). Furthermore, the severity and damage of urban heat islands is increasing due to global climate change (Kim and Brown, 2021). We hypothesize that the value of urban trees downtown may be greater because urban trees shade and cool buildings and pavement surfaces, filter airborne pollution, reduce stormwater runoff, and provide beautification and wildlife habitat (Greene et al., 2018, Middel et al., 2015, Wood and Esaian, 2020). That marginal valuation may be heightened by the fact that trees are relatively scarce in downtowns due to planting space limitations (Jim, 2003).
In addition, because lot size has routinely served as an interaction term in property hedonic analyses (Cho et al., 2009; Donovan and Prestemon, 2012a, Donovan and Prestemon, 2012b), we assess whether the marginal implicit price of urban tree canopy may vary by this factor. On large lots, screening may be disproportionately valued by owners. Further, trees offer an opportunity to significantly differentiate large properties. Authors Saphores and Li (2012) found that tree canopy cover was more valuable on larger properties in Los Angeles than smaller ones. In Quebec, researchers found that residential trees added value up to a certain degree, but that having too many trees per square meter negatively affected property values (François et al., 2002). Climatic differences my account for variation between L.A. and Quebec.
In addition to environmental factors, there is debate as to how social factors such as crime rate may influence urban tree canopy marginal value (Donovan and Prestemon, 2012a, Donovan and Prestemon, 2012b; Schusler et al., 2018; Troy et al., 2012). Tree characteristics may determine how vegetation limits or encourages crime (Schusler et al., 2018). Low, dense vegetation is positively associated with crime risk, as it may obscure criminal activity (Michael et al., 2001). Similarly, researchers found that larger trees were associated with reduced crime and smaller trees were associated with increased crime (Donovan and Prestemon, 2012a, Donovan and Prestemon, 2012b). Urban tree canopy was found to be associated with lower crime rates in both Vancouver (Chen et al., 2016) and New Haven, Connecticut (Gilstad-Hayden et al., 2015). Troy et al. (2012) found that in most neighborhoods in Baltimore, urban tree canopy had an inverse relationship with crime rates due to trees’ motivating people to gather socially more outdoors, resulting in more “eyes on the street” (Jacobs, 1961), which in turn deters criminal activity (Kuo & Sullivan, 2001). Based on this, we hypothesize that residents of areas of high crime areas will likely value tree canopy more than residents in otherwise similar low-crime areas, given the acknowledged ability of trees to get more eyes on the street and mitigate crime.
Finally, this study is important from an environmental justice perspective. There is the potential for unequal increases in home value throughout Baltimore following tree planting and urban greening initiatives if there is a premium on the added privacy, reduced urban heat, noise and pollution that urban trees provide. This study does not address the causality of how the presence of trees may result in self-selection of certain types of home purchasers. However, given the increasing concern about how neighborhood greening can unintentionally lead to gentrification and displacement (Rigolon and Németh, 2020), it is important to study spatial contextual and patterns in the variation of marginal prices, as we do in this study.
Section snippets
Study area
Baltimore is the most populous city in the US state of Maryland, with approximately 621,000 residents in 92 square miles (238.3 square kilometers). Baltimore City is the center of the Baltimore-Townson Metropolitan Statistical Area which is home to nearly 3 million people. Baltimore is majority (62 %) Black (U.S. Census Bureau QuickFacts: Baltimore city, Maryland, 2021). It is in the Mid-Atlantic region on the East coast of the United States. Areas of Baltimore remain impoverished following
Results
The diagnostics in GeoDa clearly indicated the presence of spatial autocorrelation. The Moran’s I test was significant (I= 8.67, p = 0), revealing a high degree of global autocorrelation. The LaGrange Multiplier tests were all significant. However, the higher test statistic and significance level for the Robust LM-lag statistic (36.7, p = 0) compared to Robust LM-error statistic (3.04, p = 0.08) unequivocally indicated that the spatial lag model was the appropriate model for remedying the
Discussion
Studies by Anderson and Cordell (1988), Sander et al. (2010), and others illustrate how neighborhood tree cover reliably has a significant positive effect on home sale price across neighborhoods. While we find that canopy cover in the 100–400 m scale is associated with higher property values, our analysis validates research in Los Angeles (Saphores and Li, 2012) and five California counties (Mei et al., 2017) showing that these price effects vary depending on the socioeconomic and physical
Conclusion
We demonstrate with this study that increasing urban tree canopy does not hold the same amenity value across geographic contexts in Baltimore. We confirmed our hypothesis that urban tree canopy more strongly contributes to house prices in Baltimore areas closer to downtown, near major roadways, and on larger lots. However, we reject our hypothesis that crime rate is be associated with lower marginal price, as we did not find an association between crime rates and property values. Our findings
CRediT authorship contribution statement
Ashby Lavelle Sachs: Conceptualization, Methodology, Writing – original draft, Project administration. Angela E. Boag: Conceptualization, Formal analysis, Writing – original draft, Visualization. Austin Troy: Conceptualization, Formal analysis, Visualization, Methodology, Writing – review & editing, Supervision.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References (99)
- et al.
Influence of trees on residential property values in Athens, Georgia (U.S.A.): a survey based on actual sales prices
Landsc. Urban Plann.
(1988) - et al.
The role of trees in urban stormwater management
Landsc. Urban Plan.
(2017) - et al.
Spatial impacts of multimodal accessibility to green spaces on housing price in Cook County, Illinois
Urban For. Urban Green.
(2022) - et al.
Retail land use, neighborhood satisfaction and the urban forest: an investigation into the moderating and mediating effects of trees and shrubs
Landsc. Urban Plan.
(2006) - et al.
Research note: greater tree canopy cover is associated with lower rates of both violent and property crime in New Haven, CT
Landsc. Urban Plan.
(2015) - et al.
Canopy of advantage: who benefits most from city trees
J. Environ. Manag.
(2018) Protection of urban trees from trenching damage in compact city environments
Cities
(2003)- et al.
Urban heat island (UHI) intensity and magnitude estimations: a systematic literature review
Sci. Total Environ.
(2021) - et al.
Tree cover and property values in the United States: a national meta-analysis
Ecol. Econ.
(2022) - et al.
Valuing access to urban greenspace using non-linear distance decay in hedonic property pricing
Ecosyst. Serv.
(2022)