# RECORD 721bde276de0a5af885925a9444c628cfce6d18a # TITLE: The impacts of light rail on residential property values in a non-zoning city: A new test on the Houston METRORail transit line # YEAR: 2019.0 VENUE: Journal of Transport and Land Use DOI: 10.5198/jtlu.2019.1310 # AUTHORS: Qi-Sheng Pan # PAGES: 24 CANDIDATE TABLE PAGES (1-based): [3, 14, 18] # ABSTRACT: The impacts of rail transit system on residential property values have been examined for many metropolitan areas in the U.S. But there are few studies on the effects of light rail in a non-zoning city. As the rail transit in the largest non-zoning city, Houston’s light rail transit line, or the so-called METRORail, has not received much attention from the planning research society since it opened to the public in 2004. A previous study by the author utilized 2007 household data to analyze the impacts of Houston’s METRORail line and found the net effects of the rail transit line change significantly at different distances from the rail stations. One limitation of that study was that the physical environment and neighborhood characteristics of the station areas may not have had notable changes over a relatively short time span, i.e., three years after the opening of the light rail. This study employs 2010 InfoUSA household data to re-examine the effects of Houston’s METRORail line. Similar to the previous studies, the author adopts a traditional ordinary linear regression (OLS) to investigate the contribution of a set of variables representing the physical, neighborhood, and accessibility characteristics of properties, and also employs a multi-level regression model (MLR) to examine the hierarchical structures of spatial data explicitly. In addition, this study tests the spatial autocorrelation in the modeling process and analyzes its effects on the results. The modeling results suggest that the METRORail line has had significant net positive effects on residential property values. The MLS model captures the difference of these effects with more spatial details. The spatial regression model improves model fit, but spatial autocorrelation is not completely eliminated. ===== PAGE 1 (score 10, numeric-paren 0, table-words 0, var-words 10) ===== Abstract: The impacts of rail transit system on residential property val- ues have been examined for many metropolitan areas in the U.S. But there are few studies on the effects of light rail in a non-zoning city. As the rail transit in the largest non-zoning city, Houston’s light rail transit line, or the so-called METRORail, has not received much atten- tion from the planning research society since it opened to the public in 2004. A previous study by the author utilized 2007 household data to analyze the impacts of Houston’s METRORail line and found the net effects of the rail transit line change significantly at different distances from the rail stations. One limitation of that study was that the physi- cal environment and neighborhood characteristics of the station areas may not have had notable changes over a relatively short time span, i.e., three years after the opening of the light rail. This study employs 2010 InfoUSA household data to re-examine the effects of Houston’s METRORail line. Similar to the previous studies, the author adopts a traditional ordinary linear regression (OLS) to investigate the contribu- tion of a set of variables representing the physical, neighborhood, and accessibility characteristics of properties, and also employs a multi-level regression model (MLR) to examine the hierarchical structures of spatial data explicitly. In addition, this study tests the spatial autocorrelation in the modeling process and analyzes its effects on the results. The model- ing results suggest that the METRORail line has had significant net positive effects on residential property values. The MLS model captures the difference of these effects with more spatial details. The spatial re- gression model improves model fit, but spatial autocorrelation is not completely eliminated. Keywords: Light rail, residential property values, soning, ordinary lin- ear regression, multi-level regression model, METRORail. The impacts of light rail on residential property values in a non- zoning city: A new test on the Houston METRORail transit line Article history: Received: September 17, 2017 Received in revised form: December 8, 2018 Accepted: January 6, 2019 Available online: April 15, 2019 Copyright 2019 Qisheng Pan http://dx.doi.org/10.5198/jtlu.2019.1310 ISSN: 1938-7849 | Licensed under the Creative Commons Attribution – Noncommercial License 4.0 The Journal of Transport and Land Use is the official journal of the World Society for Transport and Land Use (WSTLUR) and is published and sponsored by the University of Minnesota Center for Transportation Studies. T J  T  L U http://jtlu.org V. 12 N. 1 [2019] pp. 241–264 Qisheng Pan Tongji University Texas Southern University pan_qs@tsu.edu 1 Introduction There are a large number of studies on the impact of rail transit services in many cities However, few of them have been conducted for a non-zoning city. Zoning has been a common method of controlling land use in U.S. cities since the 1920s. It is important for decision makers and planning scholars to ===== PAGE 3 (score 21, numeric-paren 0, table-words 4, var-words 11) ===== 243 The impacts of light rail on residential property values in a non-zoning city Table 1. Mode shares of commuting trips in Houston Metropolitan Statistical Area, 1980-2009 Notes: 1. Large US metro areas have population greater than 3 million in 2000 excluding New York. Source: Census Trans- portation Planning Package (CTPP), 1980-2000 data sets. 2. 2009 data are obtained from the 2009 American Community Survey. Due to the decentralization trend (often simplified via the catch phrase “urban sprawl”), Houston’s population density is low, suggesting that it is not feasible to support a sustainable public transit system to operate in the entire metropolitan area. Transit service in Houston has been mainly limited to the core county, Harris County. Similar to the other comparable cities, Houston has a low and continually de- creasing transit mode share. All the evidence makes it a reasonable expectation that Houston’s METRO- Rail has little, if any, effect on residential property values. However, most discussions on the relationship between density and transit system ignore the variation of density within a metropolitan area (Bertaud, 2003, 2004). Many studies do not provide details on spatial variation of transit mode share in large met- ropolitan areas. It is possible that the light rail in Houston has significant effects on land-use change and residential property values in the light rail corridor due to the relatively high transit mode share in the corridor and its unique land-use policies. Table 1, for example, does not provide sufficient details on spa- tial variation of transit mode share in Houston, especially for commuting trips to Downtown Houston. A commute survey in 2009 showed a surprisingly high public transit mode share in downtown Houston (Table 2). Therefore, it is important to incorporate spatial variation into the analysis. Table 2. Mode share of downtown Houston commuting, 2009 Source: Central Houston Inc. (2009), Downtown Houston Commute Survey Report Houston Large US Metro Areas*1 Mode 1980 1990 2000 2009*2 1980 1990 2000 Drove Alone 69.4 76.1 77 78.8 65.07 72.21 73.71 Carpool 22.5 14.6 14.2 12.1 18.72 13.40 12.64 Transit 2.9 3.7 3.3 2.2 8.54 6.92 6.43 Walk 2.7 2.3 1.6 1.6 4.50 3.39 2.69 Others 2.5 3.3 3.9 5.3 3.16 4.07 4.53 Total Workers (1000) 1,512 1,769 2,082 2,709 27,981 34,744 38,009 Total Population (1000) 3,120 3,731 4,670 5,867 62,101 71,417 82,329 Mode Number of Respondents Mode Share Drive alone 5,977 48.40% Carpool 1,377 11.15% Vanpool 180 1.46% Park & Ride 3,558 28.81% Local or express bus 489 3.96% Light rail 153 1.24% Bus and rail 328 2.66% Walk 129 1.04% Rarely travel to downtown 66 0.53% Motorcycle or scooter 55 0.45% Bicycle 38 0.31% Sum 12,350 100.00% ===== PAGE 4 (score 13, numeric-paren 0, table-words 1, var-words 11) ===== 244 JOURNAL OF TRANSPORT AND LAND USE 12.1 The effects of Houston’s METRORail on residential property values were examined in very few studies. The empirical analyses by Pan and Ma (2009) and Pan (2013) employed both a Hedonic Price Model (HPM) with Ordinary Least Square (OLS) function and a multilevel regression (MLR) model to examine the impacts of Houston’s light rail system. It obtained information on home sales prices and physical characteristics of properties from InfoUSA’s 2007 household database. They reported that the opening of the light rail has had significant net positive effects on residential property values. But immediate proximity to light rail stations and bus stops had significant negative impacts on properties located within a quarter mile of rail stops. One limitation of this study was that the data collected in a relatively short time span, i.e., three years after the opening of the light rail, may not capture the changes of physical environment and neighborhood characteristics of the station areas, which usually take a long time to mature. An interesting comparison between the METRORail in Houston, Texas, with the Metro lines in Shanghai also showed that both rail transit systems had significantly positive effects on residential prop- erty values, but notable variations of their effects were observed at different distance ranges and time spans in Houston and Shanghai (Pan, Pan, Zhang, & Zhong, 2014). The HPM model with the cross-sectional data is the most common methodology in the literature for estimating the implicit value of differences in property characteristics (Debrezion, Pels, & Rietveld, 2007; Dubé, Thériault, & Des Rosiers, 2013). However, the inherent spatial nature of this approach may lead to spatial dependence and heterogeneity. Spatial dependence or spatial autocorrelation results from spatial correlation among observations that are assumed to be independent, while spatial heteroge- neity comes from the spatial correlation of the error terms (Anselin, 1988). As Higgins and Kanaroglou (2016) pointed out, it is important to take into account the spatial dependence and heterogeneity in examining the effects of rail transit since the failure to address this issue may lead to the violation of the statistical independence assumption, heteroskedasticity, biased estimates, and unreliable test statistics, etc. Moran’s I statistics and Lagrange multiplier tests have been usually employed to examine the pres- ence of spatial dependence and heterogeneity. Spatial methods such as spatial lag models and spatial error models have been adopted to control spatial dependence and heterogeneity. Spatial fixed effects have been used to capture spatial dependence and heterogeneity. Geographically Weighted Regression (GWR) has been developed to explore spatial varying relationship and examine heterogeneity between dependent and independent variables over space. The difference-in-differences (DID) estimator pro- vides a means for avoiding the bias due to omitted variables. But all these methods have advantages and drawbacks linked to their uses (Higgins & Kanaroglou, 2016). The objective of this study is to conduct a new test on the Houston METRORail Transit Line to identify impacts on residential property values by controlling spatial dependence and heterogeneity. To compare with the previous studies, the author also adopts a traditional ordinary linear regression (OLS) using recent property transaction data to investigate the contribution of a set of variables representing the physical, neighborhood, and accessibility characteristics of properties, and also employs a multi-level regression model (MLR) to examine the hierarchical structures of spatial data explicitly. In addition, this study tests spatial dependence and heterogeneity in the modeling process using Moran’s I statistics and control the spatial dependence and heterogeneity of the variables using spatial regression models. In subsequent sections, we review the relevant literature in Section 2, describe the methodology in Section 3, present the empirical study in Section 4, and summarize findings and draw conclusions in Section 5. ===== PAGE 14 (score 34, numeric-paren 6, table-words 6, var-words 15) ===== 254 JOURNAL OF TRANSPORT AND LAND USE 12.1 Table 5. Variables used in the models 4.2 Analysis and results This study employs a large set of observations to examine the effects of light rail on property values. The MLR and OLS models are applied with 100 percent observations, which include 384,368 residential properties. To test the spatial autocorrelation in the OLS model, Moran’s I method is applied to test the residuals of the residential property values in the OLS models. Spatial regression models including a spatial lag model and a spatial error model are utilized to examine the spatial dependence of the variables with 5 percent samples from the population. Moran’s I and spatial regression models in GeoDa or R’s spdep package are utilized to examine spatial dependence of spatial variables. The outputs of Moran’s I test are shown in Table 6 and plotted out in Figure 2. Table 6. Spatial autocorrelation diagnostics for the OLS residuals of the residential property values. Moran’s Index 0.4227 (0.0000) Moran’s I statistic standard deviate 18.4282 (0.0000) Lagrange Multiplier (LM) Diagnostics for Spatial Lag (Spatial lag Model as alternative hypothesis) 750.7199 (0.0000) Lagrange Multiplier Diagnostics for Spatial Error (Spatial Error Model as alternative hypothesis) 331.8029 (0.0000) Robust Lagrange Multiplier to Spatial Lag 529.23956 (0.0000) Robust Lagrange Multiplier to Spatial Error 110.3226 (0.0000) Variables Levels Descriptions Source Dependent 1st level Lgval1000 Log(Home Sale Prices ($1000)) InfoUSA Independent Homesize Home Size (square feet) InfoUSA Homeage Home Age (Year) InfoUSA Raillineop Home Sale Before (0) or After (1) METRORail Line Open GIS Map RailQMI Within one-quarter mile of METRO Rail stop GIS Map RailHMI Within one-quarter to one-half mile of METRO Rail stop GIS Map Rail1MI Within one-half to one mile of METRO Rail stop GIS Map Rail2MI Within one to two miles of METRO Rail stop GIS Map Rail3Mi Within two to three miles of METRO Rail stop GIS Map BusStQMI Within one-quarter mile of METRO Bus stop GIS Map Hwyhmi Within one-half mile of Highway intersections GIS Map Hwy1mi Within one half to one mile of Highway intersections GIS Map Hwy2mi Within one to two miles of Highway intersections GIS Map Hwy3mi Within two to three miles of Highway intersections GIS Map Distcbd Distance to CBD GIS Map Distmed Distance to Medical Center GIS Map 2nd Level Medinc Median Income ($1000) CTPP/GIS Map Popdens Population density CTPP/GIS Map JobDens Job density CTPP/GIS Map Pctminor Percentage of minority population CTPP/GIS Map JobAcess Total job accessibility CTPP/GIS Map ===== PAGE 15 (score 8, numeric-paren 0, table-words 3, var-words 2) ===== 255 The impacts of light rail on residential property values in a non-zoning city Figure 2. Moran scatter plot for the standardized logarithm of residential property values ($1000) Table 6 shows the results of testing spatial autocorrelation in the OLS residuals. Moran’s Index is calculated as the slope of the Moran scatter plot between standardized logarithm of residential property values and their spatial lag values (Figure 2). The score of 0.4227 is highly significant, which indicates significant spatial autocorrelation of the residuals in the OLS model. All the other testing results, includ- ing the Lagrange multiplier (LM) diagnostics for spatial lag and spatial error, and the robust Lagrange multiplier to spatial lag and spatial error, point to the rejection of the null hypothesis and indicate the presence of strong spatial autocorrelation in the OLS model. It also indicates the missing of lagged dependent variable and the existence of error dependence. So both spatial lag model and spatial error model need to be applied to examine the spatial dependence of the variables. Table 7 reports the results of MLR and OLS with 100 percent of residential properties and 5 per- cent random samples of properties. It also shows the results of spatial regression models including spatial lag model and spatial error model with all the explanatory variables. ===== PAGE 18 (score 133, numeric-paren 116, table-words 1, var-words 1) ===== 258 JOURNAL OF TRANSPORT AND LAND USE 12.1 Table 7. Comparison of results from MLR, OLS, SLM, and SEM Notes: *p = 0.1, significant at the 0.1 level, ** p = 0.05, significant at the 0.05 level, and *** p = 0.01, significant at the 0.01 level. 100 Percent of Properties 5 Percent Random Samples of Properties Model MLR OLS MLR OLS SLM SEM Fixed eff. Level 1 Intercept 3.2236 *** 3.8603 *** 3.9791 *** 3.9010 *** 2.9145 *** 10.8554 *** homesize 0.2331 *** 0.2877 *** 0.2480 *** 0.2830 *** 0.2474 *** 0.2619 *** homeage -0.0092 *** -0.0097 *** -0.0099 *** -0.0101 *** -0.0086 *** -0.0099 *** Raillineop 0.0350 *** 0.0398 *** 0.0301 *** 0.0375 *** 0.0380 *** 0.0353 *** RailQMI 0.4288 *** 0.3911 *** 0.6012 *** 0.4246 ** 0.3593 * 0.4394 ** RailHMI 0.2462 *** 0.3598 *** 0.5019 *** 0.4610 *** 0.3226 *** 0.4727 *** Rail1MI 0.2381 *** 0.3277 *** 0.3740 *** 0.3400 *** 0.2381 *** 0.3627 *** Rail2MI 0.1791 *** 0.3021 *** 0.2518 *** 0.2095 *** 0.1417 *** 0.2249 *** Rail3Mi 0.0884 *** 0.2121 *** 0.1668 *** 0.2175 *** 0.1502 *** 0.2155 *** BusStQMI -0.0146 *** 0.0945 *** 0.0416 *** 0.1131 *** 0.0859 *** 0.0935 *** Hwyhmi -0.0053 -0.0015 0.0067 -0.0097 0.0056 -0.0144 Hwy1mi -0.0088 0.0034 0.0187 0.0251 0.0310 * 0.0139 Hwy2mi 0.0136 ** 0.0129 *** 0.0293 0.0219 * 0.0223 * 0.0164 Hwy3mi 0.0076 -0.0055 * 0.0085 0.0047 0.0079 -0.0025 Distcbd 0.0075 ** -0.0049 ** -0.0071 -0.0070 *** -0.0087 *** -0.0062 * Distmed -0.0267 *** -0.0192 *** -0.0188 *** -0.0171 *** -0.0112 *** -0.0183 *** Level 2 Medinc 0.0062 *** 0.0034 *** 0.0048 *** 0.0033 *** 0.0021 *** 0.0035 *** Popdens 0.0030 * -0.0071 *** -0.0042 * -0.0064 *** -0.0034 ** -0.0066 *** Jobdens 0.0026 *** 0.0034 *** 0.0026 ** 0.0045 *** 0.0029 *** 0.0038 *** Pctminor -0.5325 *** -0.6479 *** -0.6615 *** -0.6687 *** -0.5012 *** -0.6629 *** Jobacess 0.0004 *** 0.0001 *** 0.0001 0.0001 *** 0.0001 *** 0.0001 *** Random eff. Level 1 Intercept 0.3393 *** 0.3256 *** Level 2 Intercept 0.0597 *** 0.0425 *** ρ (Eq. 6) 0.2420 *** λ (Eq. 8) 0.2049 *** R Square 0.4360 0.4447 0.4691 0.4581 Pseudo R Square 0.4798 0.5063 Log Likelihood -17,236.3 -16,889.8 -17,056.9 AIC 33,923.2 34,514.6 33,823.6 34,155.7 Breusch-Pagan Test 863.4998 *** 914.6319 *** LRTErr 358.8515 *** LRTLag 692.989 *** Obs. of Level 1 384368 384368 19072 19072 19072 19072 Obs. of Level 2 1427 1207 1207 ===== PAGE 19 (score 14, numeric-paren 0, table-words 0, var-words 14) ===== 259 The impacts of light rail on residential property values in a non-zoning city 5 Discussion and conclusions This research contributes to the growing literature on the relationship between light rail and residential property values in three unique respects, (1) it examines the effects of light rail on residential property values in the largest non-zoning city, i.e., Houston; (2) the effects of the Houston’s light rail are examined by utilizing the 2010 residential property database, which allows the effects to be captured after a reason- able time span, i.e., six years after the opening of the new light rail line; (3) it employs spatial regression models together with the widely used ordinary linear regression (OLS) and the sophisticated multi-level regression (MLR) model to investigate the impacts of light rail system on residential property values. In the largest non-zoning city, Houston’s light rail is expected to have little, if any, effect on resi- dential property values due to the decentralization of population, manufacturing, and office firms, low population density, and low transit mode share in the non-zoning city. However, this study shows that the opening of the light rail has significant positive effects on residential property values. It finds that Houston’s METRORail has developed significant positive effects on the values of residential properties near to light rail stations after six-year services, which cannot be identified within a three-year time span in a study by Pan and Ma (2009) and Pan (2013). These results are consistent to the findings of Lee and Sener (2017) that reported a significant increase of commercial development along the original light rail corridor in Houston, about 4 to 10 years after its opening in 2004. Policy makers and planners have considered light rail as a means to induce urban land-use change and increase land values due to the improvement of transit accessibility. However, the marginal increase in transit accessibility alone may not be large enough to change broad land-use patterns and thus raise land uses without government support (Giuliano, 1995, 2004; Cervero & Landis, 1995). The empirical studies in the literature have reported mixed results and a large variation for the effects of rail stations on property value (Debrezion et al., 2007). It is argued that most studies on the impacts of light rail have done too early to find consistent patterns of land development and land value change. The few studies that analyze the effects of light rail over a large time span have shown LRT proximity has considerable influence on the probability of land-use change (Cervero & Landis, 1997). Our study confirms that it needs times for Houston’s METRORail to develop significant positive effects on residential property values. Comparing to a zoning city, the time span for a non-zoning city to show significant change of land-use patterns and residential property values after the opening of light rail may be longer due to the lack of government assistance. The land-use study by Lee and Sener (2017) found that the spike of commercial and industrial land use in Houston’s METRORail corridor between 2008 and 2014 after a small decline between 2005 and 2008. During the same periods, residential land uses had modest gains while the number of vacant and other land parcels decreased. It implies that the increased transit accessibility after the introduction of light rail has attracted businesses and firms to locate around the station areas, which increased the attraction of the neighborhoods and thus generated the value uplift for residential properties in the light rail corridor. A limitation of this research lies in its modeling approaches. A major shortcoming of OLS and MLR is that they have not controlled spatial dependence (or spatial autocorrelation) and heterogeneity though MLR takes into account of the nesting structure of the spatial data. Though OLS and MLR models captured the effects of the light rail on residential property values, their results were considered biased and the capitalization effects of light rail were overestimated. Spatial regression models improve the model fit but they have not completely ruled out spatial autocorrelation. The difference-in-differenc- es (DID) method provides a simple way to avoid the bias in the estimated coefficients that results from the omitted variables. It is also able to treat the problem of spatial autocorrelation among residuals if it is generated by omitted variables. We plan to adopt the DID method using longitudinal or repeat sales data in our future research.