Do Multifamily unit Rents Increase in Response to Light Rail in the Pre-service Period?

Abstract

The effects of transit investments on land and housing values are a longstanding topic of interest in part because the nature and timing of those effects are important for designing anti-displacement and land value capture strategies. For these reasons, we explore whether multifamily unit rents have increased in planned station areas before the Purple Line light rail project in Maryland is operational. We employ a difference-in-difference (DID) approach to explore this question and validate the DID results with a first difference approach. We find that rents for units located within one-half mile of anticipated stations did increase well before transit service is expected to begin, but only for units with two or more bedrooms. We suggest these results imply that anti-displacement and land value capture strategies are warranted and potentially viable, but to be effective they need to be adopted well before transit service begins. Further, our results suggest that in the case of the Purple Line in Maryland, such policies should focus on units located within one-half mile of proposed stations and concentrate on preserving affordable units with two or more bedrooms.

Keywords

light rail transit housing Washington metropolitan area multifamily rent displacement

Introduction

The effects of transportation infrastructure investments on land and property values have long been a topic of research by economists, planners, and real estate scholars. For economists, the relationship between accessibility and land values undergirds much of urban economic theory in both mono- and polycentric models. For real estate scholars, accessibility is a key determinant of property values and rents, as implied by the adage: location, location, location. For planners, the coordination of land use and transportation remains the mantra for those who promote sustainability and smart growth (Knaap and Talen, 2005). But this relationship is of more than academic interest. Transportation infrastructure is expensive and unearned increases in land value are attractive sources of revenues for value capture strategies. Further, increases in land values are often the first step in the process of gentrification, where low cost housing is replaced by high cost housing and low-income residents are replaced by higher-income residents.

Despite the extensive research on this subject, the issue is far from settled. Most studies find that proximity to transit stations increases property values, retail sales, and commercial and residential rents, though some have found no effect and others found negative effects. These disparate results can be attributed to many factors, such as the type of transit service provided, the distance of the affected property from a station, the time between construction of the station and the impact on properties, and the methods used to analyse the relationship. Further, though the connection between investment in fixed rail transit and property values has been studied extensively, the link between rail transit and gentrification and displacement is underexplored and has shown conflicting results (Zuk et al., 2018).

In this paper we explore the relationship between multifamily rents and proximity to anticipated stations of the Purple Line light rail project in Montgomery County, Maryland, in the pre-service period. This relationship is important for two reasons. First, the Purple Line corridor is demographically and economically diverse, and includes several low-income neighbourhoods with high shares of minority and immigrant residents. These low-income neighbourhoods host a large portion of the Washington D.C. region’s most affordable housing. These residents could be displaced if rents rise to levels too high for their incomes to support (Slater, 2009). Second, the Purple Line is funded with county, state, and federal funds under a unique public-private partnership agreement. Montgomery County and neighbouring Prince George’s County, through which the Purple Line will run, are anxious to use land value capture instruments to fund investments in infrastructure, preserve low income housing, and mitigate gentrification. Thus, the effects of the Purple Line project on multifamily rents and timing of those effects are critical to the success of these strategies.

We proceed as follows. We first review previous research. Because there have been no empirical studies that explore the impact of light rail transit on multifamily rents in the pre-service period, we review studies that explore the impact of rail transit on land and housing prices in the pre-service periods and on rents in the post-service period. Next, we provide information on the Purple Line light rail project and its geographic and social context. We then describe the data we employ, the methods we apply, and the results of our analysis. Finally, we conclude with policy implications and recommendations.

Literature Review: The Impacts of Light Rail Transit in the Pre-service and Post Service Period on Property Prices and Rents

The literature on the effects on transit investments on property markets is voluminous. Although results vary, most studies find that investments in transit can increase residential and commercial property values near transit stations. As described by Higgins & Kanaroglou (2016) in a recent review, some studies examine effects on commercial property values, but most focus on single-family housing prices in the post-service period. Meta analyses by Debrezion et al.(2007) and Mohammad et al. (2013) suggest that the differences in findings of these effects can be attributed to differences in the type of transit service provided, the density and character of development in the surrounding area, the length of time transit service has been provided, differences in statistical methods, and more. In general, however, the evidence suggests that investments in transit generally increase property values near transit stations after service has begun, but market conditions and policy environments condition these effects. Some have also speculated that rent and price increases reflect less the value of transit services but more the corresponding investments in public infrastructure that often complement light rail investments (Chatman 2013).

Fewer studies have examined the effects of rail transit on rents in the post-service period (Table 1). Cervero (1994) explored the effects of rail transit investments on office rents, vacancy rates, and densities in Washington, D.C., and Atlanta and found rents higher and vacancies lower near transit stations than areas without a station. John (1996) found apartment rents also higher near rail stations in Washington D.C. in the post service period, and Bollinger et al. (1998) found office rents higher within walking distance of rail stations in Atlanta. In an international multi-site study, Hass-Klau et al. (2004) found housing rents and property values higher near transit stations than areas without a station in Greater Manchester, UK; Rouen, France; and Portland, Oregon. Weinberger (2001) found commercial rents higher within one half mile of a light rail station than areas without a station in Santa Clara, California.

Table 1.

Studies on the Impacts of Light Rail Transit in the Pre-service and Post Service Period on Property Prices and Rents.

Measurement period	Land and Housing prices	Housing rents
Pre-service	Agostini and Palmucci (2008) ^a	No studies in the existing literature This paper tries to fill this gap
	Boucq and Papon (2008) - or minimal
	Cao and Lou (2018) ^a
	Damm et al. (1980) ^a
	Devaux et al. (2017) ^a
	Dubé et al. (2018) ^a
	Gatzlaff and Smith (1993) - or minimal
	Henneberry (2007) - or minimal
	Hurst and West (2014) ^a
	Knaap et al. (2001) ^a
	Loomis et al. (2012) - or minimal
	McDonald and Osuji (1995) ^a
	McMillen and McDonald (2004) ^a
	Peng and Knaap (2023)
	Yan et al. (2012) - or minimal
Yen et al. (2018) ^a
Post-service	Al-Mosaind et al. (1993) ^a	Cervero (1994) ^a Bollinger et al. (1998) ^a Hass-Klau et al. (2004) ^a John (1996) ^a Nelson et al.(2015) ^a Sadayuki (2018) ^a Weinberger (2001) ^a
	Bae et al. (2003) ^a
	Cervero and Duncan (2002) ^a
	Chen et al. (1998) ^a
	Clower and Weinstein (2002) ^a
	Dubé et al. (2012) ^a
	Devaux et al. (2017) ^a
	Diao (2015) ^a
	Du and Mulley (2007) ^a
	Dueker and Bianco (1999) ^a
	Duncan (2008) ^a
	Duncan (2011) ^a
	Hass-Klau et al. (2004) ^a
	Weinstein and Clower (1999) ^a
	Welch et al.(2016) ^a

Note: ^aIndicates studies that find a positive effect of transit on property value, a “– or minimal” indicates studies that have a negative or minimal effect.

The number of studies on the effects of rail transit on land and housing prices in the pre-service period has grown in the last two decades. This growing body of evidence shows that land and/or property values begin to rise well before service begins. In an early study of Metrorail stations in Washington D.C., Damm et al. (1980) found single family, multifamily, and commercial property values were higher near Metro stations that were expected to open sooner. McDonald and Osuji (1995) found residential land values in the vicinity of the Midway line in Chicago began to increase two years before the line opened. In a more detailed study of the Midway line in Chicago, McMillen and McDonald (2004) found single family property values rose in the vicinity of light rail stations six years before the line began operation. Knaap et al. (2001) in a study in Washington County, Oregon, found vacant land values in station areas began to rise once the location of the stations were announced. Cao and Lou (2018) found the Green line in St. Paul, Minnesota, began to affect single family home prices after the station locations were announced and again after service began. Agostini and Palmucci (2008) in a study of the Santiago, Chile, metro system and found single family prices rose both after station locations were announced and when preliminary engineering began. Dubé et al. (2018) in a study of housing prices in Dijon, France, found prices did not increase in station areas until the construction phase.

Some studies, however, found the price impacts of rail transit investments to be minimal or negative in the pre-service period. Gatzlaff and Smith (1993) found single family home prices were only weakly impacted by the announcement of a new rail system in Miami, Florida. Henneberry (2007) found the announcement of the South Yorkshire Supertram in Sheffield, UK, to have negative impacts on housing prices. Yan et al. (2012) found a light rail investment in Charlotte, North Carolina, also had a negative impact on single family housing prices until the line became operational. Loomis et al. (2012) also found negative impacts during the construction phase and positive impacts after a new light rail line began operation in San Juan, Puerto Rico. Similarly, Boucq and Papon (2008) found no impacts on apartment prices of the T2 tramway in Hauts-de-Seine, France, until operation began. Yen et al. (2018) in a detailed analysis of the timing of the price impacts of the Gold Coast light rail line in Australia, found property prices in station areas start to increase after the announcement of station locations and again after a solid financial commitment was made by the government. Price escalation then slowed during construction and operation periods.

The literature further suggests that extent and timing of housing price and rent impacts have important implications for policies that address gentrification and displacement. Moeckel and Dawkins (2016) in a simulation exercise show that higher income households are likely to displace low income households in transit station areas. In a review of the empirical literature, Zuk et al. (2018) demonstrate that investments in transit not only raise housing prices and rents, but subsequently lead to gentrification and displacement. Turner and Snow (2001), furthermore, demonstrate that investments in transit are a leading indicator of neighborhood change. Pollack, Bluestone, and Billingham (2011) found increases in rents, household incomes, and vehicle ownership near transit in twelve US cities. And in an examination of 14 US cities that expanded transit systems between 1970 and 1990 Kahn (2007) found that transit-adjacent census tracts experienced disproportionate increases in property values and educational attainment. Deka (2017) analyzed changes in home values, rents, and race/ethnicity near rail transit in New Jersey, but found positive impacts only on home values.

In sum, there is ample evidence that investments in light rail transit lead to increases in housing prices and rents, but the extent, timing, and locations of those impacts often vary. There is also growing evidence that increases in housing prices and rents are a contributing cause of gentrification and displacement, which suggest that a better understanding of how light rail investments affect housing prices and rents is necessary for the design and implementation of effective anti-displacement policies to work. Finally, although there is evidence that housing prices rise in station areas even before transit service begins, there are few studies which investigate if rents of multifamily housing units rise in station areas in the preservice period. To best of our knowledge, this paper is the first to investigate preservice light rail transit effects on multifamily rents.

The Purple Line’s Planning and Construction Process

The Purple Line light rail transit system will provide transit service to the Maryland suburbs of Washington, D.C., and was expected to begin operation in 2022 but has been delayed due to construction challenges. Spanning 16.2 miles and 21 stations, it will directly link the suburban hubs of Bethesda, Silver Spring, College Park, and New Carrollton and provide a direct radial connection between the Red, Green, and Orange lines at existing stations of the Washington Metro system. It will also connect three stations of the Maryland Area Regional Commuter (MARC) rail system, and one stop on Amtrak’s northeast corridor at New Carrollton. The cost of construction is expected to exceed $2.3 billion and create 6300 jobs over the five-year construction period (Maryland Transit Administration, 2014). Estimated daily ridership by 2040 is expected to reach 74,000 per day (Maryland Transit Administration, 2014).

The project weathered two decades of political, legal and financial battles before construction began. The draft environmental impact study was completed in 2008 and the Federal Transit Administration issued a record of decision in 2014 (Maryland Transit Administration, 2014), with additional milestones thereafter. In March 2016, the Maryland Transit Administration announced the team of private companies that would build, operate, and maintain the Purple Line; in August 2017, $900 million of federal funding was officially granted by the Federal Transit Administration; in August 2017 construction began; and in December 2017, the U.S. Court of Appeals for the D.C. Circuit dismissed legal environmental challenges and allowed construction to proceed. These significant events happened between 2015 and 2018. A timeline of critical events in the history of the Purple Line is provided in Figure 1.¹

Figure 1.

Purple line planning and construction timeline.

Neighbourhoods That the LRT Will Traverse

Although the line will traverse Montgomery and Prince George’s counties in Maryland, we limit our analysis to the 10 stations in Montgomery County, which is home to over one million residents (U.S. Census Bureau, 2017) and several U.S. federal government agencies. Although the county is ranked 15^th nationally in median household income, tenants in multifamily properties are heavily cost burdened (Montgomery County, 2018). In 2015, the county ranked 9th in the nation with a median monthly rent of $1656 (Montgomery County, 2018). More than 49 percent of renter households in Montgomery County spend over 30 percent of their incomes on rent (U.S. Census Bureau, 2017).

The population of Montgomery County is diverse; more than one in three residents are foreign-born, many from Central America and Asia (U.S. Census Bureau, 2017). Socioeconomic diversity is especially pronounced in the Purple Line corridor (see Figure 2). In the census tracts along the corridor in Montgomery County median home values range from just under $400,000 to over $800,000; rents range from less than $1300 to over $2300; and median household incomes range from less than $80,000 to over $130,000 to (U.S. Census Bureau, 2017). High income, largely white neighbourhoods are concentrated in the west side of the corridor while low income, largely minority neighbourhoods are concentrated on the east side of the corridor (Figure 2).

Figure 2.

Characteristics of demographics and housing market in the rail transit corridor (left top: median home value; right top: median rent; left bottom: median household income; right bottom: proportion of foreign-born residents).

The Purple Line is widely expected to raise property values and rents, which has raised concerns about gentrification and displacement, especially on the east side of the corridor. For these reasons, the County has adopted a variety of anti-displacement policies, such as tenant protection laws, inclusionary zoning, and direct acquisition and preservation of affordable units. A unique coalition of interest groups formed the Purple Line Corridor Coalition (PLCC) in 2013. Through and extensive public participation process the PLCC drafted a Purple Line Community Development agreement that was signed by the county executives of both Prince George’s and Montgomery Counties. Among the four goals of that agreement was: “housing opportunities are available for people of all incomes in communities throughout the corridor, especially current low, middle-income, and transit-dependent residents” (Purple Line Corridor Coalition, 2017, p. 1). A subsequent Housing Action Plan prepared by the PLCC committed the coalition to constructing or preserving 17,000 units of affordable housing in the corridor (Purple Line Corridor Coalition, 2019). Members of the coalition recently received a $5 million grant to further the implementation of the housing plan (McCartney, 2019). In addition, county policy makers, including Montgomery County Executive Marc Elrich, have expressed considerable interest in capturing land value increments from the Purple Line investment to fund affordable housing programs or other infrastructure projects (M. Elrich 2019, personal communication, 29 July). For these reasons, the effects of the Purple Line on housing prices and rents are of considerable policy relevance.

Data and Methods

The data for our analysis come from three primary sources. The first source is the Montgomery County Department of Housing and Community Affairs, which by statute requires all owners of multifamily rental property to complete an annual rental housing survey. The earliest survey data available for public access is the 2015 rental housing survey, which was conducted and completed in April, 2015, and included 846 of 930 multifamily rental facilities. All facilities in the dataset are multi-unit apartment buildings (Peng and Knaap, 2021). In 2018, 923 of 1047 rental facilities were included in the survey.² In 2015, construction of the Purple Line had not yet begun, but multiple steps toward implementation had been completed, but by the completion of the 2018 survey, construction had begun. The second source is the Montgomery County Open Data portal, maintained by Montgomery County Government, which provides information on amenities such as the location of schools and other public facilities, the location of the proposed Purple Line stations, shopping centers, and more. The third source is the 2013 Maryland’s State Education Indicators Database, maintained by the Maryland State Department of Education; this database includes average reading scores of fifth graders for each public elementary school in the County.

Each multifamily unit in the survey is an observation. After we drop duplicate entries, vacant units, and outliers, we have a two-period pooled dataset of 137,547 observations in 793 multifamily complexes. We list definitions of variables in Table 2 and summary descriptive statistics of those variables in Table 3.

Table 2.

Definition of Variables.

Category	Variable Name	Definition
Dependent variable	Rent	Monthly rent of unit
Independent variable	Time	If an observation from 2018 survey, assign 1. Otherwise, 0.
	Half-mile treated	If a complex is within half mile from planned stations, assign 1. Otherwise, 0.
	One-mile treated	If the distance between a complex and planned stations is larger than half mile but less than one mile, assign 1. Otherwise, 0.
	Age of structure	Age of building structure
	Swimming pool	If a complex has a swimming pool, assign 1. Otherwise, 0.
	Gym	If a complex has a gym, assign 1. Otherwise, 0.
	Pet allowance in building	If a complex allows pets, assign 1. Otherwise, 0.
	Storage service	If a complex provide storage service, assign 1. Otherwise, 0.
	Shared laundry room	If there is a shared laundry room, assign 1. Otherwise, 0.
	Affordability program	If a complex has any housing affordability program, assign 1.Otherwise, 0.
	Neighboring cemetery	If a complex is less 0.25 mile from a cemetery, assign 1. Otherwise, 0.
	Shopping center	Distance between a complex and its nearest shopping center.
	5th reading score	Use average fifth grade reading score per school as indicator of performance of elementary school. The score varies from 0 to 100. The larger the score is, the better performance the school has.
	Foreign born percentage	Foreign born percentage at census tract level
	Household income	Median household income at census tract level
	Accessibility score by walking to jobs	An indicator to quantify accessibility of a block to jobs by walking travel mode
	Accessibility score by driving to jobs	An indicator to quantify accessibility of a block to jobs by auto travel mode
	Crime rate	Crime rate at the census tract level (2017). Crime rate indicates how many incidents (all types of police-recorded crime) occur per 100,000 people a year.
	Bus	Dummy variable indicating if an apartment complex’s distance to bus stop is less than 0.25 mile
	Structure type	A categorical variable indicating what structure type a complex is. There are four types of buildings: Garden apartments with 1–4 stories, midrise apartments with 5–8 stories, high-rise apartment with 9+ stories, and townhouses.
	Housing rental market (city)	A series of dummy variables indicating what rental housing market an apartment unit is in.

Table 3.

Summary Descriptive Statistics.

Variable Name	Mean	Std.Dev.	Min	Max
Rent	$1511	404	$154	$3470
Time	0.536	0.499	0	1
Half-mile treated	0.142	0.44	0	1
One-mile treated	0.028	0.225	0	1
Age of structure	38.9	20.2	1	133
Swimming pool	0.645	0.478	0	1
Gym	0.586	0.493	0	1
Pet allowance in building	0.769	0.422	0	1
Storage service	0.376	0.484	0	1
Shared laundry rooms	0.564	0.496	0	1
Providing affordability program	0.594	0.491	0	1
Neighboring cemetery	0.022	0.148	0	1
Shopping center(feet)	1926	1515	70	23,706
5^th reading score	91.4	6.29	73.0	100.0
Foreign born percentage	37.5	11.1	6.5	61.0
Household income	$81,396	26552	$42,955	$227,778
Accessibility score by walking to jobs	5968	5165	33	25,220
Accessibility score by driving to jobs	85,976,173	9,104,000	51,447,026	98,358,892
Crime rate	9,722	11,862	953	66,161

Note: The two-year pooled dataset has 137,547 observations in 793 complexes. The mean values of the dummy variables can be interpreted as percentages. The dummy variable, Gym, for example, equals 1 if an apartment has gym and is otherwise equal to 0. The value of Gym is 0.586, which means 58.6% apartment units offered gym facilities.

Each unit in the survey is an observation in our analysis. For each observation, we have information on the characteristics of the unit, amenities of the apartment complex where the unit sits, and social economic and accessibility characteristics of the neighbourhood where the apartment complex is located. For unit characteristics, unit rent is the total monthly rent that property managers or owners receive for the unit. These rents include both tenant payments and government subsidies. We exclude observations that are vacant or employee-occupied because the rents of these observations do not represent arms-length transactions.

Table 2 lists variables included in the analysis. For each observation, the variables contain the following: monthly rent of unit, a dummy variable indicating if an observation is from the 2018 survey, a dummy variable indicating whether the complex falls within the half-mile radial catchment area of a Purple Line station; a dummy variable indicating whether the apartment complex is between a half mile and one mile away from a Purple Line station; a continuous variable indicating the age of the complex; a categorical variable indicating structure type of the apartment complex; a dummy variable indicating if the complex has a swimming pool; a dummy variable indicating if a complex allows pets; a dummy variable indicating if the complex provides storage service; a dummy variable indicating if the complex provides a shared laundry room; a dummy variable indicating if the complex participates in any housing affordability program; a dummy variable indicating if the complex is less than 0.25 miles from a cemetery; a continuous variable indicating pertinent public elementary school quality measured using fifth grade reading scores; a continuous variable indicating distance between the complex and the nearest shopping center; a continuous variable indicating foreign born percentage of the neighbourhood where the complex is located; a continuous variable indicating median household income of the neighbourhood where the complex is located; a continuous variable indicating accessibility level to jobs by walking;³ a continuous variable indicating accessibility level to jobs by driving; a continuous variable indicating crime rate of the neighbourhood where the complex is located; and a series of dummy variables indicating which rental housing submarket (city) the complex is located.

The descriptive statistics of the variables are shown in Table 3. The mean monthly rent is $1,511, while 53.6% of the observations are from the 2018 survey. 14.2% of the observations are in the half-mile treated group, while 2.8% of the observations are in the one-mile treated group. The mean unit is 38.9 years old, feeds a school with a reading score of 91.4 for 5th grade, and is 1926 feet from the nearest shopping center. The mean values of other variables can be found in Table 3.

To test if the rents of units in the light rail transit corridor increased significantly, compared to others, we apply a difference-in-difference (DID) approach and present key results below. The DID approach may suffer from omitted variable bias, although we include as many control variables as we can into the DID models. To validate the DID results, we also apply a first difference approach and report the results in Appendix I. For both tests, we define two treatment groups: those units within a half-mile and those within one mile of a proposed station.⁴ Rental units within a half mile of a planned Purple Line station are included in the half-mile treated group. Rental units more than a half mile but less than one mile from a planned Purple Line station are included in the one-mile treated group (See Figure 3). Rental units elsewhere in the County are used as the control group.⁵

Figure 3.

The Treated Groups and the Control Group by bedroom types. (left top: Studio units; right top: 1-bedroom units; left bottom: 2-bedroom units; right bottom: 3&4-bedroom units).

Difference-in-Difference approach

The specification of the DID model is:

\begin{array}{l} p_{i} = α + β_{0} * T i m e_{i} + β_{1} * H a l f M i l e_{i} + β_{2} * (T i m e_{i} * H a l f M i l e_{i}) + \\ γ_{1} * O n e M i l e_{i} + γ_{2} * (T i m e_{i} * O n e M i l e_{i}) + ω^{T} * X_{i} + ε_{i}, ε_{i} \sim N (0, σ^{2}) \end{array}

(1)

where P_i is the rent of the unit and i indicates unit observation.⁶ The observation i is either from the 2015 survey or 2018 survey. Time _i is a dummy variable takes the value zero for observations from 2015 survey and a value of one for observations in 2018 survey. HalfMile_i is a dummy variable that indicates whether unit i is within a half mile of any Purple Line station. OneMile_i is a dummy variable that indicates whether unit i is more than a half mile and less than one mile from any Purple Line station;

ε_{i}

is the error term, which we assume to follow a normal distribution. X_i is a column vector of attributes of unit i.

α, β_{1}, β_{2} {, γ}_{1}, a n d γ_{2}

are corresponding coefficients;

ω

is a coefficient vector for X.

The control group consists of observations that are more than one mile away from all planned Purple Line stations. If the estimate of $β_{2}$ is positive and significant, units within the half-mile treated group saw a larger rent increase than their counterparts in the control group from 2015 to 2018. If the estimate of $γ_{2}$ is positive and significant, it means that units within one-mile treated group saw a larger rent increase than their counterparts in the control group.

There are three common concerns regarding the DID approach. First, the DID approach can suffer from omitted variable bias. To reduce this problem, we include as many control variables as possible, given data availability, in the DID model. Second, the DID approach may suffer from self-selection bias when dividing rental units into the treated groups and control groups based on their distance to planned rail stations. While stations were not randomly placed in space, the variation in the station-area contexts of the Purple Line stations is sufficient to permit a credible quasi-experimental design. Our pretests show that most of the characteristics of the rental units in the treated groups are comparable to the characteristics of rental units in the control group. To improve the quality of that design we performed the DID test and we have added a significant number of control variables to the DID models. Third, the DID approach could suffer from spatial autocorrelation. To account for spatial autocorrelation conventional spatial econometric models can be applied if pre-defined weight matrices are non-singular. However, a single apartment complex in our dataset could include dozens or hundreds of individual units in one location. Thus, the weight matrix based on unit geographical location is singular. Conventional spatial econometric models cannot be applied in this multifamily housing context. We also note that the observations are clustered by apartment complexes, and hence we correct for clustered standard errors in this analysis.

Findings

As a preliminary exploration we present in Figure 4 average rents for units in the half-mile treated group and the control group in 2015 and 2018. As shown average rents for each bedroom type in the half-mile treated group and the control group increased from 2015 to 2018. We pool together 3- and 4-bedroom units due to the limited number of observations of 4-bedroom units. Note that the average rents of 2-bedroom units and 3/4-bedroom unit in the half-mile treated group increased more than their counterparts in the control group. Figure 4 also shows that rents of 2-bedroom and 3/4-bedroom units in the half-mile LRT corridor increased substantially more than those in the control group. Of course, simple comparison of means does not prove causation. The rents near stations may have increased due to other factors, such as differences in social economic environment, school performance, or other location specific factors. To explore this further, we next estimate DID regressions with a several control variables for other factors to see if the premium on units in station areas still exists.

Figure 4.

Average rents comparison between treated group, control group, before, and after event (left top: half mile-treated group average rent in 2015 and 2018; right top: control group average rent in 2015 and 2018; left bottom: rents increase by bedroom type for both treated group and control group; right bottom: different in difference by bedroom type).

We estimate equation (1) and report key results in Table 4. Following Wolverton et al. (1999), who find rental housing markets segmented by number of bedroom, we analyse rents for four distinct unit-size submarkets (studio, one-bedroom, two-bedroom, and three/four-bedroom units). Column (1) to Column (4) report results that correspond to various subsamples. For example, Column (3) presents results based on the subsample of 2-bedroom units. All coefficients on control variables have expected signs and the models have R-squares with values above 0.5. The provision of some special services such as permitting pets and providing storage service increases rents.⁷ Apartment complex and neighborhood amenities such as swimming pools, gyms, and shopping centers also increase rents, while disamenities like proximity to cemeteries decrease rents. Public school quality, crime rates, job accessibility, and other neighbourhood social economic factors influence unit rents in directions we expected.

Table 4.

Results of the difference-in-difference models.

	0 bedroom (1)	1 bedroom (2)	2 bedroom (3)	3,4 bedroom (4)
Intercept( $α$ )	556.729 (10403.901)	−1382.308 (6666.536)	−2979.251 (6740.067)	16488.958 (12000.386)
Time ( $β_{0}$ )	81.387*** (26.865)	54.644*** (14.444)	67.047*** (8.757)	49.316*** (16.869)
HalfMile ( $β_{1}$ )	−73.541 (63.651)	42.572 (56.744)	−47.466 (37.661)	−117.848 (89.273)
OneMile ( $γ_{1}$ )	−64.800 (58.015)	135.371* (77.275)	38.385 (72.227)	30.634 (91.351)
$T i m e * H a l f M i l e$ ( $β_{2}$ )	35.938 (31.347)	12.849 (29.937)	37.675* (21.313)	104.940*** (37.468)
$T i m e * O n e M i l e$ ( $γ_{2}$ )	−1.572 (33.055)	1.003 (22.155)	6.068 (31.574)	50.750 (44.193)
Age	−2.383** (1.058)	−4.392*** (0.764)	−5.583*** (0.800)	−5.431** (2.749)
Swimming pool	52.877 (45.270)	148.534*** (33.803)	135.235*** (27.765)	144.471** (63.483)
Gym	133.903*** (37.706)	103.371*** (37.203)	3.166 (29.750)	76.596 (61.740)
Pet allowance in building	−23.125 (36.586)	5.741 (33.679)	52.570** (26.004)	120.488* (63.322)
Storage service	126.331*** (32.243)	82.699*** (28.462)	59.248** (23.980)	13.807 (53.988)
Shared laundry room	−12.285 (33.882)	−78.946*** (24.693)	−46.632** (20.430)	−41.815 (63.703)
Affordability program	−21.894 (42.523)	−134.759*** (26.108)	−86.968*** (19.923)	−144.105*** (48.600)
Neighboring cemetery	−397.760*** (111.524)	−78.201 (50.179)	−71.609 (44.666)	−211.772** (85.925)
Logarithm of distance to shopping center	−11.633 (18.270)	−29.153* (17.252)	−26.674* (14.807)	−38.485 (33.902)
5^th reading score	10.759*** (3.155)	2.538 (1.929)	2.937* (1.627)	6.743* (3.799)
Foreign born percentage	0.899 (2.969)	−2.104 (1.835)	−0.346 (1.615)	−7.272** (3.478)
Logarithm of accessibility score by walking to jobs	20.732 (47.623)	76.071*** (27.710)	113.006*** (21.883)	107.502*** (40.326)
Logarithm of accessibility score by driving to jobs	117.092 (548.208)	135.381 (355.805)	99.406 (360.589)	−793.093 (639.730)
Dummy indicating if distance to bus stop less than 0.25 mile	3.084 (24.819)	9.263 (27.218)	10.541 (19.949)	22.396 (47.892)
Log(crime rate)	−34.530 (19.404)	−31.961** (16.093)	14.708 (16.214)	−97.671* (46.760)
Control median household income at neighborhood level	Yes	Yes	Yes	Yes
Control housing rental market (city)	Yes	Yes	Yes	Yes
Control structure type	Yes	Yes	Yes	Yes
R-squared	0.578	0.504	0.550	0.584
Adjusted R-squared	0.575	0.503	0.549	0.582
Number of observations	6,737	57,485	61,836	11,489
Number of apartments	207	674	647	267

The dependent variable is rent for each unit. Standard error is in parentheses.

Units are clustering by complexes and standard errors are corrected.

***

p = 0.01(two-tailed). **p = 0.05(two-tailed). *p = 0.10(two-tailed).

Our primary interest lies in the coefficients on HalfMile and OneMile and Time*HalfMile and Time*OneMile. The coefficients on HalfMile and OneMile are insignificant,⁸ which suggest that in 2015 there was no observed effect of proximity to station on rents for all units regardless of bedroom size. We note that four significant events occurred between 2015 and 2018: the team of companies that were to build and operate the line was selected in 2016; $900 million in construction funding from the FTA was granted in 2017; a judge dismissed legal challenges in 2017; and construction officially began in 2017. These four events took place between the administration of the 2015 survey and 2018 survey and any one or some combination of these events could trigger the rise in rents.

To investigate that, we turn to the coefficients on the variables Time*HalfMile and Time*OneMile. We find that the coefficients on the variables Time*HalfMile and Time*OneMile are insignificant for zero- and one-bedroom units. This suggests that the start of construction had no impact on rents for zero- and one-bedroom units. The coefficients on the variables Time*HalfMile for the 2-bedroom, and 3/4-bedrooms subsamples are positive and significant with the magnitude of $37.68, and $104.94, respectively. These results imply that the rents of 2-bedroom, 3/4-bedroom units within one half mile of a planned Purple Line station all rose relative to units in the control group between 2015 and 2018. These increases in monthly rents are statistically significant at 90 percent level for 2-bedroom units and 99 percent level for 3/4-bedroom units. The coefficients on the variables Time*OneMile for the 2-bedroom, 3/4-bedroom unit subsamples, however, are insignificant. These results imply that the rents of 2-bedroom, and 3/4-bedroom units located more than one half mile of a planned Purple Line station but less than one mile from planned stations did not rise relative to units in the control group between 2015 and 2018. We corroborate these findings using a first difference approach and report the results in the appendix.

Implications of Findings

These results suggest that the rents of two-bedroom and three or more-bedroom units within a half mile of a planned Purple Line station increased more than a control group of units elsewhere in the county between 2015 and 2018—five years before transit service was expected to begin. These rent increases, compared to the control group, could impose costs ranging from about $450 per year for a 2-bedroom unit to over $1200 per year for a 3/4-bedroom unit. This raises two important questions. First, why would the rents of multifamily units within a half mile from the planned stations increase more than the control group even before service begins? And second, why would this rise occur only for two-, three-, and four-bedroom units?

The answer to the first question has two parts. The first is the anticipation effect. Plans for the Purple Line convey information about when and where light rail service will be provided; that information leads to expectations of higher rents in the future, and these future rents become capitalized into current property values. Several studies that examine property values provide evidence of this effect (Cao and Lou 2018; Dubé et al. 2018; Knaap et al. 2001). But this explanation cannot explain why rents would rise before service begins because there would be no transportation cost savings or gains in household utility before service begins. One could even argue that disruption brought by construction would cause utility and willingness to pay to fall in the pre-service period (Yan et al., 2012). Why would renters be willing to pay higher rents now? We suggest the second part of the answer lies in nontrivial moving costs. If moving costs were trivial and renters were able to move frequently and without cost, they would likely not be willing to pay more in rents for units that offer access to transit service until sometime in the future. But because moving is not costless, renters might be willing to pay more for units that will provide this benefit in the future. The American Communities Survey (ACS) also shows that renters in Montgomery County do not move very often.⁹ This is especially true for existing low-income immigrant residents that rely heavily on place-based cultural amenities in their neighbourhoods (see Appendix II). For this reason, landowners may be able to increase rents in the preservice period. Substantial anecdotal evidence and Figure 4 strongly suggest that this is the case.

The second question concerns why rents rose within a half mile of a station area in the pre-service period for two or more bedroom units and not for zero- and one-bedroom units. Again, we believe that answer has two parts. The first part stems again from nontrivial moving costs in terms of time and money—especially for larger households. Because moving costs for one-person households are relatively small, it is easier for such households to resist rising rents by moving to other locations. Moving costs for two-person households and for households with children, however, are not likely to be small. Larger households are thus more likely to prefer long-term leases¹⁰ and are more likely to pay more for units that offer new transit access sometime in the near future—especially in culturally supportive neighbourhoods.

The second part of the answer, we believe, lies in recent changes in the supply of apartment units. As part of our own work monitoring housing markets in the Purple Line corridor, we have observed that the supply of zero‐ and one-bedroom units increased more rapidly in recent years than the supply of two-, three-, and four-bedroom units. In fact, “very few new construction projects have any 3+ [or more] bedroom units at all” (Montgomery National Capital Park and Planning Commission, 2017, p. 7–7). It is likely that increases in supply have mitigated rent increases for units with fewer bedrooms, but not for units with more bedrooms. Further research could investigate the link between changes in the supply of rental housing and changes in multifamily rents more deeply.

Conclusions and Policy Implications

In this study we have used unusually detailed data on multifamily units and rents to explore the impacts of a planned investment in light rail on multi-family rents. Using the difference-in-difference and the first difference tests we find that multi-family rents did increase in the preservice period for units with two or more bedrooms within one-half mile of a planned station. We are not aware of any previous studies that have found similar or conflicting results.

We believe these results have important implications for policy makers in the Purple Line corridor and other places anticipating similar light rail investments. Some of these policy implications are now fairly well known but others have heretofore not been addressed in the planning, real estate, or regional science literature. First, our study adds to the mounting evidence that investments in transit tend to increase housing prices and rents in station areas. This lends credibility to the concern that such investments have the potential to displace residents that cannot afford higher housing prices and rents. That is, concerns about transit-induced gentrification are valid. Second, transit-induced increases in housing prices and rents can and do occur in the preservice period. This suggests that anti-displacement policies need to be adopted and enacted long before transit service begins. It also means that land value capture strategies must also be in place well ahead of transit service operations. Otherwise, increases in land value will have already been captured by land owners. Third, we find that rent increases in anticipation of transit service are more likely for multifamily units with two or more bedrooms. This suggests that families that rent—especially larger low-income families that rent—are vulnerable to displacement pressures even before transit service begins. Finally, we find that rents increase for apartments within a half mile of a proposed station but not for those between a half mile and one mile of a proposed station. This suggests that anti-displacement and land value capture strategies of the two counties and PLCC should focus first on locations with multiple bedrooms within a half mile of proposed stations.

In sum, we believe that our analysis adds nuance to a growing body of literature on the effects on transportation investments on land and housing markets. It helps to confirm principles of urban economics, real estate development, and urban planning on the relationship between accessibility and property values. More importantly, however, we believe our analysis on the specific effects of anticipated investments in light rail on multifamily rents add information that can lead to more carefully designed land value and anti-displacement policies. We do not suggest that our results imply that rents will increase for multiple bedroom units near planned stations in every light rail project, but we do suggest that analysis of the potential of such effects are necessary for anti-displacement policies to succeed.

The American Community Survey (ACS), also administered by the US Census, provides similar information gathered from a different survey at more specific levels of geography. In Montgomery County, Maryland, 38.1% of renter households had resided in their rental unit for over 5 years as of 2019 (US Census Bureau, 2022B). While it is not possible to interpolate the exact number of renters who have lived in their unit for 2 years versus 4 years, it is likely that many renters in the bracket of 52.1% for 2–5 years have lived there for at least three. Adding that to the share that have lived in a unit for over 5 years means it is likely that over half the renter population in Montgomery County has lived in their current unit for at least five years.

While it is clear from the data that the majority of current renters have moved relatively recently, it is also clear that a plurality of renters are unlikely to move. There is an ongoing secular decline in residential mobility driven by housing affordability and demographic change, made clear by the fact that as of 2019, only 1 in 10 Americans had moved in the past year, down from nearly 1 in 5 in 1985 (Frost, 2020). Demographic research indicates that older renters (age 60 and above) are much less likely to expect to move in coming years than younger renters (McHugh, Gober and Reid 1990). Bartik, Butler and Liu (1992) estimated that low-income renter households are willing to pay 10% of income to avoid being forced out of current dwellings due to the psychological costs of moving, which are greater for long-tenure, elderly, and minority households. Montgomery County is subject to the same demographic pressure as the rest of the country, driving down residential mobility rates. As shown, Montgomery County has a high number of long-tenure renter households. Further, Montgomery County’s rental households are majority-minority: 73,061 of 128,212 (57.0%) total rental households indicated a race other than white alone (US Census Bureau, 2022B). Interviews with affordable housing and minority serving officials confirm that minority residents of the Purple Line corridor have very limited housing choices due because of limited employment or rental records, immigration status, and cultural support systems. For these residents the social costs of moving far exceed the pecuniary cost of doing so.

Supplemental Material

Supplemental Material - Do Multifamily unit Rents Increase in Response to Light Rail in the Pre-service Period?

Supplemental Material for Do Multifamily unit Rents Increase in Response to Light Rail in the Pre-service Period? by Qiong Peng, Gerrit Jan Knaap, and Nicholas Finio in International Regional Science Review

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Gerrit‐Jan Knaap

Supplemental Material

Supplemental material for this article is available online.

Notes

References

Agostini

C. A.

Palmucci

G. A.

. 2008. “The anticipated capitalisation effect of a new metro line on housing prices.” Fiscal Studies 29 (2): 233–256.

Al-Mosaind

M. A.

Dueker

K. J.

Strathman

J. G.

. 1993. “Light-rail transit stations and property values: a hedonic price approach.” Transportation Research Record 1: 90–94.

Bae

C.H.C.

Jun

M.J.

Park

. 2003. “The impact of Seoul's subway Line 5 on residential property values.” Transport Policy 10 (2): 85–94.

Bartik

T. J.

Butler

J. S.

Liu

J.-T

. 1992. “Maximum score estimates of the determinants of residential mobility: Implications for the value of residential attachment and neighborhood amenities.”Journal of Urban Economics 32 (2): 233–256. doi:10.1016/0094-1190(92)90007-8

Bollinger

C. R.

Ihlanfeldt

K. R.

Bowes

D. R

. 1998. “Spatial variation in office rents within the Atlanta region.” Urban Studies 35 (7): 1097–1118.

Boucq

Papon

. 2008. “Assessment of the real estate benefits due to accessibility gains brought by a transport project: the impacts of a light rail infrastructure improvement in the Hauts-de-Seine department.” European Transport 40: 51–68.

Cao

Lou

. 2018. “When and How Much Did the Green Line LRT Increase Single-Family Housing Values in St.” Paul, Minnesota? Journal of Planning Education and Research 38: 427–436.

Cervero

. 1994. Rail transit and joint development: Land market impacts in Washington, DC and Atlanta. Journal of the American Planning Association, 60 (1), 83–94.

Cervero

Duncan

. 2002. “Benefits of proximity to rail on housing markets: Experiences in Santa Clara County.” Journal of Public Transportation 5: 1–18

10.

Chatman

D. G

. 2013. Does TOD Need the T? Journal of the American Planning Association, 79 (1): 17–31.

11.

Chen

Rufolo

Dueker

K. J

. 1998. “Measuring the impact of light rail systems on single-family home values: A hedonic approach with geographic information system application.” Transportation Research Record 1617: 38–43.

12.

Clower

T. L.

Weinstein

B. L.

. 2002. The impact of Dallas (Texas) area rapid transit light rail stations on taxable property valuations. Australian Journal of Regional Studies, 8(3): 389.

13.

Damm

Lerman

S. R.

Lerner-Lam

Young

. 1980. “Response of urban real estate values in anticipation of the Washington Metro.” Journal of Transport Economics and Policy 1: 315–336.

14.

Dawkins

Rolf

Moeckel

. 2016. Transit-Induced Gentrification: Who Will Stay, and Who Will Go?, Housing Policy Debate. 26:4‐5 801–818. DOI: 10.1080/10511482.2016.1138986

15.

Deka

2017. Benchmarking gentrification near commuter rail stations in New Jersey. Urban Studies. 54(13) 2955–2972. https://doi.org/10.1177/0042098016664830

16.

Debrezion

Pels

Rietveld

. 2007. “The impact of railway stations on residential and commercial property value: a meta-analysis.” The Journal of Real Estate Finance and Economics 35 (2): 161–180.

17.

Devaux

Dubé

Apparicio

. 2017. Anticipation and post-construction impact of a metro extension on residential values: The case of Laval (Canada), 1995–2013. Journal of Transport Geography, 62: 8–19.

18.

Diao

. 2015. “Selectivity, spatial autocorrelation and the valuation of transit accessibility.” Urban Studies 52: 159–177.

19.

Mulley

. 2007. The short-term land value impacts of urban rail transit: Quantitative evidence from Sunderland, UK. Land Use Policy 24: 223–233.

20.

Dubé

Des Rosiers

Thériault

. (2012). Impacts of commuter rail transit on property values-The Montreal North Shore case (No. eres2012_294). European Real Estate Society (ERES).

21.

Dubé

Legros

Devaux

. 2018. “From bus to tramway: Is there an economic impact of substituting a rapid mass transit system? An empirical investigation accounting for anticipation effect.” Transportation Research Part A: Policy and Practice 110: 73–87.

22.

Dueker

K.J.

Bianco

M.J

. 1999. “Light-rail-transit impacts in Portland: The first ten years.” Transportation Research Record 1685: 171–180.

23.

Duncan

. 2008. “Comparing rail transit capitalization benefits for single-family and condominium units in San Diego, California.” Transportation Research Record 2067: 120–130.

24.

Duncan

2011. The impact of transit-oriented development on housing prices in San Diego, CA. Urban studies, 48 (1): 101–127.

25.

Elrich

2019. Personal communication, phone discussion with Gerrit Knaap. College Park, MD.

26.

Frost

Riordan

. 2020. ““Are Americans Stuck in Place? Declining Residential Mobility in the US.” Joint Center for Housing Studies of Harvard University.” https://www.jchs.harvard.edu/sites/default/files/harvard_jchs_are_americans_stuck_in_place_frost_2020.pdf

27.

Gatzlaff

D. H.

Smith

M. T

. 1993. “The impact of the Miami Metrorail on the value of residences near station locations.” Land Economics 69: 54–66.

28.

Hass-Klau

Crampton

Benjari

. 2004. “Economic impact of light rail-the results of 15 urban areas in France, Germany, UK and North America. Great Britain: Environmental & Transport Planning. Access from.” https://www.researchgate.net/publication/317941738_Economic_Impact_of_Light_Rail_the_Results_for_15_Urban_Areas_in_France_Germany_UK_and_North_America

29.

Henneberry

1998. Transport investment and house prices. Journal of Property Valuation and Investment, 16 (2): 144–158.

30.

Higgins

C. D.

Kanaroglou

P. S

. 2016. “Forty years of modelling rapid transit’s land value uplift in North America: Moving beyond the tip of the iceberg.” Transport Reviews 36: 610–634.

31.

Hurst

Needham B.

West

Sarah E.

. 2014. “Public transit and urban redevelopment: The effect of light rail transit on land use in Minneapolis, Minnesota.” Regional Science and Urban Economics 46: 57–72.

32.

John

Sirmans

G. Stacy

. 1996. “Mass transportation, apartment rent and property values.” Journal of Real Estate Research 12: 1–8.

33.

Knaap

G. J.

Ding

Hopkins

L. D.

. 2001. Do plans matter? The effects of light rail plans on land values in station areas. Journal of Planning Education and Research, 21: 32–39.

34.

Knaap

Gerrit

Talen

Emily

. 2005. “New urbanism and smart growth: A few words from the academy.” International Regional Science Review 28 2: 107–118.

35.

Loomis

Santiago

de Jesus

Y. L

. 2012. “Effects of construction and operation phases on residential property prices of the Caribbean's first modern rail transit system.” Urban Public Economics Review 17: 56–77.

36.

Maryland Transit Administration . 2014. Press Release: Governor O’Malley Announces Purple Line Receives Federal Environmental Approval.

37.

McCartney

Robert

. “2019. JPMorgan Chase commits $5 million to combat gentrification along the Purple Line.” https://www.washingtonpost.com/local/trafficandcommuting/jpmorgan-chase-commits-5-million-to-combat-gentrification-along-the-purple-line/2019/10/21/6a7bd8dc-f42f-11e9-ad8b-85e2aa00b5ce_story.html

38.

McDonald

J. F.

Osuji

C. I

. 1995. “The effect of anticipated transportation improvement on residential land values.” Regional Science and Urban Economics 25: 261–278.

39.

McHugh

Kevin E.

Gober

Patricia

Reid

Neil

. 1990. “Determinants of short- and long-term mobility expectations for home owners and renters.” Demography 27, 1: 81–95.

40.

McMillen

D.P.

McDonald

, 2004. Reaction of house prices to a new rapid transit line: Chicago’s midway line, 1983–1999. Real Estate Economics, 32 (3): 463–486.

41.

Mohammad

S. I.

Graham

D. J.

Melo

P. C.

Anderson

R. J

. 2013. “A meta-analysis of the impact of rail projects on land and property values.” Transportation Research Part A: Policy and Practice 50: 158–170.

42.

Montgomery County . 2018. “Montgomery County Annual Rental Facility Occupancy Survey: Overview of the 2018 DHCA Rental Survey” CountyStat.

43.

Montgomery National Capital Park and Planning Commission . 2017. Rental Housing Study Technical Document: Montgomery County, Maryland.

44.

Nelson

A. C.

Eskic

Hamidi

Petheram

S. J.

Ewing

Liu

J. H

. 2015. “Office rent premiums with respect to light rail transit stations: Case study of Dallas, Texas, with implications for planning of transit-oriented development.” Transportation Research Record 2500: 110–115.

45.

Peng

Knaap

. (2023). When and Where Do Home Values Increase in Response to Planned Light Rail Construction?. Journal of Planning Education and Research. 0739456X221133022.

46.

Peng

Qiong

Knaap

Gerrit J.

. 2021. “Investigating the effects of service and management on multifamily rents: a multilevel linear model approach.” Journal of Housing and the Built Environment 36: 991–1009.

47.

Price

Cheung

Friedman

Singer

. 2005. The world settles in Washington, DC: as an immigrant gateway. Urban Geography, 26: 61–83.

48.

Purple Line Corridor Coalition . 2019. Purple Line Corridor Coalition Housing Action Plan 2019-2022. http://purplelinecorridor.org/wp-content/uploads/2019/12/HAP-Full-Report-06-Dec-2019.pdf

49.

Purple Line Corridor Coalition . 2017. Pathways to Opportunity: A Community Development Agreement for the Purple Line Corridor. http://purplelinecorridor.org/wp-content/uploads/2017/10/PLCommunityDevelopmentAgreement.pdf

50.

Sadayuki

Taisuke

. 2018. “Measuring the spatial effect of multiple sites: An application to housing rent and public transportation in Tokyo, Japan.” Regional Science and Urban Economics 70: 155–173.

51.

Slater

Tom

. 2009. “Missing Marcuse: On gentrification and displacement.” City 13: 2–3. doi:10.1080/13604810902982250.

52.

Turner

Snow

. 2001. “Leading Indicators of Gentrification in D.C. Neighborhoods.” Presentation at the Urban Institute D.C. Policy Forum, Washington, D.C., June 14, 2001. Washington, DC: Urban Institute.

53.

U.S. Census Bureau . 2017. ACS Demographic and Estimates. 2013-2017 American Community Survey 5-Year Estimates. U.S. Census Burea

54.

US Census Bureau . 2022a. American Housing Survey (AHS) American Housing Survey Table Creator. Washington, DC: U.S. Census Burea. https://www.census.gov/data/data-tools/ahs-table-creator.html

55.

US Census Bureau . 2022b. American Communities Survey 5 year Tables (ACS). U.S. Census Burea. https://data.census.gov/cedsci/advanced

56.

Weinstein

B. L.

Clower

T. L

. 1999. The initial economic impacts of the DART LRT system. Denton: University of North Texas. https://digital.library.unt.edu/ark:/67531/metadc30378/

57.

Weinberger

R. R

. 2001. Light rail proximity: Benefit or detriment in the case of Santa Clara County, California? Transportation Research Record, 1747 (1): 104–113.

58.

Welch

T. F.

Gehrke

S. R.

Wang

. 2016. “Long-term impact of network access to bike facilities and public transit stations on housing sales prices in Portland, Oregon.” Journal of Transport Geography 54: 264––272.

59.

Wolverton

M. L.

Hardin

W. G.

Cheng

. 1999. “Disaggregation of local apartment markets by unit type.” The Journal of Real Estate Finance and Economics 19: 15.

60.

Yan

Delmelle

Duncan

. 2012. “The impact of a new light rail system on single-family property values in Charlotte, North Carolina.” Journal of Transport and Land Use 5: 60–67.

61.

Yen

B. T.

Mulley

Shearer

Burke

. 2018. “Announcement, construction or delivery: When does value uplift occur for residential properties? Evidence from the Gold Coast Light Rail system in Australia.” Land Use Policy 73: 412–422.

62.

Zuk

Bierbaum

Chapple

Gorska

Loukaitou-Sideris

. 2018. “Gentrification, Displacement, and the Role of Public Investment.” Journal of Planning Literature 33 (1): 31–44.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.91 MB