|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 IMPACT OF LIGHT RAIL TRANSIT PROXIMITY ON SINGLE FAMILY HOME VALUE APPRE CIATION IN PORTLAND, OREGON 1998 2007 By STEPHEN ANDERSON A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS IN URBAN AND REGIONAL PLANNING UNIVERSITY OF FLORIDA 20 10
2 20 10 Stephen Anderson
3 To all those who work harder because they have to and to all those who still believe that this is a great country and are devoted to making it better
4 ACKNOWLEDGMENTS I would like to thank my committee chair, Dr. Ruth Steiner, who nurtured and encoura ged me throughout the process. I would also like to acknowledge Professor Keith Yearwood for motivation and discussion. I would also like to thank Professors Latimer, Zwick and Dr. Macedo for their input and knowledge. Of course, I would like to thank Mom and Dad, my sister and my brother. My parents provided a very intellectually nurturing environment that has directly affected me. Special thanks to Mom for the support and Dad for the inspiration.
5 TABLE OF CONTENTS Page ACK NOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 8 LIST OF FIGURES ................................ ................................ ................................ ......................... 9 LIST OF ABBREVIATIONS ................................ ................................ ................................ ........ 10 ABSTRACT ................................ ................................ ................................ ................................ ... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 14 Why Study Light Rail? ................................ ................................ ................................ ........... 14 Light Rail Patronage Enforces a Sustainable City Core ................................ ......................... 16 Land Use and Density + Light Rail Transit (LRT) = Sustainable, Successful Downtowns ................................ ................................ ................................ ......................... 17 Light Rail is Popular ................................ ................................ ................................ ............... 19 Research Question ................................ ................................ ................................ .................. 20 2 BACKGROUND INFORMATION ................................ ................................ ....................... 27 Modes of Transport ................................ ................................ ................................ ................. 27 Heavy Rail ................................ ................................ ................................ .............................. 28 Commuter Rail ................................ ................................ ................................ ........................ 29 Buses and Bus Rapid Transit (BRT) ................................ ................................ ...................... 30 History of Transportation and Land Use ................................ ................................ ................ 30 Sprawl ................................ ................................ ................................ ................................ ..... 32 Transit Taking Us Downtown ................................ ................................ ................................ 34 Home Valuation Tied to Bundle of Factors ................................ ................................ ............ 35 Trends in Home Appreciation ................................ ................................ ................................ 36 ................................ ................................ ......................... 38 Physical Constraints and Economic Concerns ................................ ................................ ....... 39 How Portland Changed ................................ ................................ ................................ ........... 41 Land Use Transportation Air Quality (LUTRAQ): The Stimulus for a Regional Approach ................................ ................................ ................................ ............................. 42 3 LITERATURE REVIEW ................................ ................................ ................................ ....... 45 Theoretical Framework ................................ ................................ ................................ ........... 4 5 What is Walkability? ................................ ................................ ................................ .............. 47 Determining Value ................................ ................................ ................................ .................. 48 Methods to Isolate LRT Impacts ................................ ................................ ............................ 49
6 Previous Literature on LRT and Residential Property ................................ ........................... 51 LRT and Neighborhood Income ................................ ................................ ............................. 54 Implementation of a New Transit Station ................................ ................................ ............... 54 Negative Externalities Affect Property Value ................................ ................................ ........ 56 LRT Effect on Commercial Properties ................................ ................................ ................... 58 Conclusi on of Previous Studies ................................ ................................ .............................. 59 Attracting LRT Ridership ................................ ................................ ................................ ....... 60 Multimodal Integration ................................ ................................ ................................ ........... 60 Intensity of Land Use Around Stations ................................ ................................ .................. 61 Density of Population around Stations ................................ ................................ ................... 61 Parking ................................ ................................ ................................ ................................ .... 62 Levels of Service and Connectivity ................................ ................................ ........................ 62 Area Climate ................................ ................................ ................................ ........................... 64 Auto Dependence ................................ ................................ ................................ .................... 65 Public Perception of Transit ................................ ................................ ................................ ... 65 Conclusion ................................ ................................ ................................ .............................. 66 Criticisms Made About LRT ................................ ................................ ................................ .. 67 Roads, Like Transit, Are Not Free ................................ ................................ .......................... 67 Critics: LRT Does Not Decrease Congestion ................................ ................................ ......... 68 Criticism: LRT does not Reduce Sprawl nor Increase Density ................................ .............. 70 Criticism: LRT Only Redistributes Growth, Not Create It ................................ ..................... 70 Conclusion ................................ ................................ ................................ .............................. 71 4 METHODOLOGY ................................ ................................ ................................ ................. 78 Why Study Portland? ................................ ................................ ................................ .............. 78 Why Examine Single Family Residential (SFR)? ................................ ................................ .. 80 Dollar per Square Foot Justification ................................ ................................ ...................... 81 Study Design ................................ ................................ ................................ ........................... 81 Case Studies/ LRT Scenarios ................................ ................................ ................................ .. 83 Data Elements ................................ ................................ ................................ ......................... 85 Data Exclusions ................................ ................................ ................................ ...................... 86 Case Studies/ LRT Scenarios ................................ ................................ ................................ .. 87 Conclusion ................................ ................................ ................................ .............................. 88 CASE STUDIES/RESULTS ................................ ................................ ................................ ......... 89 Case Studies ................................ ................................ ................................ ............................ 90 Hollywood and 42 nd Street, High Income Area with Existing LRT ................................ ...... 90 East 172 nd Street and Burnside Avenue, Low Income Area with Existing LRT ................... 91 Martin Luther King Boulevard and Killingsworth Street, High Income Area with No LRT ................................ ................................ ................................ ................................ ..... 92 Parkrose/ Sumner, Low Income Area with New LRT ................................ ........................... 93 Polk Avenue and Smith Street, Low Income Area with No LRT ................................ .......... 94 Results ................................ ................................ ................................ ................................ ..... 96
7 5 DISCUSSION ................................ ................................ ................................ ....................... 110 Hollywood/42 nd Street: Many Factors Add to Promising Results ................................ ........ 110 Parkrose/Sumner: New Station Yields Uneven Results ................................ ....................... 111 Polk and Smith: No Transit, Not Much Appreciation ................................ .......................... 113 East 172 nd Street: LRT Access Does Not Automatically Equat e Appreciation ................... 115 MLK Blvd. / Killingsworth Street Proximity to Central Business District (CBD) reaps valuation benefits? ................................ ................................ ................................ ............ 117 6 CONCLUSIONS AND RECOMMENDATIONS FOR FURTHER RESEARCH ............. 121 REFERENCE LIST ................................ ................................ ................................ ..................... 124 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 129
8 LIST OF TABLES Table page 1 1 Rail track miles by mode in the United States, 2002 2006 ................................ ................ 23 1 2 Light rail routes under construction and traffic congestion in select American cities as of 2006. 23 1 3 Light rail transit agencies ranked by unlinked passenger trips, fiscal year 2006 (thousands) ................................ ................................ ................................ ......................... 24 1 4 Light rail transit agencies ranked by passenger miles, fiscal year 2006 (thousands) ........ 25 2 1 Cost and other selected characteristics for selected transit modes ................................ .... 44 3 1 Composite of studies: Effects of light rail proximity to property values ........................... 73 3 2 Transit (f ixed route) service frequency level of service measurements ............................ 74 3 3 Transit (fixed route) hours of service level of service measurem ents ............................... 74 3 4 Operating cost per mile by transit mode and cost increase of selected transit modes in the United States, 2000 2007. ................................ ................................ ............................ 74 3 5 Rail route mileage and status of selected future rail projects in United States, 2006 ........ 76 3 6 Land value appreciation by land type near light rail s tops in Dallas, Texas, 1997 200 1 ................................ ................................ ................................ .......................... 76 5 1 Characteristics for Portland light rail transit lines, 2008 ................................ ................... 99 5 2 Case study transit stops and levels of service ................................ ................................ .... 99 5 3 Light rail transit scenario/case study area characteristics ................................ ................ 100 5 4 Dollar per square foot comparison, 1997 and 2007 per findings ................................ .... 108 5 5 SFR appreciation percentage rate per square foot in Portland, Oregon, per findings ..... 109
9 LIST OF FIGURES Figure page 1 1 Relationship between land use and transit patronage ................................ ........................ 26 2 1 Percentage of transit trips in United States in 2000. ................................ .......................... 44 3 1 Percentage increase in mean sales price of existing sing le family residential units in United States by region, 1997 2007. ................................ ................................ .................. 75 3 2 U.S Median Home Prices 1975 2009, (adjusted for inflation) ................................ .......... 77 5 1 Regional map of Portland, Oregon ................................ ................................ .................. 101 5 2 Parcel Map of Smith Street and Polk Avenue ................................ ................................ 102 5 3 Parcel Map of Killingsworth Street /MLK Boulevard. ................................ .................... 103 5 4 Parcel Map of Parkrose/Sumner Area ................................ ................................ .............. 104 5 5 Parcel Map of Hollywood/42 nd Street ................................ ................................ .............. 105 5 6 Parcel Map of East 172 nd Ave ................................ ................................ .......................... 106 5 7 Results, Single family residential parcels value in dollars per square foot, per l ight rail transit scenario in Portland, Oregon, 1997 2007 ................................ ....................... 107 5 8 Single family residential appreciation percentage rate per square foot in Portland, Oregon per findings ................................ ................................ ................................ ......... 108
10 LIST OF ABBREVIATIONS APTA American Public Transportation Association BART Bay Area Rapid Transit BRT Bus Rapid Transit CBD Central Business District DART Dallas Area Rapid Trans it EMV Estimated Market Value FHA Federal Housing Administration GIS Geographic Information Systems HOV High Occupancy Vehicle LOS Level of Service LRT Light Rail Transit LUTRAQ Land Use Transportation Air Quality MARTA Metropolitan Atlanta Regional Transit Authority MAX Metropolitan Area Express (Portland) MBTA Metropolitan Boston Transit Authority MPO Metropolitan Planning Organization MSA Metropolitan Statistical Area ODOT Oregon Department of Transportation RLIS Regional Land Inform ation System RMV Real Market Value SFR Single Family Residential SR State Road
11 TCRP Transit Cooperative Research Program TOD Transit Oriented Development UGB Urban Growth Boundary VMT Vehicle Miles Traveled
12 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements For the Degree of Master of Arts in Urban Regional Planning IMPACT OF LIGHT RAIL TRANSIT PROXIMITY ON SINGLE FAMILY HOME VALUE APPRECIATION IN PORTLAND, OREGON 1998 2007 By Stephen Anderson December 20 10 Chair: Ruth Steiner Co chair: Joseli Macedo Major: Urban and Regional Planning The mobility and accessibility offered by living in walkable proximity to transit cannot be understated. Enter, Portland, Oregon, the city that pioneered its Urban Growth Boundary (UGB) highest per capita light rail uses in t he United States (American Public Transportation Association, 2008) Currently three lines (soon to be four) and nearly forty stations serve this metro area (numbering two million people) connecting jobs, entertainment, and the necessities of life with residents. Numerous studies examine the impact transit access has on property values as far back as the 1960s (Hess & Almeida, 2006) A large body of work exists on how just the preparation of planning a transit line can affect property values. This thesis examines the single family home real estate appreciation rates around five light rail stations throughout the region. The proximity to LRT should reflect in higher valuation of the land. The five sto ps are also categorized as low or high income areas as well as having new, existing or no light rail access. This thesis examines via parcel data home value appreciation rate through 1997 2007 and compare s the appreciation rates between single family resi dential units in areas that are within a quarter mile of a rail
13 station versus those who are not. It also examines appreciation rates in areas that have different light rail transit (LRT) scenarios (those areas with existing transit, those with new transit improvements, and those without transit). The benefits of having transit within a five minute walk should be illustrated in higher single family residential units (SFR) values near those This study differs from previous work in that the actual appreciation will be compared in a dollar per square foot format. Also departing from past studies is a comparison of home value appreciation rat es in area with or without walkable transit access. Wh at happens to home appreciation in single family home areas with existing LRT versus areas that do not have light rail access? The empirical study ultimately contradicts the previous studies that areas with LRT access tend to be worth more, appreciate at a higher rat e in Portland and vicinity than those that do not. This research shows that proximity to the Central Business District has a stronger correlation to higher residential property values than light rail proximity. In the case studies it was found that a high income area with no light ra il access appreciated the most. A low income area with no light rail access appreciated faster than a low income area with access over the same ten year period. Additional research c ould include examine areas that are an identical distance from downtown to determine the affect of proximity
14 CHA PTER 1 INTRODUCTION Light rail, busways, and streetcar lines, like historic boulevards and main streets, are pedestrian friendly and serve as catalysts to the formation of m ixed use neighborhoods and centers. They do not form barriers within communities; in fact, they often unify a place by creating a focus and a common destination. Additionally, they form benign connections between communities; visitors without a car are al ways w elcome to places trying to over come the tyranny of parkin g lots and traffic congestion. -Peter Calthorpe The Regional City To quote Alex Marshall, author of the book, Cities and places are created by the methods we choose for moving people and goods and services from place to systems these systems drive the economy of a city and determin e (Marsh all, 2001) Marshall, while not a professional planner, decries the growth of suburban sprawl that (Marshall, 2001) He discusses how the loss of place has changed social interactions, i.e., quality of life in his hometown of Norfolk, attributed to the advent of the automobile, but more so the adven t of the highway. To undo the impact of the highway and the spr awl that it offers, returning to utilizing other forms of getting from point A to point B may be needed (Marshall, 2001) Light rail may be part of the solution to many of the problems of loss of place. Walking, bicycling and/or using light rail transit are all ways to conserve fossil fuel and lead to more sustainable future (Giuliano & Hanson, 2004) Why S tudy Light Rail? Why not? LRT is a popular form of transit that could lead to more sustai nable, equitable citie s (Giuliano & Hanson, 2004) LRT is a key motor for economic revitalization because it can reinforce the density of CBDs and provide transportation equity to citizens (Giuliano & Ha nson,
15 2004) From conserving fossil fuels to promoting downtown businesses, there are many reasons to study LRT. A healthy LRT system can connect more people to jobs (Calthorpe & Fulton, 2001) According to Figure 1 1 heavy patronage of LRT leads to more mixed uses of land, larger residential densities, as well as more options of transport that do not use automobiles (Giuliano & Hanson, 2004) These developments tend to increase more LRT us e (Figure 1 1). Public transit has an appeal to those who are concerned with transportation equity (Giuliano & Hanson, 2004) Transportation equity is the availability of transport to those who do not own cars or for whateve r reason cannot drive them (Calthorpe & Fulton, 2001) Older citizens and young people who lack autos longer in the C entral B usiness D istricts ; they are often on the periphery (Calthorpe & Fulton, 2001) Many decades of decentralization has left the central city with a decreasing share of economic activity, but an increasing share of disadvantaged population (Pucher & Renne, 2003) are looking to light rail to solve these problems. Table 1 2 highlights the congestion ran kings of cities and their corresponding new light rail track construction. Table 1 2 also illustrates that cities are utilizing light rail to provide another form of transportation. Notice the difference between the miles of track constructed between the two modes. Much of the new construction is located in the Sunbelt, which is one of the, if not the fastest growing areas in the nation. The Sunbelt is defined as the region the south of the 37th or 38th parallels, north latitude (American Public Transportation Association, 2008) These cities are mid size to large; the densities may not yet support commuter or heavy rail, but are likely to be as congested as any other large metro area. All have some form of rail transit; highlighted
16 areas are constructing new LRT tracks as of 2008 (Table 1 2). What is significant is that rapidly growing cities are looking more and more to LRT to fulfill their transit needs. L ight Rail Patronage E nforces a S ustainable C ity C ore Consider Alex Marshall, author of How Cities Work : To have a cohesive downtown, or even reall y older neighborhoods, a city has to have a cohesive system of mass trans it, and has to make it dominant; u rbanism is a space. Only mass transit has the ability to rais e the pressure to enough people per square inch (Marshall, 2001) It is important for the country to rebuild and revitalize its core cities to undo the sprawling patterns of the last fifty years. The health of our CBDs depends on multimodal transit and light rail is an increasingly popular mode of transit (Calthorpe & Fulton, 2001) The design of a city and its transportation o ptions are intrinsically linked (Calthorpe & F ulton, 2001) Strong urban cores are reinforced because of their density (Marshall, 2001) The increased economic value derived from increased accessibility should lead to higher property values in our core areas, which can lead to revitalization of older suburbs (Marshall, 2001) A n ideal scenario that illustrates th is idea for city X : A LRT connects the underutilized d owntown to the s prawling suburbs utilizing old railroad infrastructure As auto commute costs increase, transportation savings increase for employers and employees living near an LRT station. The metro area experiences a collective push to move back downtown. The downt own revitalize s the desire t o be closer to the CBD increases prices in the CBD. As jobs return to the CBD, and those closer to downtown older suburbs become attractive again, this will increas e the municipal tax base as rbon blueprint by reinforcing density If light rail use comprises more and more of our daily commute, we will be using less fossil fuels as LRT runs overwhelmingly on electricity (Marshall, 2001) The decreased dependency on fossil fuels makes LRT an attractive option for transit/ transportation planners.
17 Because most transit systems have the greatest impact on accessibility to downtown, compared to other areas of the region, the greatest impacts on development have been seen in downtown areas: within downtowns, rail transit investments have stimulated redevelopment and T ransit C ooperative R esearch P rogram (Transportation Research Board, 2003) Vesalli (1996) found that public sector involvement, including land assembly, high density zoning allowances, restrictions on parking, and financial incentives, played an important role in most successful examples o f develo pment around transit stations (Handy, 2005) There are many factors that contribute to successful, sustainable downtowns. Land Use and Density + L ight R ail T ransit (LRT) = Sustainable Successful Downtowns A sustainable downtown can mean many things. For Alex Marshall it is a place to buy hardware or underwear without getting in a car (Marshall, 2001) By emphasizing LRT investment, a center cit y can rely less on fossil fuels and reduce its p ollution from automobiles. However, a sustainable city core can also refer to a tight network of narrow walkable streets, an efficient LRT system that gets people where they need to go. Suc cessful LRT requires a successful downtown. The hub of an LRT is A successful downtown i s achieved through many factors -perhaps the most important is a density of mixed uses. Mixed land uses can yield a number of transportation benefits: The more complementary land uses are mixed, (i.e., offices, shops, restaurants, banks ) the more likely people are to walk and less likely to drive. With more mixed land uses, t rips are potent ially more spread out throughout the day and week instead of clumped during the m orning and evening peak periods (Halifax Regional, 2002) Several studies have concluded that land use mix is not as significant in influencing tr ansit mode choice decisions in neighborhoods as density, although it is difficult to sort out the eff ects of land use mix because it is so strongly correlated to density
18 (Halifax Regional, 2002) Land use mix, however, is partic ularly significant for walking and biking modes, which are primary methods of accessing transit service. Because pedestrian and bike trips are typically shorter than transit or auto t r ips, the presence of a land use mix within work to home trip s is essenti al in ensuring accessibility to shopping, restaurants, recreation and les s glamorous places like a hardware store. To access shopping, recreation and restaurants, people need to have the ability to safely walk there once they leave their car, bus, or train they are referring to the ease of access to a particular destination by those on foot. Walkability is transportation infrastructure meant to move car s eff iciently ( e.g., raised highways, triple left turns at intersections) could make wal king three blocks an adventure. Many obstacles can prevent direct access on foot these obstacles may not be readily apparent on a road map. An LRT system can add to what is considered Perhaps the most basic factor in determini ng walkability of an area is the presence of sidewalks. It may be as simple as a wealth of pedestrian amenities including side walks combined with low road traffic speeds. If walkable access to LRT is desirable to the point the consumer is willing to pay more -and the amenities easily accessed through such transit -than it is logical that the same property is attractive for a hi gh er level of development (Diaz, 2001) Such trends often change the land uses naturally around stations. An existing low density use such as S ingle F amily R esidential (SFR) can be converted to a higher density use or another type altogether (Diaz, 2001) These developments are indicative of the rise of Transit Orien ted (TOD) and Mixed Use Developments based on the idea that many employers and employees will wish to capitalize on the transportatio n savings afforded by the land.
19 Light Rail is Popular As an impetus for an increasingly popular urban form TODs -LRT has been also increasing in popularity -if o ne considers new track mileage -it could be considered the most popular form of transit currently. Table 1 1 illustrates the popularity of LRT in this country by showing LRT construction as the largest percentage of total rail construction between 2002 and 2006. Table 1 1 shows that as a percentage LRT is growing faster than other forms of rail transit Over the last few years more and more cities that have relied on ot her modes are also constructing or expanding their LRT systems The perception is that LRT is safer, cleaner, and has a relatively low impact on the environment compared to heavier forms of rail (Transportation Research Board, 2002) Table 1 2 illustrates the cities that are constructing n ew LRT lines or ar e planning to; n otice the cities in the table that are in bold type; these cities are located in the Sunbelt which is normally considered the region of the U.S located south of the 37 th or 38 th parallel north latitude (INRIX, 200 6) Table 1 2 also highlights the INRIX congestion rating for these cities. These Sunbelt cities are growing and traffic congestion is a growing problem. The aim for many transit planners is to relieve these same congested highways. It is argued that rail transit has more potential for attracting discretionary riders out of their cars than bus transit (Giuliano & Hanson, 2004) There are over thirty cities in the country that now employ some form of LRT. Many are experiencing increased ridership, and more sprawling cities such as Phoenix, Dallas, and Charlotte are planning to implement an LRT system or are enhancing an existing one. This is important because these areas are not part of the connec ted compac t Northeast Corridor. T hey are Sunbelt cities, and are rapidly growing. As the next section will discuss the research question, it is important to remember that LRT is arguably the
20 most popular form of rail transit and will continue to be a central figur e in transportation plans in this country for the foreseeable future. Research Question Now that the importance of examining LRT effects have began to be discussed, another theme of this study is the nature of land value. Unlike infrastructure, oil, or pre cious metals, land is a commodity that cannot be transported. It only has value because of where it is. It has been maintained that land value is determined by a bundle of factors, one of which is location (Alonso, Location and Land Use: Toward a General Theory of Land Rent, 1964) Early cities sprouted along transportation routes such as a road, a river or port. The location of the area made it more valuable due to the access the location provided (Alonso, Location and Land Use: Toward a General Theory of Land Rent, 1964) In the beginning of the 20 th century, our fledgling cities experienced an intensification of residential densities becau se of technological limitations since there were no automobiles, a city could only grow as quickly as it took to travel by foot, or horse (Marshall, 2001) Eventually technology improved, one form of transit begat another, and employees were able to live fur ther and further away from jobs (Marshall, 2001) Our current situation is a result of technological development, and perhaps a lack of thoughtful planning for those who lack cars or seek an alternative. We are now in a transi tion, due to many developments, where having access to a transit station is an option over the automobile congestion plaguing our cities For the sake of this study, if this theory holds true, then residential areas near LRT could be experiencing larger p roperty appreciation rates than comparable areas without transit. Or at the very least, they would significantly have a larger dollar per square foot value as a whole than neighborhoods without LRT access. This leads to the question of what effect a new LRT station has on nearby residential areas that previously lacked LRT access? It would follow that areas that depend more
21 on transit (i.e., low income and/or zero car households) could see the largest valuation compared to transit dependent areas that do not have LRT access. It also would follow that high income areas with LRT access may receive an extra valuation than those same high income areas that lack LRT. These are relationships that this study will examine using Portland, Oregon and its LRT syst em, Metro Area Express (MAX), as a case study. This study will also conduct separate scenarios that seek to isolate different LRT access levels to answer the following questions: What effect does LRT access have in a high income area versus a low income area with existing LRT transit? What effect does new LRT access have on a low income area? How do these results compare with high and low income areas that have no LRT access? These scenarios will be quantified in case study format and compared to each other. The independent variable will be 1997 2007 single family residential values taxlot data provided by regional tax collectors. This study will hopefully illustrate a significant valuation benefit to neighborhoods th at have LRT access. As indicated within Table 1 2, some of our most congested c ities are try to relieve that congestion through LRT expansion (Debrezion, Pels, & Rietveld, 2004) Table 1 2 identifies cities that have ( construct ed/are constructing ) LRT and their congestion rankings. Rapidly growing metropolises have turned to LRT that may have lacked rail transit on any level before. LRT can help reinforce a city core and counteract sprawl (Calthorpe & Fulton, 2001) LRT is popular because it is more sustainable and less in trusive than the other forms of rail: heavy rail, commuter rail, rapid rail tend to be noisier and more expensive to construct that LRT (American Public Transportation Association, 2008) This is important because newer Sunbelt cities may lack the older rail infrastructure of older, northern cities (Marshall, 2001) To implement rail in these places m ay require new track, as older areas can
22 simply run new service on existing track. Examining the figures in Table 1 1, while all rail modes have had increases in the ir operating cost per passenger mile, LRT has increase d these costs at a slightly lower rate perhaps pointing to a more sustainable form of trans it for large to mid size cities (Giuliano & Hanson, 2004) It is possible LRT may become cheaper over time, especially compared to rising fuel costs. As to this chapter will identify background information that is necessary to understand the context of the study. Could this desirability translate to higher property values around LRT stations? This th single family residential (SFR) parcels over a ten population is just under two million inhabitants -yet it leads the country in LRT passenger trips per capita (American Public Transportation Association, 2008) Although cities like Boston and San Francisco have larger total yearly ridership, the d ata shown in Table s 1 3 and 1 4 indicate that Portland is a leader is both passenger miles and passenger trips (Federal Transit Administration, 2002) they rel y on their light rail more than any other major metro area in the United State s. T he facts is the 23 rd largest Metropolitan Statistical Area (MSA) (U.S Census Bureau, 2008) As s hown in Tables 1 3 and 1 4, t he area r anks fourth in passenger miles and fourth in passenger trips. In summary, LRT is a popular form of transit for many reasons. Transportation savings make a real difference and impetus in housing choices to many househ olds in the country. Portland is a city that has embraced LRT like no major American metro yet.
23 T able 1 1. Rail track miles by m ode in the United States, 2002 2006 Year Commuter Rail Heavy Rail Light Rail Other Rail Total 2002 7,267.1 2,179.2 1,113.6 29.7 10,589.5 2003 7,433.9 2,209.5 1,147.2 30.0 10,820.6 2004 7,284.1 2,209.5 1,321.2 30.3 10,845.1 2005 7,947.5 2,277.3 1,385.1 30.3 11,640.2 2006 8,016.7 2,277.3 1,463.8 38.3 11,796.1 Percent c hange 2002 2006 10.3% 4.5% 31.4% 29.0% 10.2% Source: American Pub lic Transportation Association. (2008). 2008 Fact Book Table 1 2. Light r ail routes under c onstruction and traffic c ongestion in select American c ities as of 2006. Location Miles of T rack Congestion Ranking Charlotte, NC 9.6 27 Dallas, TX 16.0 4 Denver, CO 19.2 16 Los Angeles, CA 15.1 1 Oceanside, CA 23.7 15 Phoenix, AZ 64.3 13 Portland, OR 0.6 23 Salt Lake City, UT 0.8 50 San Francisco, CA 5.4 7 Seattle, WA 15.5 9 Tempe, AZ 20.0 13 TOTAL 190.2 Source: American Public Transportation Association. (2008). 2008 Fact Book Source: INRIX National Traffic Scorecard 2006 (2006) Oceanside (suburb of San Diego, CA) and Tempe (suburb of Phoenix, AZ) are considered part of the same city for their respective con gestion rankings
24 T able 1 3 L ight rail transit agencies ranked by unlinked passenger t rips, fiscal y ear 2006 ( t housands) Transit Agency/ Urbanized Area (Primary City) Passenger Trips Massachusetts Bay Transportati on Authority (MBTA) Boston, MA 80,278 San Francisco Municipal Railway (MUNI) San Francisco, CA 43,678.8 Los Angeles County Metropolitan Transportation Authority (LACMTA) Los Angeles, CA 42,020.8 Tri County Metropolitan Transportation District of Oregon (TriMet ) Portland, OR 34,591.5 San Diego Trolley, Inc. (MTS) San Diego, CA 33,829.8 Southeastern Pennsylvania Transportation Authority (SEPTA) Philadelphia, PA 25,445.5 Dallas Area Rapid Transit (DART) Dallas, TX 18,581.1 Bi State Development Agency (METRO) St. Louis, MO 16,573.9 New Jersey Transit Corporation (NJ TRANSIT) New York, NY 15,767.1 Utah Transit Authority (UTA) Salt Lake City, UT 15,203.7 Sacramento Regional Transit District (Sacramento RT) Sacramento, CA 14,452.1 Metropolitan Transit Authority of Harris County, Texas (Metro) Houston, TX 11,333.1 Denver Regional Transportation Dis trict (RTD) Denver, CO 11,277.9 Metro Transit Minneapolis, MN 8,957.9 Santa Clara Valley Transportation Authority (VTA) San Jose, CA 8,279.8 Port Authority of Allegheny County (Port Authority) Pittsburgh, PA 7,510.6 Maryland Transit Administration (MTA) Baltimore, MD 5,927.1 Niagara Frontier Transportation Authority (NF T Metro) Buffalo, NY 5,631.9 The Greater Cleveland Regional Transit Authority (GCRTA) Cleveland, OH 3,791.3 Memphis Area Transit Authority (MATA) Memphis, TN 959.3 Central Puget Sound Regional Transit Authority (ST) Seattle, WA 885.6 New Orleans Regional Transit Authority (NORTA) New Orleans, LA 605.9 Hillsborough Area Regional Transit Authority (HART) Tampa, FL 520.3 Charlotte Area Transit System (CATS) Charlotte, NC 175.3 Central Arkansas Transit Authority (CATA) Little Rock, AR 154.4 Kenosha Transit (KT) Kenosha, WI 52.9 Island Transit (IT) Galveston, TX 37 Source: Feder al Transit Administration. (2007 ) National Transit Database. Note: Data e xcludes transit agencies not reporting data to the N ational T ransit D atabase.
25 T able 1 4 L ight rail transit agencies ranked by p assenger m iles fiscal year 2006 ( t housands) Transit Agency/ Urbanized Area (Primary City) Passenger Miles Los Angeles County Metropolitan Transportation Authority (LACMTA) Los Angeles, CA 302,183.5 Massachusetts Bay Transportation Authority (MBTA) Boston, MA 215,711 San Diego Trolley, Inc. (MTS) San Diego, CA 208,875.5 Tri County Metropolitan Transportation District of Oregon (TriMet) Portland, OR 179,875. 4 Dallas Area Rapid Transit (DART) Dallas, TX 136,797.1 Bi State Development Agency (METRO) St. Louis, MO 119,769.5 San Francisco Municipal Railway (MUNI) San Francisco, CA 112,916.6 Utah Transit Authority (UTA) Salt Lake City, UT 86,039 Sacramento Regional Transit District (Sacramento RT) Sacramento, CA 78,181 New Jersey Transit Corporation (NJ TRANSIT) New York, NY 72,899.5 Southeastern Pennsylvania Transportation Authority (SEPTA) Philadelphia, PA 64,430.8 Denver Regional Transportation Distr ict (RTD) Denver, CO 59,137.1 Metro Transit Minneapolis, MN 52,584.6 Santa Clara Valley Transportation Authority (VTA) San Jose, CA 41,913.3 Maryland Transit Administration (MTA) Baltimore, MD 33,831.9 Port Authority of Allegheny County (Port Authority) Pittsburgh, PA 32,902.8 Metropolitan Transit Authority of Harris County, Texas (Metro) Houston, TX 27,517.6 The Greater Cleveland Regional Transit Authority (GCRTA) Cleveland, OH 22,147.1 Niagara F rontier Transportation Authority (NFT Metro) Buffalo, NY 13,867.3 Central Puget Sound Regional Transit Authority (ST) Seattle, WA 983 Memphis Area Transit Authority (MATA) Memphis, TN 919.6 Hillsborough Area Regional Transit Authority (HART) Tampa, FL 838.4 New Orleans Regional Transit Authority (NORTA) New Orleans, LA 763 Charlotte Area Transit System (CATS) Charlotte, NC 279.5 Central Arkansas Transit Authority (CATA) Little Rock, AR 249 Kenosha Transit (KT) Kenosha, WI 59.5 Island Transit (I T) Galveston, TX 47.6 Source: Feder al Transit Administration. (2007 ) National Transit Database. Note: Data excludes transit agencies not reporting data to the National Transit Database.
26 Figure 1 1 Relationship between land use and transit patronage Source: Adapted from Halifax Regional Planning Paper, 2002 Transit Patronage Residential Density Pedestrian Amenities/ "Walkability" Mix of Uses Location of Uses Site Design Size and Density of Centers
27 CHAPTER 2 BACKGROUND INFORMATI ON Modes of Transport Before delving into any further research or literature on L RT, it is necessary to define what constitutes LRT The A merican P ublic T ransportation A ssociation Fact Book ( 2008 ) defines light rail (streetcar, tramway, or trol ley) as lightweight passenger rail cars operating singly (or in short, usually two car, trains) on fixe d rails in a right of way that is not separated from other traffic for much of the way. Light rail vehicles are typically driven electrically with power being drawn from an overhead electric line via a trolley or a pantograph. A light rail car (or street car, tram, or trolley car) usually operates much of its route on non exclusive right of way. Sometimes older cars are refurbished (vintage trolley cars) or newer cars are built to look like older cars (heritage trolley cars) (Am erican Public Transportation Association, 2008) Case to make their developments look more appealing (Brown & Weiner, 2009) For so me, LRT is aesthetically appealing and characterized by lower speeds and lower capacity tha (American Public Transportation Association, 2008) Newer systems tend to operate on private rights of way separat e from city automobile traffic (American Public Transportation Association, 2008) Light rail may be defined to somewhat erroneously to include older streetcar systems (trolleys or tramways) that often run on city streets as well as modern light rail systems. Light rail systems can typically handle steeper inclines than heavy rail, and curves sharp enough to fit wi thin street intersections LRT is generally powered by electricity, usually by means of overhe ad wires but sometimes by a live rail also called third rail (a high voltage bar alongside the track), requiring safety measures and warnings to the public not to touch it. The type of technology used for the modern light rail
28 systems of today (the stand ard gauge, the electric wires over head connecting to the car, always over ground) are typically what people associate with LRT. Some examples of LRT systems What i s not LRT? Streetcars and trolleys that are not used over long distances, do not operate a t fast speeds nor have a large number of riders compared to the other modes of rail transit are not LRT The difference between streetcar/tram systems and LRT are di fficult to ascertain because of a technology overlap Some systems referred to as light rail are actually streetcar systems and vice versa. traditional trolley where the tracks and trains run along the streets and share spac e with road traffic. Stops tend to be frequent, and little effort is m ade to set up special stations. A true, often separated from ro ad traffic and run along their own right of way Tracks are highly visible, and in some cases significant effort is expended to keep traffic away through the use of special signaling, and even grade crossings with gate arms. Many light rail systems eve n fairly old ones have a combination of the two, with both on road and off road sections. In some countries, only the latter is described as light rail (American Public Transportation Association, 2008) In those places, trams running on mixed right of way are not regarded as light rail, but considered distinctly as streetcars or trams. However, the requirement for saying that a rail line is "separated" can be quite minimal sometimes just with concrete buttons to dis courage automobile drivers from getting onto the tracks. Heavy Rail The subways that were utilized early in the 20 th century are the predecessors to LRT These systems were effective partly because of lack of space and the population density unique to N other examples
29 These systems tend to travel at higher speeds and have lar ger carrying capacity t han LRT. These systems differ from commuter rail in that they offer less service for commuters from suburbs and more on transport in tra city, instead of intercity. The American Public Transportation Fact Book 2008 defines heavy rail (metros, subways rap id transit or rapid rail) as A n electric railway with the capacity for a heavy volume of traffic. It is characterized by high speed and rapid acceleration passenger rail cars operating singly or in multi car trains on fixed rails; separate rights of way from which all other vehicular and foot traffic are excluded; sophisticated signaling, and h igh platform loading. If the service were converted to full automation with no onboard personnel, the service would be considered an automated guide way. A heavy rail car has motive capability, is driven by electric power taken from a third rail or (rarely, overhead wires), and is usually operated on exclusive right of way (American Public Transportation Association, 2008) Commuter Ra il Northeast Corridor trains are an example of commuter rail in the United States. T his train carries those who wish to travel intercity through the Northeast chain of cities. The American Public Transportation Fact Book 2008 defines commuter rail as A n electric or diesel propelled railway for urban passenger train service consisting of local short distance operating between a central city and adjacent suburbs. Service must be operated on a regular basis by or under contract with a transit operator for the purpose of transporting passengers within urbanized areas and outlying areas (American Public Transportation Association, 2008) An example of c Rail, a rail line that serves the metro areas of Miami Fort Lauderdale and West Palm Beach. As with most rail transit, this system emphasizes quicker headways (time between running trains) during peak periods (generally during AM and PM commuter rush hours) (American Public Transportation Association, 2008)
30 Buses and B us R apid T ransit (BRT) The large majority of Americans take their transit trips on bus, not rail. As Figure 2 1 identifies, bus transit will always have its purpose. Being that it is not a fixed way route gives it some advantages over LRT. B us R apid T ransit (BRT) has been util ized around the world as a flexible, less expensive alternative to LRT. BRT buses may have lower floors and can be very similar to LRT in appearance and function. Longer distance between stops, higher operating speeds and longer routes are all characteri stics of BRT -which can make it appealing as a LRT alternative (Jacobs Engineering, 2008) Table 2 1 identifies the capital cost per vehicle in having a dedicated lane for BRT, which is slightly lower than The seate d capacity of BRT is slightly lower. Even if the costs are lower -c n competition for new expansion -to stimulate property values. The many modes of tra nsport that used fixed guideway systems (LRT, heavy and commuter rail) offer benefits that buses and BRT may not. While buses will always serve a major need, there is a desire by growing metros to establish passenger rail transit where there was none previ ously. BRT tends to offer a cheaper alternative to these cities as well. The variety of modes o f transport all rely on the same things (people, energy source) and all seek to produce the same result to increase accessibility and mobility. In our larger m etro areas these modes tend to work together to achieve the intended results. As mentioned before there is some competition in areas between LRT and BRT because of the similar density requirements and the differing capital costs. History of Transportatio n and Land Use The following are necessary to understanding the history of transportation in America, particularly as it concerns urban and later suburban dwellers. The suburb, while not an American invention, was aided by the American expressway in the wa y the resulting land uses made land
31 value no longer tied to distance to the workplace. As our transit modes evolved, the need to live close to mobility was redefined. The home/ land values that accompanied access to mobility -be it streetcar, light rail or bus have influenced the form and fashion of our cities. Before any mechanized transit, water bodies were early transit lines; their accessibility made their location desirable, especially in this country. The first large American towns were located on major bodies of water. As technology advanced moving inland became possible, and cheaper. Towns grew into cities and since the form of transport was predominantly by foot then the challenge w as to keep everything close. However for many early city dweller s needed to live close to where they worked, or access to transport Thus l i ving near a transit stop was hardly a new concept. When the streetcar was invented in the lat e nineteenth century, there was a push to squeeze as many people as could fit near t he lines; inadvertently creating the first TOD Marshall notes that his hometown of Norfolk, Virginia, as many other cities that urbanized during the early 20 th century, was predicated upon being near the streetcar line (Marshall 2001) In time, horse drawn omnibuses were eventually replaced by the horse drawn streetcar, which in turn, was replaced by the electric streetcar in the 1890s (Giuliano & Hanson, 2004) As the streetcar g ained more promi nence in the inn er workings of our cities, the land use surrounding the streetcar tended to create a market for land near the stations. In time, as each technological advancement replaced the preceding one, cities began to take up more and more land. When serv ing the increasing urban population growth technological advancement s in transport ation made living further out more and more possible, if not desirable. When the first subways appeared in our country, it allowed the working man increased mobility due to its higher speeds; instead of living six blocks from his place of work he now might live a mile or so; the commuter was born. By the early to mid 20 th century, in our larger cities, subways were
32 being utilized in combination wi th city buses and the streetcar (Giuliano & Hanson, 2004) The larger cities were then connected by rail. During this time, our old industrial cities reached some of their highest populations cities like Detroit, Cleveland, Buffalo and St. Louis -which is indicative of the density and the dependence on rail transit (Marshall, 2001) As the automobile began to increase in popularity, the old streetcars were disassembled in many places (Marshall, 2001) City buses would still carry on, but the new spirit of American freedom was the highway. The highway gave way to the expressway. These devel opments allowed workers to live ten miles or more away from their workplaces. The expressway then enabled increased accessibility by eliminating the slower drive on Main Street on the way to the job (Marshall, 2001) Transit by its very nature, does not utilize the highway. The auto centered city now made pasture land that was never valuable before, extremely valuable. This highlights the nature of highways versus the nature of transit whereas the highway tends to spread people out, transit pulls people in, similar to the spokes on a bicycle wheel. Sprawl By the 1950s and 1960s it was the desire of many Americans to escape the noisy, dirty, denseness of the city proper. The highways were designed to make auto use more efficient and decrease congestion. In 1955, President Dwight Eisenhower started the interstate highwa y system with national defense in mind. The idea behind the highway was moving people in cars quickly and efficiently. It enhanced suburban flight as now the worker could commute to his job in the CBD and return home to suburbia A n idea many people boug ht into this idea -whereas before they could live ten miles from work now they could live twenty or thirty. However the expressway construction carved up vibrant downtown neighborhoods. B ulldozing for
33 expressways along with urban renewal programs also en couraged bulldozing vibrant working class neighborhoods, which displaced minorities, sometimes disproportionately. Sprawl is characterized as low density, automobile centered development with a lack of grid like connectivity in the street network (Marshall, 2001) Those types of circumstances have led to our current situation of sprawl and lack of vibrant city cores. The interstates prompted sprawl but even President Eisenhower s on appending six lane suburban boulevards and accompanying subdivisions harbor ing cul de sac after cul de sac (Marshall, 2001) These are symptoms of urban sprawl; while convenient for auto use, these neighborhoods tend to be connected by pedestrian unf riendly intersections that make walking, jogging, or bicycl ing anywhere dangerous (Marshall, 2001) P aradoxically, the most valuable land in a metro area could now be in area where previously there was no accessibility. As suburbia continued to expand further and further out, the interchanges of various expressways and beltways became the new downtown th at was where Americans worked, shopped and were entertained (Marshall, 2001) For the typical American commuter, t he congestion created by sprawl was bearable as long as transportation costs remained (namely gasoline) inexpensi ve. That changed after 9/11 -but some people saw these problems coming; as early as the many transportation professionals realized that adding another lane to an expressway might not solve the congestion problem (Calthorp e & Fulton, 2001) The transit agencies in our cities had suffered by then, with money being spent more on roads than transit (Calthorpe & Fulton, 2001) As more and more people drove cars transit in general became to be asso ciated with poverty and crime. This attitude was signified in property studies in early 1990s Atlanta and Miami. In these areas, that attitude is reflected in lower than expected valuations in home prices The attitude toward transit by the late
34 20 th centur y has changed -living near a transit stop was now in some areas seen as a detriment, instead of an asset. Towards the turn of the century g overnments began to re emphasize public transit spending, but the local nature of government tended to prevent a com prehensive approach to the congestion problem (Calthorpe & Fulton, 2001) As we will see later, Portland was significantly ahead of many metrpolitan areas in transit investment. Transit Taking Us Downtown Planners and concerne d policymakers are now speaking to a more concerned audience than twenty years ago. More and more cities are turning to LRT, to their neglected city cores in the hope of taking us back downtown. Regionalism would become more and more reflective of the new paradigm in transportation planning. The dilemma for policymakers stems from needing money to support both transit and highway, since the two often work in concert. All transit projects are not funded, and most transportation budgets are comprised of a m ajority highway construction and maintenance costs. A delicate highways to transit balancing budgeting job is necessary; many of us living in single family homes tend to use those same highways to access transit anyway. Some Americans are looking to eli minate their dependence on automobiles entirely. Recently there has been a change in attitude supporting the return to city cores that have been under utilized for the majority of the last fifty years. These new downtowns may not be where people work as in the downtown new densities are forming as Americans seek relief from un walkable sprawl and high gasoline prices. Transportation costs now eat up a considerable chunk of househ old budgets (U.S Census Bureau, 2008) These developments have led policymakers, planners, and ordinary commuters to seek other forms of transportation. Such forms include HOV lanes, carpooling, walking, bicycling and increasingly public transit. BRT Commuter rail, and LRT -are all forms of transit that have become favored
35 as Americans look to maximize the efficient use of their transportation dollar. Some commuters in our large metro areas have given up the two hour commute in exchange for living near a tran sit stop. More new miles of track of LRT are being proposed than any other form of transit and many of these areas proposing LRT are in rapidly growing cities. Have we found that we no longer able to sustain a lack of transportation options, due to expen sive gas, declining air quality and two hour commutes? One solution may very well be expanding our nations LRT and fostering connectivity with all transit modes. While transit use has not greatly increased per capita, there are positive signs in the re cre ation of TODs. The TOD seeks to make getting around without an automobile a major feature of living in that development this is exemplified by sidewalks, bikeways, and most importantly walkable access to transit, which discussed earlier, is roughly a mil e away in network distance. While t his study examines SFR, the TOD desire for a return to density, transport options and most of all a divor ce from big box sprawl. The TOD s values are reflected in thi s consumer desire. All of this is good news for landlords who own SFR near transit. Home Valuation T ied to Bundle of Factors Reiterating the research question, d oes having light rail access make a difference in single family home value appreciation? There are many factors to say LRT access is the main impetus, given that houses appreciate in value for many reasons. The externalities are made up of a bundle of benefits. Factors such as number of bathrooms, added home improvements, the numbers of quality sc hools in the area are all hard to quantify It is more likely given how well the LRT system performs. The logic stems from a well utilized system; if LRT is making the population mobile without the automobile, walkable access to the station is desirable If it is desirable to the point the consumer of housing is willing to pay more for increased accessibility,
36 than it is logical that the same property is attractive for a high er level of development (Diaz, 2001) If the property around a station is extremely desirable a n existing low density use such as (Jackson, 2008) SFR can be converted to a higher density use or another type altogether (Diaz, 2001) Trends in Home Ap preciation Home appreciation also reflect s national economic trends. Property values are highly related to state, regional and neighborhood factors and trends. Among growing metropolitan areas a majority are located in the Sunbelt. Logically, single family residential land appreciation reflects the desirability of the region in general. However, land values (and accompanying appreciation rates) can vary greatly from street to street. The cyclical regional and national trends in rea l estate such as the booms and busts of the 2000s can also effect home value appreciation. As shown in Figure 3 2, the country experienced higher home appreciation rates in the 2003 2005 housing burst than any time in the previous ten years (National Association of Realtors, 2008) Figure 3 1 highlights regional trends: t he robust Sunbelt population growth, and the appreciation of SFR values by 11.6 percent in 2004 and 12.4 percent in 2005 in the West. In the Northeast mean SFR values appreciated 11 percent in 2003, 11.7 percent in 2004, and 11.11 percent in 2005. Housing value growth in the U.S. as a whole was higher during these years than any other point since the late 1970s and the South, and the Midwest also posted strong numbers (National Association of Realtors, 2008) Figure 3 2 shows that home appreciation rates were steadily increasing throughout the United States right until the housing bubble burst in 2006 and 2007. As shown in Figure 3 .2, the median sale price decline d in each of these years. All regions saw a decline in the rate of appreciation by 2006, compared to the last three years (National Association of Realtors, 2008) By 2007, the mean national S FR housing value began to actually
37 depreciate which followed another year of depreciation in 2008 something that had never happened before, since data was collected in 1968 (National Association of Realtors, 2008) Concerning this ten year study in Portland, it is important to recognize that the national h ousing value had grown as well. Portland echoed the national trends, but it remains to be seen whether or not LRT access can help home prices withstand the more volatile pric e fluctuations. Nevertheless, there are signs; the current housing depreciation is less acute in certain areas in Denver, Colorado. Homes near light rail stations along the southeast line, which opened in November 2006, have increased by an average of near ly 4 percent over the past two years -but the rest of the Denver market declined an average of 7.5 percent (Jackson, 2008) According to the National Association of Realtors, mean housing values are higher in the West, then the Northeast, then the South and the Midwest being typically the least expensive. Interestingly the Northeast avoided the housing depreciation wave to the same extent that plagued the other regions, as well as the country on the whole. In summary, the fact ors corresponding to home valuation are diversified; the price range can fluctuate so greatly in adjacent neighborhoods that access to transit may be a small reason X house is valued at X dollars. What is significant about the regional price fluctuations i s the areas in the country (the Northeast) that more auto dependent regions (the South). The next section focuses on the different modes of transit and their particular effect on land as well as some factors that affect transit in general. While the South and West saw their home values fluctuate more during the recent home value most historically urban and dense area of our nation.
38 Our old core cities, especially in the Northeast Corridor have blocks and blocks of TOD; though they are not referred to as such. They are remnants of the urban form of a city designed around a streetcar, one of the early forms of transit (Giuliano & Hanson, 2004) Density is probably the single most important factor, although density will not do it alone. The TCRP report recommends that instead of determining minimum density requirements for which transit service types are feasible, the relationship between density (as well as other aspects of la nd use form) and the cost at which transit service can be provide d should instead be determined (Halifax Regional, 2002) This is because decisions about providing services are made in corridors and locations where land use characteri stics, including density, types and mix of uses, are quite varied (Halifax Reg ional, 2002) Further, trip purpose also influences transit service type (Halifax Regional, 2002) In determining how la nd use form such as residential density and employment center size influences light rail and commuter rai l transit demand and service cost, T CRP found that residential densities have more influence on light rail ridership and cost than on commuter rail (Halifax Regional, 2002) In turn, CBD employment density was found to be more important for supporting commuter rail ridership than light rail ridership (Halifax Regional, 2002) Thoughtful transportation planning, i.e., zoning for intense mixed uses a round LRT stations is a good way of ensuring that it will. If light rail, in theory, drives up pr operty value near a station, the n the resulting increase values will result in highest and best use, which would increase density. As far as creating density, the regional picture must be taken into consideration. In Smart growth and the Transportation Land use connection: What does the research tell us ? S he discusses the Knight and Trygg study which points to the necessity of th influence where in the region growth occurs, but only given the right conditions and policies
39 (Handy, 2005) The Knight and Trygg study explored th e importance of four different factors in influencing the impact of transit on land use: local government land use policies, regional development trends and forces, availability of developable land, and the physical characteristics of the area (Knight & Trygg, 1977) They concluded that all of these factors influence the likelihood of development. There is a general logic and a history of the land use transportation connection: the older the city, the more mature and diversifie generally speaking. The land value in these mature transit cities also tends to be high, matching the degree of use. Transit planning may differ in different areas depending on how strong the land use transportation connection is. Generally, the mature transit city will have a high density. Physical Constraints and Economic Concerns While density is the main ingredient, other factors can either enhance or destroy the recipe. It is no coincidence that certain aspects of the land itself lend themselves easily to sprawl. The cities on either coast are limited in some aspect by mountains, bodies of water and other geographic features. Cities located on rolling expanses of flat te rrain are not ideal in keeping density intact. Cities like Houston and Phoenix -some of the most sprawling congested cities exhibit similar characteristics. Kain discusses this in a study comparing ridership statistics of San While Sa n Diego s small er CBD tends to attract few er ridership numbers but this disadvantage is off set by the area (Kain & Zvi, 1999) In contrast t o Houston, which is built on an unbroken expanse, San Diego is h emmed in by the Pacific Ocean on one side and a steep mountain range on the other. As a result, a large fraction of San Diego s jobs and households are concentrated along North South corridor along the bay and a large fraction of the remainder in a narrow East West valley. Not surprisingly, the fi rst and second lines of San Diego s light rail system were built in these corridors (Kain & Zvi, 1999) When transit technologies develop, they tend to accelerate the decline of older m odes of
40 transport. Our present day transportation modes will be discussed with an emphasis on the rail modes, especially LRT. Portland exhibits the type of physical characteristics that induce a vibrant city core. These characteristics include physical f eatures that ser ve as barriers to sprawling development: t hese include mountains, oceans and rivers be New York without these prohibitive fea tures the same is true for San Francisco These features are absent in cities like Phoenix, Orlando and Houston Portland has no shortage of physical constraints on development the city is divided by the Williamette River and by the Columbia River to the north. It is also bounded by a significant mountain range di rectly northwest of the CBD. Consider Buffalo, New York over a twenty year period) the light rail line has no s i g nif i cant effect on land use, employment or central city economic development (Banister & Berechman, 2000) current T is going to bring back our CBDs then a population decline in total metropolitan area (people leaving for the suburbs and people leaving the suburbs altogether) is difficult to overcome. If there is chronic high unemployment than an LRT can have all the accessibility in the world but if jobs are leaving the area the n an LRT will not help revitalize a CBD. Economic impacts are highly localized and tend to occur in fast growing, heavily congested core areas (Giuliano & Hanson, 2004 ) Portland has grown while Buffalo has shrunk. of high tech as well as traditional industries native to the Pacific Northwest. Portland also is
41 home to major employers such as Intel and Nike. Portland also, in terms of tonnage, is the third busiest port on the West Coast (Portland Business Alliance, 2005) Portland has historically had a lower unemployment rate than the national average but a significantly higher cost of living (Portland Business Alliance, 2005) Portland suffe red an overall decline in population between 1950 and 1980, however not to the extent of Buffalo. The city seemed to be able to reinvent itself an attract people back to the core c ity by 1990 the population had increased 19.4 p ercent from a decade earlier. By 2000, the population had grown by another twenty percent (U.S Census Bureau, 2008) Again, the local economic situation precludes any stand alone effects of LRT on home prices. LRT can help rejuvenate but is not likely to drive the engine of an economy entirely. How Portland Changed Portland was facing similar challenges in the 1970s a nd 1980s. In 1973 the state of provisionally in 1976 and finally in 1979 (Calthorpe & Fulton, 2001) The boundary was conceived primaril y to protect farmlands from speculation and development (Calthorpe & Fulton, 2001) However, Portland at this time was still similar to any other metro area; the UGB boundary was so large that it took development nearly twenty years to reach it (Calthorpe & Fulton, 2001) A common misconception about the UGB is that it is a fixed boundary, when actually it is elastic it must be adjusted periodically to provide enough land for twenty years of growth (Calthorpe & Fulton, 2001) Metro, which began in 1979, finally began in 1992 to address the alternative transportation modes as a response to this sprawl (Calthorpe & Fulton, 2001) This effort was initiated by a new state Transportation Planning Rule that required cities with populations larger than 25,000 to revise their transportation plans to provide more emphasis on such modes,
42 requiring pedestrian friendly designs along transit lines and street connectivity in subdivisions (Calthorpe & Fulton, 2001) The MPO s in lieu of the Transportation Planning Rule had to adopt these plans that sought to decrease VMT per capita (Calthorpe & Fulton, 2001 ) It was a united front against auto dependency Metro was addressing VMT per capita in its Regional Urban Goals and Objectives, which called for a new regional plan (Calthorpe & Fulton, 2001) The plan would be known as Region 2040. Around the same time, a nonprofit environmental advocacy group called 1000 Friends of Oregon began its study of an alternative land use transportation strategy (Bartholomew, 1997) L and U se T ransportation A ir Q u ality (LUTRAQ) : The Stimulus for a Regional Approach he was the defeat of the Western Bypass Highway i n 1988 another example of a concerned citizenry defeating a major high way plan (Bartholomew, 1997) The proposed bypass was an attempt to relieve congestion in Washington County, where rapid growth had been to that point. The political leaders of Washington County were solidly behind the proposed extension. This was a departure from the norm for the region building more highways to alleviate traffic congestion. Comprised of citizens, activists and developers, L and Use Transportation Air Quality (LUTRAQ) was born in a res ponse to this proposed hig hway (Bartholomew, 1997) The group had to convince the state that their alternative land use transi t to not building the b ypass could support the densities that the projections called for (Bartholome w, 1997) Workshops were Instead of building more highways LUTRAQ proposed alternatives that included more emphasis on pedestrian access to regional transit, and moderate density (Bartholomew, 1997) It was a very politically charged situation Transportation (ODOT) found the LUTRAQ alternative to be equal or superior to the b ypass in virtually every category. This was followed in September
43 1995 with ODOT's official recommendation from the study process, c alling for various smaller road improvements, but no bypass (Bartholomew, 1997) This defeat of a highway was significant because it was one of the few occasions that a land use transportation alternative was taken seriously enough to defeat a new expressway (Bartholomew, 1997) LUTRAQ is significant because the voters m ade a conscientious decision to advocate transit options over expanding the highway system. This would set the stage for the LRT expansions that Portland would add every few years. approach to transportation, there are a bundle of factors that are tied in home valuation and increasing rail transit ridership. The factors tied to home valuation range from everything to location to number of bathrooms in the unit. Some of the factors t ied to increasing transit ridership (which can signify a healthy downtown ) include density and intensity of land use, the physical geography of the area, the economic trends of an area but perhaps most importantly citizens towa rd transit.
44 Figure 2 1. Percentage of transit t rips in United States in 2000. Source: The Geography of Urban Transportation Third Edition, 2004. Table 2 1. Cost and other selected c haracteristics for selected transit modes Mode Typical capital cost per vehicle (m illions of dollars ) Annual operating cost per revenue m ile Seated c apacity (p assengers) Average s peed (MPH) Maximum s peed (MPH) Light rail 3 25 Between $7 $15 150 300 per train/ 100 per streetcar 15 25 60 Heavy rail 20 250 Between $8 $10 60 80 per car plus standees 30 70 Commuter r ail 5 9 Between $11 $13 1,250 per train (depending on length) 30 90 Bus rapid t ransit 0.2 0.5 (shared lane) 8 25 (dedicated lane or guideway) Between $6 $8 60 100 per bus 30 60 Source: Florida Depa rtment of Transportation SEFCC S tudy (Jacobs Engineering, 2008) Bus services 58% Heavy Rail 30% Light Rail 4% Suburban Rail 5% Other 3% Bus services Heavy Rail Light Rail Suburban Rail Other
45 CHAPTER 3 LITERATURE REVIEW This literature revie w is divided in three sections. T he first section discuss es the theoretical framework established by Von Thunen in the mid 19 th century and furthered by others in the latter part of the 20 th The second part discusses the previous studies of the effects of LRT on property values. The third section examine s the myths and claims about LRT (Bani ster & Berechman, 2000) This thesis aims to quantify desirability in terms of housing appreciation rates. Land that offers increased alternative transportation options may have higher appreciation rates. Some urban planning studies examine how land use and transportation are connected, and in urban economics studies one may learn that accessibility and mobility affect location value. In real estate price studies, one learns that as a location become s more attractive, as a result of certain characteristics, demand increases. This will result in increased price valuations as being close to a transport facility increases the accessibility of the property and thus the value of the transport facility is capitalized in the property value (Debrezion, Pels, & Rietveld, 2004) There are many factors or characteristics that determine how much a house worth, as access to transit is just one of many. Theoretical F ramework Although in terms of total track mileage, light rail has a long way to go, the wave of rail transit future points towards light rail as opposed to other forms. In 2006, new light rail construction has most likely surpassed new commuter rail construction in the United States. More mid size cities have p lanned light rail systems than commuter rail. Interestingly enough the cities that have new light rail routes are Sunbelt cities that have seen increased population growth in the last twenty years.
46 The motivation for living near LRT has been greater as tr ansportation costs per household begin to rise. As gas prices increase, one motivation for living next to a LRT stop is the savings afforded by using transit compared to rising costs at the pump. For the consumer of land the choice could to move (for a va riety of reasons) to an LRT stop interests economic geographers. Location theory is concerned with the where and why of economic activity. Johann Von Thunen, who in 1863 for a given and of a given fertility, land value differentials are derived as quoted in (Debrezion, Pels, & Rietveld, 2004) This is one of the first theories concerning the bundle of the CBD, he theorized that the most important land uses would be centered in the central core. Von Thunen was primarily concerned with balancing land costs and transport costs. weary commuter -trading the ranch house in the suburbs for the more expensive town home downtown. Economist William Alonso conducted an important study in the 1960s improving Von rent model is that every agent is prepared to pay a certain amount of money depending on the location of the land. This leads in equilibrium to a rent gradient that declines with distance from the CBD for sites that yield equal equity (Muth, 1969) In the early studies, a dominant factor explaining the difference between land (property) values was the accessibility as measu red by the distance to the CBD and the associated transportation costs (Alonso, 1964) The money saved from not having to travel is what motivates the intense land uses found in our most dense city cores. The people who live in these areas often are able to perform many daily trips w ithout using an automobile for longer trips to work may be used by transit. Even longe r trips may be used by a combination of different modes.
47 Von Thunen, Muth and Alonso understood that transportation and land costs involve a trade off The one thing that all transit systems have in common is their attempt to improve Accessibility refers to the number of opportunities called (Giuliano & Hanson, 2004) Activity sites would include inc luding employment, entertainment, recreation and the CBD. Mobility refers to the most efficient way to move people between different activity sites (Giuliano & Hanson, 2004) The goal of transportation planning has been to inc mobility (Giuliano & Hanson, 2004) Thus, mobility puts a high premium on collective modes of transport and the accessibility of places has a major impact upon their land values (and hence the use to which the la nd is put) and the location of a place within the transportation network determines its accessibility. Accessibility also includes the concept of attractiveness, the opportunities or activities that are located in a given place. Thus in the long run, th e transportation system (and the travel on it) shapes the land pattern (Giuliano & Hanson, 2004) Alex Marshall and others expressed the importance of this coordination A way to measure the value of transit proximity, and mo reover, success and health of a city is to examine its house value appreciation rates. If more people want to live in a place, that place becomes inherently more valuable, and the price of the land will increase over time. Based on new construction from re cent years, LRT is quickly becoming the mos t popular form of rail transit. As mentioned earlier, a fixed route service may be a better engine for realizing economic potential, more so than a highway or bus. What is Walkability? Much is made in this study on the benefits of walkability. A quarter mile distance is theoretically the furthest distance people are willing to walk to transit as opposed to getting in their cars (Hess & Almeida, 2006) The terms actual distance and network distance are refer to
48 how long it takes to get somewhere (be it car or on foot) network distance can be quite longer than actual (Hess & Almeida, 2006) gated com munities can make more difficult than it might appear on a map. Through these street often make using an automobile necessary for what might have been a short walk Also, a satisfying the mile buffer) if to get to the stop busy inters ections or freeway junctions between the person and the stop. In previous studies, observers have noticed that inconsistent results and weak evidence of a land use transport connection may be due to the complexity of metropolitan development and unpredictable travel patterns (Ryan, 1999) It was also found in the P a rsons B rinckerhoff (PB) fects on property (Parsons Brinckerhoff, 1999) Researchers have attempted to isolate the effects of access to transit in an eff ort to try to predict. This thesis seeks to correlate the two factors of access and home appreciation, and then compare these correlations to income and length of transit service. Determining Value R eal M arket V alue (RMV) or E stimated M arket V alue (EMV) is the dollar amount that real estate professionals use to conduct property transactions (National Association of Realtors, 2009) Home appreciation rates fluctuate greatly in any given area due to the factors mentioned earlier. balance in rapidly increasing property values the more the house is worth; the more its owners have to pay in property taxes (Jackson 2008) A national trend in single family home value growth welcomed the nation in the early to mid 2000s due to many factors, and as we have seen, desirable Sunbelt areas tend to populate
49 more robustly than their northern neighbors. As of 2008, howeve r, there are coastal areas or Sunbelt areas that are depreciating in SFR value just as fiercely as they appreciated. Consider that in some regions, mean SFR value depreciated in back to back years, all American cities would do well to seek to eliminate any unnecessary detractors from the growing housing values. Methods to Isolate LRT Impact s Researchers use three approaches to separate out the effects on transit on land values; (1) matched pairs; (2) repeat sales analysis; and (3) hedonic price models (Cervero & Duncan, 2001) All of these methods implicitly assume that markets have fully captured information abou t the contamination and that stigma impacts are fully reflected in prices (Kilpatrick, 2005) The first two approaches are the easiest but also the most problem prone (Cervero & Duncan, 2001) Repeat sales extract value trend changes immediately after a station implementation, thus isolating of the event itse lf and controlling for exogenous factors (Kilpatrick, 2005) Matched pairs determine whether the scores of the same participants in a study differ under different conditions. For instance, this sort of t test could be used to determine if property values improve more after an implementation of a new light rail stop after taking a writing class than properties they did without a transit stop. In this study, the Parkrose/Sumner case study could be examined as far as what was the median home price before and after 2001 (when the LRT stop was implemented). Hedonic modeli ng or other mass appraisal techniques, preferred when large data sets are available, use data from both within and without a n affected area and a dummy variable to capture the effect of stigma on value. When distance to a source of effect is a factor, then some sort of spatial analysis or distance function is included in the hedonic pricing model (Kilpatrick, 2005) Gatzlaff and Smith (1993) used repeat sales data to examine the impact of the system on land values. They argued th at a change in land value is a precursor to change in a land use, so is
50 a good measure of accessibility related impacts. They conducted a series of repeated sales analysis, which compares the change in sales prices of a given property in successive sales. Repeated sales within station areas and outside station areas (1 square mile) and outside station areas were compared; there was no significant difference between groups (Gatzlaff & Smith, 1993) Pioneered by Bailey et al. (1963) this method estimates the effect of an exogenous amenity shock on home value by analyzing the change in a hom tra nsactions (Kent & Parilla, 2007) According to Kent, the repeat sales methodolog several advantages over a hedonic based analysis. First of all housing price changes are grounded in real estate market transactions, which avoid the potential inacc uracy of third part (Kent & Parilla, 2007) This methodology seeks to eliminate the arbitrary nature of properties that were sold. Hedonic price models introduce more rigorous controls, and are th us widely to be considered the best method available of ascribing benefits associated with factors like proximity to transportation facilities (Cambridge Systematics, Inc., 1999) Cervero and Duncan in their study of rail transit effects on commercial land values in Santa Clara County, California state that Hedonic price models apply the technique of multiple regressions to apportion real estate values to various explanatory variables, shedding light into the marginal contr ibution of factors like accessibility, land use type and neighborhood quality to sales values For purposes of gauging land value benefits, a normative hedonic model is generally of the form: P i = f (T, N, L C) Where: P i equals the estimated price (per square foot) of parcel i. T is a vector that gauges proximity to transportation facilities. N is a vector of neighborhood characteristics (e.g., presence of mixed land uses; median housing income L is a vector of location and regional accessibility attributes (e.g., accessibility to jobs) C is a vector of controls (e.g., fixed effect variables) (Cervero & Duncan, 2001, p. 4)
51 Single family residential was used i n many studies due to the nature of residential land and accessibility. Location theory holds that home prices rise in synch with travel time savings, thus to the degree transit expedites travel, properties near stations should sell for more. It is interes ting to note that a 2002 study completed by Cervero and Duncan found that light rail conferred the greatest land value benefits to m ulti f amily r esidential and commercial properties (Cervero & Duncan, 2002) Nelson also l uses than for commercial ones (Parsons Brinckerhoff, 1999) Landis noted three reasons for this problem: (1) a lack of comprehensive and reliable data; (2) a smaller zone of impact that limits the number of observa tions; and (3) while housing va l u es are determined in the market place, the values of individual corporate transactions may represent only the value of one pair of buyers and sellers (Land is, et al., 1995). It has been theorized that proximity to light rail lines could have negative effects on p roperty; a study by Chen did not find a ny statis ti cally significant nuisance e ffects (Parsons Brinckerhoff, 1999) Dia z maintains one of the more prominent ways that people understand the value of property is through the price or value of a home or in the rent they pay (Diaz, 2001) Another reason for examining SFR is that the number of real estate consumers (owners/renters) is greater than for oth er types of real estate (Diaz, 2001) All in all, the effects are more acutely felt in the residential sector (Diaz 2). Previous Literature on LRT and Residential Property concerning LRT and home values. Studies began to appear in the late 1970s and 1980s. Previous research on light rail and residential proper ty values suggests that light rail can have two opposing influences on property values: a positive accessibility effect and a negative nuisance effect (Garrett, 2004) The positive effects have been well covered throughout this thesis and the
52 nuisance effects include lighting, noise the closer the home is to the tracks (Garrett, 2004) However the various studies that have examined LRT impacts on property values do not offer a consistent relationsh ip. A list of similar studies is listed in Table 3 1 Referring to Figure 3 2, a graph of inflation adjusted American housing values, the housing boom of the mid 2000s will be reflected in higher appreciation rates across the board regardless of how clos e the home is to an LRT stop (National Association of Realtors, 2009) Numerous studies have been conducted on Portland, seemingly since its constructed light rail in 1986. One of the first studies was in 1988 when Al Mosaind c onducted a study done on from stops on the corridor. Al Mosaind actually conducted two studies; the first used sales of homes that were located within a 1,000 mete r width along the line but distinguished those that are within 500 meters of actual walking distance to a station; the second model used a distance measure to detect price gradient of homes within 500 meter actual walking distance. Al Mosaind only examine d houses that were sold in the study period this could influence data -the study did not delve into how homes in or outside the buffer appreciated but Al Mosaind concluded that a net benefit indicated that the positive effects of accessibility are stronge r than the nuisance effects. Studies after the Al Mosaind report have focused generally on the Eastside MAX now termed the Blue Line and have found that proximity yields positive effects on property values; Lewis Workman and Brod in 1997 found that assess ed property values increase $76 for every 100 feet closer within one station (Lewis Workman & Brod, 1997) In 1998, Chen found that starting 100 meters away from a station property value decreased $32.20 per meter further (Chen, Rufolo, & Dueker, 1998) A year later, Dueker and Bianco using sales prices determined
53 (Dueker & Bianco, 1999) They used distance to the station as a proxy of accessibility and distance to the line itself as a proxy for nuisance effects. They found that The real estate market has re sponded positively to LRT. The model estimates that a house that is valued at $82,800 (median price of housing in sample) at an LRT station would be valued at $80,500 a distance of 200 feet away, $78,554 a distance of 400 feet away, $76,961 a distance of 600 feet away, $75,721 a distance of 800 away and $74,835 (10% reduction) a distance of 1000 fe et away from a rail station. This willingness to pay a premium for single family housing having LRT access is a significant and positive land use benefit of the LRT investment (Dueker & Bianco, 1999) llas study used median residential values and then examined the changes in property values near Dallas Light Rail between the years 1997 and 2001 (Weinstein & Clower, 2002) In Table 3 5 the median residential home value of 4 ,393 properties outside of a quarter mile buffer zone appreciated 19.5% of their 1997 value compared to a 3,262 homes inside mile buffer zone increasing 32.1% over the same time period (Weinstein & Clower, 2002) Table 3 5 al and industrial parcels (21.5% to 13.0%) actually appreciated more in areas without access to rail transit in the same time period (Weinstein & Clower, 2002) This study catego rized quarter mile buffer near light rail than those not within (24.7% to 11.5%) (Weinstein & Clower, 2002) Two things Weinstein and Clowe enhancing service most keenly affecting the market values of properties where people live and where there are comparatively high concentrations of non industrial jobs (Weinstein & Clowe r, 2002) It also implies that LRT may be more consistently beneficial to residential than other types of land.
54 LRT and N eighborhood I ncome As mentioned earlier, the economic status of prospective transit neighborhoods may have an impact on LRT ridership and LRT proximate property values. Researchers have found that proximity to a railway station is of higher value to low income residential n eighborhoods due to their heavier reliance on transit (Bowes & Ihlanfeldt, 2001) The dependency of a neighborhood's residents on transit may play a role is how valuable that transit is residents This value ma y not translate to higher values in poorer neighborhoods, however. It would follow that a transit stop is more valued in neighborhoods that depend on them. For every foot closer to an LRT stop in an economically depressed area, it would make sense that th ose properties would be more valued. Not so. Hess and Almeida found that in Buffalo, New York, the proximity benefits were more acutely felt in high income areas than low ones. The study on Buffalo found that within a half mile radius of LRT station, prope rty values were $2.31 greater for every foot closer to the station (Hess & Almeida, 2006) These results were promising but the overall picture in Buffalo is mixed. The study found that based on income level the effects of be ing close to LRT were varied. For three station neighborhoods that had below $20,000 median income the housing value decreased for every foot closer -one st ation area neighborhood decreased by $42.31 for every foot closer (Hess & Almeida, 2006) In Buffalo, the conditions of this Rustbelt city (population decline, crime, economic downturn) overpowered the presence of a rail system (Banister & Berechman, 2000) Truly, the economic situation affecting a metro area has an effect on home appreciation regardless of the LRT access. Implementation of a New Transit S tation Even though the economy holds sway over hom e values and home appreciation, i t bears watching what the effect of the announcement, constru ction and general speculative anticipation of a light rail station can have on these same property values. Transit systems take shape in
55 several stages: route selection, site selection, clearance and displacement, construction, early operation, and mature operation. The timing of impact is to be fairly measured and this measurement used to determine financing for transit projects. Speculative activity may begin well before construction of a new system (Gatzlaff & Smith, 1993) However, research fails to explain the causality of an LRT station opening on property value (Kim, 2008) or basically it does not necessary follow that if land prices jump once a rail service begins that transit caused this ap preciation (Cervero & Duncan, 2002) Spikes in land values could be attributable to other factors, like an upswing in the regional economy, improved highway conditions, or better schools (Cervero & Duncan, 2002) There is enough data to suggest that the implementation of a new light rail station could inflate the prices of nearby residential parcels in anticipation of increased accessibility. The Hiawatha Line study (Minneapolis St. Paul) conducted by Kent and Parilla examined the effect on SFR near seven of seventeen new light rail stations in 2004 in the city (Kent & Parilla, 2007) They compared EMVs from 2003 2006 to a comparison period spanning 1997 2000. Working with straight line distance, they used a buffer system to classify parcels as being a half mile, half to one mile, a mile, or two mi les from a station stop (Kent & Parilla, 2007) The study found a similar appreciati on rate inside and outsi de the half mile buffer (Kent & Parilla, 2007) They found that after the stop (2003 2006) that the mean EMV of parcels within the half mile buffer experience a significantly larger increase in mean EMV across all time periods (Kent & Parilla, 2007) The mean EMV within the half mile buffer were consistently valued higher than outlying parcels. However when considering the percentage change in EMV, the parcels outside the ha lf mile buffer appreciate significantly more than those within (Kent & Parilla, 2007) That suggests that the Hiawatha Line had no effect on EMV for homes close to
56 the light rail; higher income neighborhoods with more desirable amenities may explain the disparity Gatzlaff and Smith conducted a similar study in Miami in 1993. The study used sales of homes, known as repeat sales analysis, through tax records and only included homes that were sold between 1971 and 1990 ne ar it is commuter rail (Gatzlaff & Smith, 1993) This study also differs because it examined the effect of the implementation of a new station, whereas many studies te nd to examine data after the fact. The study concluded that at best there was a limited positive effect on a new rail station in proximate residential a reas. The study is significant in that the study used repeat sales analysis to highlight the differenc es of appreciation rates of homes in proximity to commuter rail stops and the median home price in Miami. Between 1978 and 1981, the study showed a (Gatzlaff & Smith, 1993) In 1981, (Gatzlaff & Smith, 1993) This implies that commuter rail transit could be a nuisance -that is have a negative effect on SFR property value -and is often cited as one of the few studies to do so. Negative Externalities Affect Property Value As sho here are potential Perhaps the home is a high crime area, or surrounded by abandoned homes? The negative ex ternalities can lower home values or slow home appreciation rates for the same exact home in a different area. Al Mosaind delved into these negative externalities that include noise, increased pedestrian and automobile traffic near the station, attracti on of undesirable groups to neighborhoods, and the
57 disruption and noise associated with the construction of such stations (Al Mosaind, Dueker, & Strathman, 1993) Al net positive impacts could be observe d if the market viewed improved accessibility more as a benefit than a nuisance (Al Mosaind, Dueker, & Strathman, 1993) Diaz wrote a comprehensive paper that summarized twelve rail projects including both light and heavy r ail in North America and found that some studies show that proximity to industrial uses or highway facilities may limit the extent to which pro p er ty values are increased (Diaz, 2001) In a study done on Atlanta properties within a quarter of a mile from a rail station were found to sell for 19 percent less than properties bey ond three miles from a station. However, properties that were between one and th ree miles from a station had a significantly higher value compared to those farther away. These resul ts suggest that houses that are very close to stations are affe cted by negative externalities, but those at an intermediate distance are bey ond the externality effects and b enefit from the transportation access p rovided by the stations (Bowes & Ihlanfeldt, 2001) affirms the possibility that rail stat ions may affect property values indirectly by increasing crime or retail a ctivity within the (Bowes & Ihlanfeldt, 2001) There are many reasons for the results -first of all, commuter rail is noisier and creates more poll ution that light rail. The second, and perhaps more important, is that in Miam i the Metrorail was not implemented in a centralized corridor and had no discernible developmental impact around the stations indicating the rail had little effect on accessibility (Gatzlaff & Smith, 1993) Cervero and Duncan found that land value premiums only accrued to single family housing units within a half mile of rail stations on two corridors, and that they generally fell within half mile ring of stations on other corridors (Cerv ero & Duncan, 2002) This study also
58 found that being near commuter rail stops in the wealthier areas tended to depress multi family real estate prices. That same study found that multi family housing in areas closer to the CBD accrued higher capitaliz ation benefits (Cervero & Duncan, 2002) This same study found that depending on whether the units were for sale or rental units there was a value added of some $85,000 per condominium in the richer North County corridor the same corridor that conferred negative effects on commercial parcels. rail in Atlanta study (MARTA) study, using sales prices, he found that rail transit proximity increased in low income neighborhoods bu t decreased in high income neighborhoods (Nelson A. C., 1992) A MARTA study in by Bowes and Ihlanfeldt (2001) found that properties decreased by 19 percent within a quarter mile compared to properties beyond three miles (Bowes & Ihlanfeldt, 2001) LRT E ffect on Commercial Properties Studies on light rail systems e ffect on commercial properties are scarce (Cervero & Duncan, 2001) A study of the DART system compared d ifferences in land values of loosely near and not near LRT stations (Weinstein & Clower, 2002) The average percentage change in land values from 1994 to 1998 for retail and office properties near DART stops was 36.8 percent and 13.9 percent (Weinstein & Clower, 2002) Examining the same area, Weinberger in 2000, conducted a study examining the relationship and found while properties within one half mile to a station commanded a premium, properties that were one quarter to one half mile co mmanded an even higher premium (Weinberger, 2000) Seeking to improve on this work, Cervero and Duncan conducted a stud y on the effect of rail transit on commercial properties in Santa Clara County, California (Cervero & Duncan, 2001) Santa Clara County has both commuter and LRT in i ts system; both were examined. Using the quarter mile distan ce settings, it was found that the capitalization benefits
59 near LRT were around per square foot, smaller than those associated with commuter rail in business districts (Cervero & Duncan, 2001) These findings also reinforce t he effect transit has on more intense land uses as opposed to lower density uses such as SFR. Conclusion of Previous Studies In summary, Table 3 1 highlights some of the previous studies that have been conducted examining the effect LRT proximity on prop er ty values. The effect of LRT on property value is an inconsistent picture. While t his thesis examines yearly home appreciation, fewer studies dealt with that aspect specifically Referring to Table 3 5, the Dallas study that dealt with land value appr eciation found that from 1997 2001 residential parcels within walking distance did appreciate more than the parcels that were outside the buffer P roximit y to a LRT station can have an effect on property values residential or commercial be it positive or negative. Von Thunen, were he alive today, would point to the proximity as a methodto decrease transportation costs Some studies indicating a contradiction to Von Thunen would point to the decreased desirabi lity of living next to LRT due to the negative externalities (noise, pollution) of rail transport. Generally the literature found that the LRT correlation is strongest and most positive among residential parcels. Based on the literature, MFR has shown to have greater benefits than SFR. Other forms of rail transit also had positive effects on residential land as well. d picture, there were cases of a positive correlation as well. The implementation of a new LRT station may cause an unsustainable temporary spike in adjacent SFR values. A new resident may gladly pay extra for a home near LRT for increase d accessibility to employment locations, entertainment, recreation, and other amenities Because o f increased accessibility to these places, l andowners seeking to rent homes may bid up those properties and capitalize those benefits into the pr ice of the property (Hess & Almeida, 2006)
60 ccess to L RT does not actually guarantee land valuation on a bundle of factors that are out of the realm of human control. The following section will highlight characteri stics that attract LRT ridership: multimodal integration, intensity of land use, density of residential population, parking and auto dependence, area climate physical and economic characteristics Attract ing LRT R idership To maximize land valuation benefits of proximity to an LRT, the LRT must be utilized by the public it is intended to serve. Certain characteristics of a region can determine how Those factors include multimodal integration, land use, land density, physical and economic characteristics, levels of transit service, area climate, and All these factors often are apparent in successfu l transit systems around the world. The assumption for this thesis is that the more riders on an LRT, the more likely homes are to benefit from increased values by being near an LRT. The following section seeks to clarify that LRT will not be successful by itself; it needs a few complimentary factors and/or situations to ensure highest possible patronage. The most important of these is density without density; it is unlikely that any LRT would be successful. The rest of this section reads much like a checklist for transportation planners who are inquiring The purpose of this section is to underline the Multimodal Integration It is apparent that no successful city transit system only relies on one form of transit. For a successful transit system, the many different forms of transit must be accessible and integrated (Calthorpe & Fulton, 2001) A transit center or hub is an area that has one or more accessible
61 connections to bus, BRT, commuter rail, light rail, heavy rail, air traffic, ferryboat, pedestrian and bicycle ways in a s pecific location This would seem to indicate to lead to an increase in LRT use (as well every other mode) because the accessi bility and mobility is enhanced (Kuby, Barranda, & Upchurch, 2004) With multimodal integration, pedestrian, bicycle, or bus trips are easily integrated and accessed fr om an LRT access point. Intensity of Land Use A round Stations from transportation savings the location of the land offers. Regardless of mode, the early economic geographers were correct when offering that the pull of the CBD would establish valuation based on distance from downtown. The most intense land use and highest densities in this country occur in New York City. A mi x of residential and commercial land use is most desirable for the areas adjacent to an LRT statio n W ithin compact urban regions, transit service in corridors that contain a variety of residential and non residential activities will prove especially attr active and competitive (Parsons Brickenhoff Quade and Douglas, Cambridge Systematics, 1996) successes from the not so successful. Density of Popula tion a round Stations Density follows suit in the aspect that high density residential is optimum for TODs as well as high density employment centers Density could very well be the most important factor for transit success the more trips made per mile the more the need for alternative forms of transport. Light rail needs density to succeed. Pushkarev and Zupan note that nine dwelling units/acre is the minimum residential density for a light rail system (Puhskarev & Zupan, 1982) The most successful transit systems are metropolitan areas with high density. Density and land use are intertwined because as density increases, land use intensifies. As land use intensifies,
62 there is potentially a need for a more intense form of rail t ransit. Pushkarev and Zupan recommended that if a LRT has five minute peak headways then nine dwelling units per acre are required through a 25 100 square mile corridor (Puhskarev& Zupan, 1982) Generally density around LRT stops will be apparent because of the nature of utility maximization this can be illustrated in the growth of multi use TOD. The physical characteristics also can enable higher density of settlement as well as affect land valu es positively Parking As mentioned earlier, auto access has limited capitalization effects on property values. However, p ark and ride access also enhances transit ridership by making stops more accessible (Litman, 2008b) Some may suspect auto use does not promote and in fact discourages transit ridership. Kuby noted in Factors I nfluencing Light Rail Boardings in the United States t he two intermodal variables number of bus connections and number of park and ride spaces are both significant at the 0.001 level [ b ] oth have positive b coefficients that make a good deal of sense. Each additional bus route interse cting with a station yields 123 weekday boardings, while each additional park and ride space nets 0.77 boa rdings (Kuby, Barranda, & Upchurch, 2004) Parking is indeed a crucial issue for many who use a combination of auto and transit commuting, especially. The balance between too much and not enough parking, as well as whether to charge for parking in attempt to stimulate ridership numbers -is a challenge for transportation/transit planners (Litman, 2008b) To summarize, park and ride lots can help increase ridership while removing people from their automobiles. Levels of Service and Connectivity A good transit system is typically characterized by good Level of Service grades. Level
63 frequency of servic e. Examine Tables 3 2 and 3 3; these two aspects of transit length and frequency of service are designed with the passenger in mind. The more convenient the service is the higher the grade. Systems that do not run very often or very long typically may n ot capture potential ridership as well as systems who score high in these areas. Connectivity refers to the ease as well as the number of connections between transit modes (Wilbur Smith Associates, et al., 2006) Connectivity is strengthened if the different transit modes schedules are coordinated (Wilbur Smith Associates, et al., 2006) For example if a home to work trip requires a bus to light rail transfer, then it requires the trip ma ker to coordinate the trip. To continue the example if the LRT has thirty minute headways, the rider needs the bus to arrive before the LRT departs. The schedule should provide some leeway if the bus arrives at the same time the LRT departs then the ride r does not have time to make the transfer. Shorter headways (less time between bus/rail arrivals) make the connectivity issue less important than say, hour long headways where missing the train results in intolerable delays (Wilbu r Smith Associates, et al., 2006) To conclude, the connectivity of transit systems requires s chedule coordination -which is only possible if service runs on time (Wilbur Smith Associates, et al., 2006) Reliability matters -i f the transit is repeatedly late and/or misses connections -riders will seek alternatives and likely return to their cars (Wilbur Smith Associates, et al., 2006) However if the LRT does not connect people to where they need to g o, then more frequent headways alone will not increase LRT ridership. Major destinations such as airports, e mploymen t cen t ers entertainment areas and riders homes should be connected, if not LRT ridership may suffer (Wilbur Smith Associates, et al., 2006) D oes the LRT go where people need to go? Does it offer a reasonable choice to car or bus? A 1996 study used a cross sectional regression analysis of station level data to study land use characteristics that make a c orridor
64 conducive to light rail (Parsons Brickenhoff Quade and Douglas, Cambridge Systematics, 1996) They combined data from 261 stations on 19 lines in 11 metropolitan areas, and their dependent variable, daily st ation boardings (Parsons Brickenhoff Quade and Douglas, Cambridge Systematics, 1996) centers, transit systems that link the CBD with sub regional employment centers will be especially cost effective, offering opportunities for two direction flows at all times of the day (Parsons Brickenhoff Quade and Douglas, Cambridge Systematics, 1996) (soon to be four) light rail li nes connect major destinations in the area and exhibit a regional connectiv ity that services the suburbs (TriMet, 2008) Area Climate Transit ridership depends on many factors, some of which are beyond the realm of human control. An an other st rong indicator of LRT ridership (Kuby, Barranda, & Upchurch, 2004) Perhaps the most important predictor of transit success is a temperate climate in which the rider feels comfortable waiting f or transit A 2002 study found that degree days carried a negative coefficient of 1.52, meaning that more extreme temperatures discourage LRT ridership (Kuby, Barranda, & Upchurch, 2004) Cities blessed with temperate climates such as San Diego, can expect up to 300 more boardings per st ation than average (Kuby, Barranda, & Upchurch, 2004) Cities like Buffalo, Cleveland and Salt Lake City which experience more extreme temperatures would see boardings at each station reduced by a similar amount (Kuby, Barranda, & Upchurch, 2004) Kuby also states that cities with extreme climate s like Phoenix and Minneapolis ities, where walking (Kuby, Barranda, & Upchurch, 2004) Given the nature of the Sunbelt LRT expansion, planners would hopefully take notice of the preceding study. Is it
65 a coincidence, at least in this country, that warmer climate cities are also some of our most auto dependent? Auto Dependence Cities that have large patronage of transit are m ore suitable for rail transit than auto centered citie s such as Miami or Los Angeles (Mackett & Sutcliffe, 2003) A study conducted by Mackett indicated that Miami was particularly unsuitable for large transit patronage due to its auto centered nature Mackett constructed a table that highlighted factors conducive to a city having successful commuter rail ridership. Miami also scored low in terms of the collective public attitude toward s public transit. American cities consi stently scored lower than their counterparts in Canada and the United Kingdom (Mackett & Sutcliffe, 2003) In Cockerill and meta stu dies in both Miami and Dallas, where the effects of residential properties were actually orrelation between car use and public perception of transit (Cockerill & Stanley, 2002) Public Perception of Transit In general, cities that have a positive image of public transport are more likely to patronize light rail than cities where the public perception of transit is of criminal activity (Mackett & Sutcliffe, 2003) The image of pub lic transport, which is based on socioeconomic factors, such as adverse effects of the route alignment on a particular community can hinder that success of LRT (Mackett & Sutcliffe, 2003) The most important predictor of LRT r idership may be public willingness to use the transit system -that willingness is a predictor and an outcome since the factors reciprocate ridership, and higher ridership paints a more positive picture of the transit system (Tay lor & Fink, 2002)
66 populace can of a transit system -a rapidly growing area can still have a virtually non existent transit system. Auto centered cities such as Miami, Houst on and Atlanta may provide special c hallenges for transit providers (Mackett & Sutcliffe, 2003) These cities are k nown for their auto centered nature and traffic congestion (INRIX, 2006) The Mack centered nature an attitude that lends itself toward a different collective public attitude toward transit, and perhaps living in close proximity to a transit station (Mackett & Sutcliffe, 2003) While it is difficult to quantify the public attitude and transit ridership correlation, there have been studies on the characteristics of rail riders. A recent study of TOD dwellers found that rail riders are more physically active than non riders (Brown & Weiner, 2009) The same study also revealed that rail riders walk more, use their cars less, and have a lower prevale nce of obesity than non riders (Brown & Weiner, 2009) There is a considera ble amount of evidence supporting a theme of transit reinforcing transit patronage as more to using transit the improved attitudes may spur expansion and system growth. Conclusion The previous section highlighted characteristics that attrac t LRT ridership: multimodal integration, intensity of land use, density of residential population, parking and auto dependence, enablers they do not alone gu arantee successful LRT -they tend to be present in areas with high transit use. If we examine our larger cities with mature, well utilized transit systems one likely finds a combination of these factors working together. Of course there is a great deal of debate about the actual benefits of light rail, as well as debates about the cost efficiency of transit in general. The next section discusses literature that refutes the aims and benefits of LRT. The following section an evaluation of LRT criticisms will highlight the arguments against LRT
67 expansions. A discussion on how Portland exhibits many of the LRT ridership attraction factors will explain why Portland was an ideal choice for the study. C riticisms Made A bout LRT This section presents arguments refuting the claims that LRT proponents make about LRT. Advocates of LRT will argue that LRT is safe, quiet, promotes economic development and most importantly counteract the negative externalities o f auto use (congestion, sprawl, pollution, reliance on f ossil fuel). Critics of transit may use literature that equivocate transit expansions with negative externalities -higher crime and lower property values in order to d efeat plans at the ballot box. These attitudes may be illustrated in referendum votes that vote against rail expansions (Mackett & Sutcliffe, 2003) City residents may not see the value of transit because it does not achieve projected ridership, economic benefits or is too costly. Critics have countered automatically decrease congestion, reduce sprawl and increase density (O'Toole, 2004) Roads Like Transit, Are Not Free Some critics of LRT conten d that transit does not wo rk; a repeated criticism is t hat it does not pay for itself (O'Toole, 2004) While most transit systems do not make money, some would argue that roads do not pay for themselves either (Litman, 2008a) To counter, some would argue a majority of transit is subsidized from t axes just as other services like police, fire, and schools are. Given the thousands of public transportation agencies in the country the consensus may be transit is a needed servic e that the taxpayer may rely on directly or indirectly (American Public Transportation Association, 2008) Transit, and rail transit in particular are often criticized because they do not pay for themselves; it is true that these systems are heavily subsidized by taxpayer money. The fare does not cover the full cost, just as the gallon of gas cost does not cover the full cost of driving. Do
68 we ever ask the road to pay for itself? The road is just as subsidized as transit, perhaps more so when referring to expressways that do not collect tolls. Some critics advocate sacrificing multimodal accessibility for more highways (O'T oole, 2004) While many agree on the need for alternative modes of transport, Randal support s less public expenditure for transit and more money for highways offer s tend to present data on how driving a car is less costly per passenger mile than LRT or bus (Holahan, 2008) This line of argument would assume we pay the full cost of driving. The costs of road maintenance, time lost to traffic congestion, and the costs of pollution are not includ ed in the price per gallon -costs that are absorbed by the taxpayers and the natural environment. Do we ask schools or cou rthouses to pay for themselves? The roads we drive on, with the exception of tolls roads are likely subsidized more so than transit In 2006, the State of Texas was actively pursuing a massive new transportation initiative called the Trans Texas Corridor, which had toll highways as a major project component. As pa rt DOT examined the issue of whet for themselves. Their conclusion, shared in the newsletter Keep Texas Moving pays f or itself in gas taxes and fees (Texas Department of Transportation, 2006) The newsletter cited State Highway 99 will cost $1 billion to build and maintain over its 40 year life cycle, but only generate $162 million in gas tax revenue during this same timeframe (Texas Department of Transportation, 2006 ) According to these calculations, it would take almost 250 years for State Highway 99 to generate enough gas tax revenue to pay for itself. This article severely weakens a favorite argument against transit subsidies. Critics: LRT Does Not Decrease Congestion It is still not clear if LRT decreases congestion The studies show that light rail could actually increase congestion due to the delays that occur at LRT at grade intersections (Litman,
69 2008a) Chandler and Noel did indicate that the average additional delays from LRT crossings increase with increasing light rail crossing frequencies and increasing traffic volum s the road (Chandler & Noel, 2004) The Chandler article des c ribes how LRT can actually increase congestion. The Chandler and Noel study also states is also dependent on the degree of coordination and preemption of tra ffic signals within the network (Chandler & Noel, 2004) If traffic signals are set to al low no conflicting phases to (Chandler & Noel, 2004) These developments point to the possible benefits coordination between regional government and good transportation planning lending itself to an efficient LRT. Does light rail use decrease congestion? By encouraging people to relocate downtown, it would follow that more destination options not requiring auto use are available to those same people. One could also contend that increased LRT use should get people out of their vehicles. Does light rail actually get peopl e out of their cars? Dueker and Bianco (1999) indicated in a study comparing two corridors, bus and rail, that actually saw an increase toward more two car households and a loss of zero and one car households (Dueker & Bianco, 19 99) Both shifts were detrimental to transit ridership and LRT "was not reversing the trend of increased auto availability in the inner portion of the study area" (Dueker & Bianco, 1999) This study compared the bus/rail mode s within an inner portion of a corridor to bus/rail within the outer portion in terms of increasing zero and one vehicle households. The reasoning stands that an increase in zero car households naturally lends itself to less congestion. The study showed th at the bus
70 corridor showed an overall slight decrease in zero and one vehicle households whereas the rail corridor showed a slight increase (Dueker & Bianco, 1999) The study results were promising; it indicated that multiple v ehicle ownership had stabilized in the outer corridors (Dueker & Bianco, 1999) While not entirely refuting the criticisms of LRT not decreasing auto dependence, the r educing tendencies. Criticism: LRT does n ot Reduce Sprawl n or Increase Density Like the Knight and Trygg study, many planning e xperts agree that transit can influence the where and how development occur s A more relevant thought is what types of development is occurring around LRT stations. Outlying stations can have the inadvertent affect of actually increasing sprawl (Handy, 2005) Handy asserts that transit systems impact development in two ways, just as highways do : by reducing transportation costs and by ch anging relative accessibilities (Handy, 2005) First, if a transit system reduces travel times, it may enable residents to live farther out, thereby increasing rather than decreasing sprawl (Handy, 2005) This idea would have merit if it were easily illustrated. However, most new light rail systems are designed to serve areas of existing development and may hav e little impact on travel times (Handy, 2005) This effect can help to increase ridership and may serve as a catalyst for redevelopment in selected areas. Criticism : LRT Only Redistributes Growth, Not Create It In addition, by reducing transportation costs, a transit system might increase overall development in the region, leading to a net gain for the region (though probably at the expense of some other region) (Handy, 2005) This finding implies that LRT while decreasing transportation costs does not actually decrease de velopment, just redistribute it (Handy, 2005) This finding was echoed in a 1995 report from the TCRP concluded more typically redistribute growth that would
71 have taken place without the investment (Huang, 1996) This is a potentially potent charge because many transit advocates argue that the economic benefits are one of the main reason s to continue to expand systems. Redistributive t heory thus suggests that transit systems may have conflicting effects on development patterns, encouraging sprawl in some ways and acting as a counterforce to sprawl in others (Ha ndy, 2005) In determining the net effect of transit, it is difficult to separate out the effect of transit from the other forces influencing the amount and location of development in a region. Despite this challenge, the impacts on development of trans it systems -particularly rail rapid transit systems and light rail have been evaluated and summar ized by a number of researchers (Handy, 2005) Table 3 1 refers to major studies that occurred in the last few decades. Many of these systems rank among the more successful LRT systems and they might credit an increased regional vision in their long range transportation plans. Conclusion The best laid LRT expansion plan may not necessarily enhance an LRT ridership -land use, zonin g, crime prevention are all things that must be compatible. The cri tics are often eager to repudiate any perceived benefit from LRT, no matter how many studies highlight there being a positive correlation. Garrett ( 2004) suggests that policymakers should consider why economic development is not taking place in a certain area: The general consensus from the academic literature and the findings presented in this report is that light rail is not a catalyst for economic development, but rather light rail can help guide economic development. Rather than relying solely on light rail to create economic development, city planners and officials should first address a key question: Why is economic development not occurring in a given area in the first place? Possib le reasons include relatively high cost to business start ups, unattractive locations (crime, poor infrastructure) and unnecessary zoning and regulations. Unless these barriers are lowered or removed, the long run economic development objectives, with or without light rail will not be fully met. (Garrett, 2004)
72 As Garrett warns, each of these things must work in concert as LRT is not normally a big enough motor to spur development on its own. Cities that have recently put in an LRT tend to se slow ridership in their formative years (American Public Transportation Association, 2008) Critics of LRT have studies to enforce their claims that LRT does not reduce congestion, reduce sprawl or have any r eal effect on land use and/or land density. There are studies in the literature review that critics of LRT would cite as proving a lack of economic benefits that transit systems LRT is too costly and how roads are cheaper. What the critics of LRT fail to mention is that roads are just as if not more expensive and the associated costs (environment) may actually multiply costs in the future. and willingness to support transit has shown to be possibly just as important as density, physical and/or economic elements of a metropolitan area (Mackett & Sutcliffe, 2003) Still the economic valuation promise of LRT attrac ts new TOD plans, new construction and new opportunities for LRT expansion as evidenced by the studies of the last twenty years.
73 Table 3 1. Composite of studies: Effects of light rail p roximity to property v alues City region(s) transit system a uthor (s) Property value data source Access measurement Findings: effect of proximity to rail transit on property value or average home price Dallas, Texas (DART) Weinstein and Clower, 2002 Assessed value (Dallas County Central Appraisal District) Straight line distance to station, one quarter mile from station, compared with properties located in the control group Property value increased 32 percent near DART stations compared with 20 percent in control group areas not served by rail Portland, Oregon (Eastside MAX) Chen et al., 1998 Sales prices (RLIS Li te and Metroscan) Straight line distance to station one km radius Property value decreased $32.20 per meter further from station beginning at a distance of 100 meters from station Portland, Oregon (Eastside MAX) Lewis Workman and Brod, 1997 Assessed value (city property tax rolls) Network distance to station, one mile radius Property value increased $76 for every 100 feet closer (within a one half to one mile radius) to three stati ons that were studied Portland, Oregon (Eastside MAX) Al Mosaind et al., 1993 Sales prices Distance rings based on walking One km radius (Model 1) Network distance to station One half km radius (Model 2) Property values were $4.32 higher within the 500 meters than outside the 500 meter radius (Model 1) Property value decreased $2175 for every 100 meters further from station (Model 2) Sacramento, California (Sacramento Light Rail) Landis et al., 1995 Sales prices (TRW REDI data services) Network distance to station All property transactions in the city No statistically significant effect on home prices San Jose, California (San Jose Light Rail) Landis et al., 1995 Sales prices (TRW REDI data services Network distance to station All property tran sactions in the city Property value decreased $197 for every 100 meters closer to station Effect may be due to commercial and industrial uses Adapted from Hess & Almeida: Impact of Proximity to Light Rail Transit on Station Area Property Values in Buffalo, New York 2006.
74 Table 3 2 Transit ( fixed r oute ) service f reque ncy l evel of service m easurement s Source: Transit Capacity and Quality of Service Manual Second Edition Transportation Resear ch Board, Washington, DC, 2003. T able 3 3 Transit ( fixed r oute ) hours of s ervice l evel of service m easurement s Source: Transit Capacity and Quality of Service Manual Second E d ition Transportation Research Board, Washington, DC, 2003 Table 3 4. Operating cost per m ile by transit mode and c ost increase of selected transit modes in the United States, 2000 2007 Mode Operating cost per mile, 2000 Operating cost per mile, 2007 Percentage increase Light r ail $ 0.45 $ 0.60 33.3% Bus $ 0.59 $ 0.80 35.5% Heavy rail $ 0.28 $ 0.40 42.9% Commuter rail $ 0.29 $ 0.40 37.9% Source: American Pub lic Transportation Association. (2008). 2008 Fact Book
75 Figure 3 1. Percentage increase in mean sales price of existing single family r esidential units in U nited S tates by r egion, 1997 2007 Source: National Association of Realtors, 2008 Note: Dat a reflects U S and regional statistics for re sale of existing homes and excludes new construction. Not seasonally adjusted for inflation 4.00% 2.00% 0.00% 2.00% 4.00% 6.00% 8.00% 10.00% 12.00% 14.00% 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Percentage Increase of Mean Single Family Home Value Year Northeast Midwest South West U.S.
76 Table 3 5 Rail route mileage and s tatus of s elected f uture r ail p rojects in United S tates 2006 Mode Construction Design Open Planning Proposed Total Light r ail 190.2 58.8 706.3 511.3 300.0 1,766.8 Commuter r ail 126.7 118.1 4,265.7 969.3 885.6 6,365.4 Heavy r ail n/a n/a 1,303.8 93.8 102.6 1,500.2 2008 Public Transportation Fact Book Source: APTA Transit Infrastructure Database (a) Data as of September 2006. Table 3 6 Land value appreciation by land type near light rail stops in Dallas, Texas, 1997 2001 Residential Retail Office Industrial Inside mile buffer 32.1% 28.3% 24.7% 13.0% Outside mile buffer 19.5% 30.4% 11.5% 21.5% Source: Weinstein, B., and Clower, T. (2002) An Assessment of the DART LRT on Taxable Property Valuations and Transit Oriented Development
77 Figure 3 2. U.S Median Home Prices 1975 2009 (adjusted for inflation) Source: National Association of Realtors, 2009.
78 CHAPTER 4 METHODOLOGY The first chapter illustrated the benefits of LRT and why it is importan t, and the next chapter provided information on what LRT is and what it is not -as well as insight on factors that affect home prices. The third chapter reviews previous research on the land value transit access correlation. This fourth chapter will examine the why and the how of the data collection process. decisions to only examine SFR detached, only examine properties that have been improved for the whole of th e study years and to divide values by the square feet of the home. This is an examination of Portland, Oregon -a city considered a planning success story -using a combination of collection and examination methodologies. The objective was to examine the t en year period set of RMV appreciation for SFR in different scenarios. Five neighborhoods were carefully chosen to reflect different LRT scenarios whose results (home appreciation) could be compared. The scenar ios are existing LRT, new LRT, and no LRT comp ared in low income versus high income study areas. Why S tudy Portland? Urban planners have long known that a transportation system has a role in fostering growth and affecting the urban structure. Investments in improved transit are the subject of much d ebate among policymakers and planners, especially at the local level. Portland has an advantage over many other metro areas in this country because of the nature of its UGB. tains would be sprawl and may force would be employment centers to consider locating near a LRT stop more so than say an Atlanta or a Los Angeles, comparatively. airport. Many metro areas lack the extensive network that
79 to the benefit of a regional body making transportation land use decisions. metropolitan area numbers around two million in population -the 23 rd largest metropolitan area -but it ranks fourth in L RT use right among larger metropolitan areas such as Boston, Los Angeles and San Francisco (American Public Transportation Association, 2008) This is noteworthy because Los Angeles and San Francisco are significantly larger and have more mature transit system s These larger cities are part of a larger rail transit system that utilizes three or more forms of t ransit mentioned in this study -which is something Portland emulates. Portland is an ideal ci ty for LRT when examining the factors that stated earlier, some cities lack. Portland operates a downtown trolley and has links to commuter rail both which are integrated with TriMet (TriMet, 2008) These factors contribute to th integration. As of 2008, the most newly constructed form of rail transit in the country and when a mid sized city ranks fourth in LRT trips it makes any Portland a logical choice for an examination of LRT proximity and home appreciati on correlation (American Public Transportation Association, 2008) Portland is a relatively unique place for several reasons, all of them directly related to the ed regional government in this transportation planning that is enforced across county jurisdictions (Marshall, 2001) The local governments in Multnomah, Clackamas and Washington counties have turned over land use decisions to Metro; they decide what goes where (Marshall, 2001) They (Metro) answer directly to the governor of the state. Metro designed the light r ail that is essential to the region. A central element of any regional design must be the balance between them, the ease of connections, and the appropriate land use complement (Calthorpe & Fulton, 2001) Some policymakers thi nk that
80 can only be achieved when the regional government has the power to make the decisions affect all jurisdictions within the region. A county or an MPO can make as many proposals and plans as its wants but if those plans are not in coordination with n eighboring jurisdictions than its success will be hindered. Why E xamine S ingle Family R esidential (SFR) ? SFR was used in many previous studies due to the nature of residential land and accessibility. Location theory holds that home prices rise in synch with travel time savings, thus to the d egree transit expedites travel properties near stations should sell for more (Cervero, 1984) It is interesting to note that 2002 studies done by Ce rvero and Duncan found that light rail conferred the greatest land value benefits t o multi family residential and c ommercial properties. This is one justification for the research. influence has been demonstrated more clearly for residential uses than for commercial ones (Parsons Brinckerhoff, 1999) Landis noted three reasons for this problem: (1) a lack of comprehensive and reliable data; (2) a smaller zone of impact that limits the n umber of observations; and (3) while housing values are determined in the market place, the values of individual corporate transactions may represent only the value of one pair of buyers and sellers (Landis et. al., 1995). It has been theorized that proxim ity to rail lines could have negative effects on p roperty (Gatzlaff & Smith, 1993) ; a study by Chen did not find a ny statis ti cally significant nuisance ef fects (Chen, Rufolo, & Dueker, 1998) Diaz fi one of the more prominent ways that people understand the value of property is through the price or value of a home or in the rent they pay (Diaz, 2001) Another reason for examining SFR is that the number of real estate c onsumers (owners/renters) is greater than for other types of real estate (Diaz, 2001) All in all, the effects are more acutely felt in the residential sector (Diaz, 2001) Thus the sample size is greater and the market price should be more responsive to changes.
81 The method chosen is a dollar per square foot method that may present a clear er capitalization effect when comparing LRT scenarios. T hese studies are important t o show a positive connection/economic capitalization benefit correlation of being close to a LRT stop in residential areas. Dollar per Square Foot Justification In addition to evaluating percentage change in appreciation values over a ten year period, th is study uses a dollar per square foot method that attempts to combine all SFR square footage to obtain a base dollar per square foot. Conversely, analyzing appreciation rates may not account for the different size of tax lots. By combining the sum of sing le family home values and dividing them by the sum of square footage, perhaps a more accurate picture of land value before and after can be obtained. The dollar per square 2001 study to analyze the capitalizati on effects in of LRT proximity in Santa Clara County (Cervero & Duncan, 2001) Property values are determined, appraised and assessed by various public agencies, namely the city/county property appraiser. Appraisers are licens ed professionals who consider many different factors when evaluating a home; they are important to land value studies. Assessed, or taxable, values were used in the studies we examined. Study Design The data table looked like: Address ID number Land Use Ra il Line Square Foot Case Area/ Name 1997 RMV 1998 RMV 123 Main Street 11223344 SFR Blue Line 5000 Main Street Station 56000 57500
82 the years in question. For example, in 1998, the total Real market Value of a ll SFR parcels within a mile buffer of LRT case area C was valued at Y. This Y value is the composite 1998 SFR value of case area C. To be able to compare this number more easily, this potentially large number is divided by the total amount of sq uare foo tage of case study C. This number equals the dollar per square foot value for stop C that year, value K. Y = $15,000,000 (Total RMV of SFR parcels within case area C for year in question) F = 1,000,000 (Total Square Footage within case area C for year in question) $15,000,000/ 1,000,000 = $ 15 dollars (Value per square foot of total SFR parcels for year at stop C) K = $15 This computation was made for every year 1997 and after, until 2007. The preceding per square foot value for each is area was then subtracted from the current year to determined yearly valuation change. These computations were done for every year in question. For example if L is the 1998 per square foot value of all SFR parcels for stop C, then subtracting K from L wou ld equal the per square foot valuation increase between 1997 and 1998 for stop C. This resulting number would then be divided into K to give the appreciation percentage rates found in the results. For example say L= $16.50: ($16.50 $15.00) = $1.50 = P er square foot Valuation increase ($1.50 / $15.00) = 10% appreciation rate for stop C between years 1997 and 1998. The ten percent would represent the aggregate valuation rate of increase all SFR parcels in a mile buffer area near stop C. This number could then per compared to different stops as well as different time periods. Each year presented a different rate of appreciation. The fluctuating appreciation rates form the statistical basis for the findings later in the study. Again,
83 this study diffe rs from other in that it examines the singe family home appreciation value in different scenarios to see whether a long term affect on property values and light rail proximity have some interdependent relationship Many studies in this field of planning, as mentioned before, use hedonic and regression models to examine a distance/value relationship inside (or outside) a mile buffer. Some studies such as Al Mosaind (1988) examine property values further out as far as two miles away from a LRT station (Al Mosaind, Dueker, & Strathman, 1993) This study examines the difference in appreciation value percentage in different transit scenarios to see whether appreciation rates are indeed greater in areas within mile of a LRT stat ion. The mile buffer (a five minute walk) used in this study is set in preceding studies as the greatest distance people are comfortable/willing to walk to access LRT (Kim, 2008) A number of the studies mentioned use cross sectional design with hedonic modeling and they examined various independent (explanatory) variables such as number of rooms and square feet and others to estimate probable property values (Kim 3). Cross sectional research is a common way of examining how explanatory variables affect dependent variables (Kim 3). Cross studies which began in the latter part of the 20 th century, have tended to use models to predict land value impact s given all things being equal (Kim, 2008) Howe ver, the most elaborate hedonic model cannot account for all factors (Kim, 2008) The literature does support the general contention that having access to light rail does improve residential values, sometimes improves commercia l values and the effect upon industrial is relatively unstudied and unclear. Case Studies/ LRT Scenarios limitations on yearly home values. Certain stops were analyzed in th is study based on 2007 real market home values in the surrounding area. Case study areas were chosen, based on a contrast between income levels as well as the amount of SFR parcels compared to commercial, industrial
84 or civic use parcels. The five case stud y areas are broken down by income level: Hollywood/42 nd Street which is a high income area with existing LRT, Martin Luther King Boulevard and Killingsworth Street which is also a high income area but lacks LRT. Three study areas are defined as low income; Polk and Smith Streets lack transit, the Parkrose/Sumner area has new transit, and East 172 nd umner is a unique case to examine the before and after effect of LRT construction, something that has began to be examined. To study the effect of the proximity of LRT stops on single family home value appreciation in Portland the RMV listed on the w ebsit e www.portlandmaps.com were used. T his ment, and distributed by RLIS, the regional land information authority. This information is free, although GIS data has a nominal c harge The following variables were collected from the website, and were recorded in accompanying data tables. Another important factor in choosing Portland is that their data is relatively accessible compared to other metros whose property value records are not as easily accessible or detailed. With the RLIS system, the different property parcel information is open for anyone to view. This study differs from others in that it examines the single family home appreciation value in different scenarios to see whether a long term affect on property values and light rail proximity have some interdependent relationship. Many studies in this field of planning, as mentioned before use hedonic and regression models to examine a distance/value relationship inside a mile buffer. Some studies su ch as Al Mosaind 1988 study and others examine property values further out as far as two miles away from a li ght rail station (Al Mosaind, Dueker, & Strathman, 1993) This study examines the difference in appreciation value
85 pe rcentage in different transit scenarios to see whether appreciation rates are indeed greater in areas within mile of a light rail station. The mile buffer ( a five minute walk) used in this study is set in preceding studies as the greatest distance peop le are comfortable/willing to wal k to access LRT (Kim, 2008) Data Elements To stu dy the effect of the proximity of LRT stops on single family home value appreciation in Portland, the RMV listed on the website www.portlandmaps.com were used. This Information System (RLIS), the regional land information authority. This information is free, although obtaining G eographica l I nformation S ystem (GIS) data requires a payment of a nominal fee. The variables in the study design were collected from the website, and were recorded in accompanying data tables. Another important factor in choosing Portland is that their data is rel atively accessible compared to other metropolitan areas whose property value records are not as easily accessible or detailed. With the RLIS system, the different property parcel information is open for anyone to view for free. The next step in the method ology is to identify all light rail stops in Portland that qualified for the parameters of the study. The exact location of the quarter mile buffer (actual distance) was determined with GIS. GIS was used to join data values to a spatial location. This ill ustrated what parcels were mile away (actual distance) as well as which parcels were zoned commercial, industrial or public utility Utilizing parcel GIS helped determined the buffer zones as they applied to the mile actual distance from LRT stations. Unlike some of the previous studies, no dollar amounts in this study were adjusted for inflation T he dollar per square foot method was used as a way to easily compare the different LRT scenarios and as a way to compensate for dif ferent sized parcels Per square foot also
86 captures the particular density of settlement in the case studies; larger parcels would not be at the same advantage had just property values been examined. The appreciation percentage was then lar per square foot value. GIS was utilized to map all parcels located on/or within a quarter mile (actual distance) of a light rail station. The actual distance was determined via GIS. After parcels are identified via their property appraisal parcel num ber, real market values were recorded for ten years (1997 2007). A distincti ve aspect of the study is the comparison of SFR appreciation values in low income and high income areas per LRT scenario. How does one define a rich or poor area? This was done bec ause there were limitations in the economic analysis provided by the U.S. Census based on zip codes -given that stops tended to have more than one zip code contained in a mile distance in any given direction The Census was used to determine generally wh ere lower income areas were, however home values tended to not be consistent with this method. It was thought that the areas chosen would be determined based on home values, per RLIS data. ar an LRT stop built before 2007, the time period of the mile buffer of an LRT stop. Data Exclusions Which stops are to be chose n? The parameters of our study require a comparison of low income and high income neighborhoods. The stops adjacent to LRT were chosen particularly if (after analyzing the land use via GIS) there were a large amount of residential areas. Generally, stops with the more residential land use s with in the mile buffers were more likely to be chosen, given all other things being equal. Many stops had civic buildings, hospitals, schools, parks and/or shopping centers adjacent to the stops. These areas around st ops were avoided, due to lack of sample size these places provided and the lack of clear data. It should be noted that no
87 neighborhoods were used that were adjacent to the Yellow Line mainly because of time data limitations ; it was finished in 2004. Since the data was available from 1997 to 2007, and the study ex amines SFR detached values before and after a LRT stop is built, certain data was excluded. Only 1997 2007 data was readily accessible, no data before or after that time period was used. Also, only single family residential parcels that have values for all the years between 1997 and 2007 were included. The following values were excluded -any parcel housing that may have been built after 1997, and a ny parcel that does not have a ten year set of value s. This was done to eliminate data errors; new properties built since 1997 will be excluded. Attached housing was excluded due to accompanying fluctuation in square feet. Certain parcels, for various reasons, had no data on the website. Multifamily housing was excluded from this study because the lack of accurate living unit data. All commercial and i ndustrial parcels were excluded due to data limitations. Also, as data was being collected there were parcels that had huge variations that could not be includ ed there were parcels that appreciated or depreciated by hundreds of thousands of dollars in consecutive years. These parcels could have been unimproved, and then improved (or vice versa) during the span of years in question. Regardless, these parcels were excluded. Case Studies/ LRT Scenarios limitations on yearly home values. Certain stops were analyzed in this study based on 2007 real market home values in the surrounding area. Case study areas were chosen, based on a contrast between income levels as well as the number of SFR parcels compared to commercial, industrial or civic use parcels. The five case study areas are broken down by income level: Hollywood/42 nd Street wh ich is a high income area with existing LRT, Martin Luther King Boulevard and Killingsworth Street which is also a high income area but lacks LRT. Three study areas are
88 defined as low income; Polk and Smith Streets lack transit, the Parkrose/Sumner area ha s new transit, and East 172 nd unique case to examine the before and after effect of LRT construction, something that has began to be examined. Conclusion To review, the data three major factors in selecting the city for examination. Using SFR, while in previous studies may not have reflected the greatest conferred economic valuation benefits, was the easiest to analyze. The ten years of property values were then divided by the total number of square foot in each study area to give the whole area a composite value. This value could then be examined over time to ascertain valuation benefits more efficiently. Util izing parcel GIS helped determined the buffer zones as they applied to the mile actual distance from LRT stations.
89 CHAPTER 5 CASE STUDIES/RESULTS system, referred to as the MAX, began service in 1996. This segment served primarily the Por tland city core and east along U.S Highway 30 As the service grew, the west side of the line was completed in 1998, stretching all the way out to Hillsbor o. As shown in Table 5 1, the Red Line travels from downtown to the Portland International Airport. I n 2004, the Yellow Line was completed, runn ing from downtown north along Interstate 5. As shown in Table 5 2 evening to serve most commuters that work different hours; as well a s the 9am 5pm commuters. To enhance any study of LRT stations and their surrounding property values, the characteristics of the LRT lines and stations are shown in Table 5 1. The Portland MSA has a population of over 2.2 million people between seven count ies in northwestern Oregon and extreme southwestern Washington as shown in Figure XX (U.S Census Bureau, 2008) The 23 rd largest MSA in the nation, Portland has the only directly elected M etropolitan P lanning O rganization (MPO ) in the United States (Calthorpe & Fulton, 2001) Metro is responsible for land use planning and maintaining the UGB, managing TriMet which includes the MAX, some city parks, cemeteries, landfills, conv ention centers, city zoo and the (Calthorpe & Fulton, 2001) promoting mixed scale replications of the CBD. Th e political structure of Metro may help keep a regional balance with regards to development in various areas of the city. Other areas may promote growth in the suburbs at the expense of the city core. Other areas could be immobilized by county vs. county in fighting with regards to land use decisions these are precisely the issues Metro was set up to eliminate.
90 Case Studies Nevertheless, Portland still has very different neighborhoods, with some rich and poor areas in the suburbs and core areas. The following chapter will give a snapshot of each neighborhood. The following five case studies are composed of three areas that within a quarter mile of a LRT station and two areas that are not. At the same time, three of the five areas are low income, w hile the other two are classified as high income. The two areas that are classified as low income in thi s study are by no means poverty level; they just represent areas that are ian household income of $48,2 88 (U.S Department of Housing and Urban Development, 2002) Hollywood and 4 2 nd Street High Income Area with Existing LRT This area is 2.52 miles distance from the CBD, the closest out of all study areas. The Hollywood/42 nd Street area is an older section of to wn that lies directly south of major retail destinations on the corner of Sandy Boulevard and Broadway. The 1999 median family income of this area was $59,288, significantly higher than the regional median (U.S Census Bureau, 2008) Table 5 3 shows that Hollywood/ 42 nd has the second oldest housing stock (56.9 percent of housing built before 1939) and a smaller percentage of SFR detached homes (U.S Census Bureau, 2008) This indicates a possibility that attached housing and MFR have developed over the years, indicating a den sity and intensity of land uses Figure 5 5 illustrates some aspects of the urban form of this area. The study area is cut in half by Inte rstate 84 as well as the MAX line. The northern half of this study area features moderate retail and office land with busy arterial intersections. The southern half of the study area is referred to as Laurelhurst neighborhood, and is separated from the r etail by Interstate 84 and the MAX line. The SFR parcels are uniform typically right around 5,000 square feet. As
91 Figure 5 5 shows there is hardly any non residential land use in the southern half of this study area. This area also has twenty percent of its residents using transit (Table 5 3), which ranks highest of case studies As Table 5 3 shows two LRT lines serve the Hollywood/42nd area, though at the station they share the same track so residents enjoy more frequent service as well as access to mor e destinations than any other study area. The MAX station is also designated city Transit Center which can be defined as having stops/connections of more than one transit mode. As Figure 5 5 shows, bus transit serves this area quite well. As Table 5 2 indicates, trains stop very frequently and as shown in Figure 5 full accessibility to destinations. East 172 nd Street and Burnside Avenue, Low Income Area with Existing LRT Another station located on the MA X Blue Line, t his station is 8.96 miles from the CBD, significantly farther away than any other case study area. The 1999 median family income of this area was $38,758 -significantly lower than the regional median (U.S Census Bu reau, 2008) This area is characterized by a less connecting pattern of streets, larger lots and newer housing only 3.3 percent of the housing was built before 1939 (Table 5 3). As Figure 5 6 shows, t here is some major retail on the major streets Burnsi de Avenue and Stark Avenue to the south. This area has many irregular sized lots and lacks a strict grid pattern of streets. This area lies on the border of east Portland and the suburb of Gresham. Burnside is a major corridor but it is only a two lane r oad that is lined with trees and SFR with some MFR that are oriented clearly towards the street. The track is in the median that divides the road. There is not a large commercial to speak of this is a very middle class residential area. The land use and i ntensity are not what they are, say, closer into town. There are no bike routes, but there is bus service on Burnside and Stark Avenues -two east west arterials. As shown in Table 5 3, the income and home values are
92 second lowest in the study, but 12.2 pe rcent of the residents report using transit regularly (Table 5 3). More than half of the houses are SFR detached T his area is served by the Blue Line, albeit the Blue Line extension, that began operations in 1998. This area does not have large amounts o f commercial and maintains a rustic feel Martin Luther King Boulevard and Killingsworth St reet, High Income Area with No LRT This neighborhood is filled with some inconsistencies. There are pockets of slightly dilapidated SFR units with stately old two story houses mixed in. The area also has a tight grid like network of streets as well as uniform lot sizes. Street to street, there are different sizes of homes and presumably home values. This area in consistent in one aspect these residential units are 1939. As to the median family income of $70,629 a year -e asily the highest in the study -one could not tell from pictures (U.S Census Bu reau, 2008) This is an area in flux, with a few new condominium type apartments that are obviously new mixed with some homes that are not maintained. The overwhelmingly majority of the housing is SFR detached. The main thoroughfare is Martin Luther K ing Boulevard this street is lined with older commercial units that appear to be underutilized. East of MLK is a predominantly African American neighborhood, although with a city that is nearly eighty percent white, that statement does not hold the same effect as it would if referring to a Sunbelt or east coast city. This case study area is 2.72 miles from the CBD. Not served by LRT, and the nearest light rail line is the Yellow Line (opened in 2004) which is two miles away. This neighborhood lies outside the walking distance to LRT. Bus transit serves MLK Boulevard and Killingsworth and Alberta Streets. As shown in Table 5 3, this area ranked second in the study with 12.3 percent of the residents using transit regularly Although not served by LRT, thes e residents utilize bus and are relatively close to downtown. While the housing stock is not overly impressive this area appreciates anyway
93 which may point to the importance to being close to the CBD. With regards to property values and appreciation this area may benefit economically through the age/size/ desirability of its housing stock. Parkrose/ Sumner, Low I ncome Area with New LRT As Table 5 3 indicates, even with its new station, the residents of this area utilized transit less than the other area s. As Figure 5 4 will also indicate the study sample may be influenced by the interchange carving the studied parcels in half. This station is 5.63 miles from the CBD and is served by the Red Line, which op ened in 2001. Parkrose/ Sumner refers to the na me for this station and the two neighborhoods Parkrose (to the east) and Sumner (to the west) that make up the area. As Figure 5 4 illustrates t his ar ea is characterized by a major highway interchange ( Interstate 205 and Columbia Boulevard ) as well as an adjacent intersection of Sandy and Columbia Boulevards. This further highlights the high percentage of transportation infrastructure that exists in the study area. The LRT station itself is located in the median of Interstate 205 (Portland Freeway). To a ccess the station riders use a flyover which is connected to a Park & Ride lot on the eastern side of the I 205 interchange on Sandy Boulevard. Th e Park & Ride lot is also served by buses that run along the mentioned arterials going to different parts of the city As shown in Table 5 3, the median family income is $46,580 which is close to the median regional income of $48,288 (U.S Census Bureau, 2008) The housing stock is newer with 11.4 percent of the stock built before 1939, but seemingly smaller in size than the other case study areas. Nearly seventy percent of the residential units are SFR detached. Although there are plenty of bike lanes and major transit amenities, only 8.4 percent of study area residents use trans it regularly.
94 North of the area are frei ght rail yards and other medium industrial land uses that border the southern extent of the airport. This study area contains some air travel type land uses, including major hotels directly east of the I 205 interch ange. This area has some unimproved roads and some unimproved lots directly bordering the station area to the west. Noticeable are the lack of sidewalks on the residential side streets. The improved roads themselves lack drainage and shoulders. It appe ars that the interstate construction disabled what was once a tight network of streets, evidenced by the continuation of street names on either side of the interchange Polk Avenue and Smith Street, Low Income Area with No LRT This station is 5.92 miles from the CBD This area has no access to LRT, although it is served by bus. North and east of downtown Portland; this area is somewhat isolated geographically from the rest of the city. The Polk and Smith case study area reported a median family income $38 ,452 even though nearly thirty percent of the homes were constructed before 1939 (U.S Census Bureau, 2008) The median income of this area was lowest of all study areas and nearly ten thousand dollars less than the regional me dian. Referring to the parcel map of the area, (Figure 5 2) this area is dominated by residential parcels, some 68 percent of them SFR detached (U.S Census Bureau, 2008) As shown in Figure 5 2, Polk and Smith have uniform lot sizes, a strict grid pattern of streets and parks and schools are nearby. The neighborhood is typical middle class, much of the housing units are SFR detached. The area also features bus service, prominent bike lanes and sidewalks There is a commercial corridor to the south of the study area on Lombard Street. Bus service runs along Fessenden Street, which is in the extreme north of the study area. Sidewalks are aplenty; bike lanes run through the study area along Central, Smith, Fessenden and Buchanan Streets. 9.2 percent of residents reported using transit (U.S Census Bureau, 2008) There are
95 more parks nearby in this area than the other areas, which may be indicative of the physical landscape that surrounds this area. S eemingly this area suffers from its relative isolation from downtown and its amenities to exhibit strong land values/ land appreciation. Hollywood/42 nd is a high income area that has had rail transit service since the 1980s. Close to downtown but separated from commercial and industrial uses, this area income wise represents upper middle class to middle class America. Our high income, with no transit area, MLK/Killingswo rth is located on a major arterial in an older section of town that is relatively close to the CBD. Overall, the income level suggests that the area is at least middle class. The area lacks LRT service in a mile radius but there is LRT a mile and a half away. This is an area in transition, possibly pushing out long term residents and/or gentrifying due to relative proximity to the CBD. Polk and Smith is farther away from the CBD, lacks LRT, and is lower income when compared to the other study areas. The area lacks even a moderate density of uses. It represents the typical middle class neighborhood in the older, outer suburb old enough to have a grid pattern of street network but not older enough to capture the capitalization effects of older housing stoc k. East 172 nd Street is also like Polk and Smith in that it is far away from downtown, but one difference the area has been served by LRT since 1998. East 172 nd probably echoes suburban life for many Americans lack of prefect grid patterns in the street network, a commuter suburb. Parkrose and Sumner is a unique case study area because it has a major interchange in the middle of it. The area is low density and a uto centered in nature, with industrial uses nearby.
96 Results As Table 5 3 indicates t he study areas appreciated rather inconsistently in the year before 2005. 1998 was a particular ly slow year, with two areas showing negligible value growth. One stop, ML K/Killingsworth, grew 7.4 percent in 1998 that was greater than the 1998 regional home appreciation rate of 7.2 -the only study area better than the median (U.S Census Bureau, 2008) Table 5 3 shows that i n 1999, the MLK/Kill ingsworth stop appreciated the fastest out of all study areas All areas in 1999 exhibited a higher appreciation rate than the previous year. The regional home appreciation rate slowed in growth (U.S Census Bureau, 2008) All areas experienced a d ecline in appreciation rates from 1999 to 2000, which may have reflected regional or national economic trends. As the table indicates, t he regional growth in home appreciation percentage rose 7.2 percent in 1998, 4.5 in 1999, and a s low 2.1 percent in 2000 (U.S Census Bureau, 2008) In 20 01, four of the five case areas appreciated by a larger percentage than the year before. In 2002, the growth continued for the study areas. The Hollywood/42 nd Street rose 7.9 percent more than it did it 2001, MLK/Killingsworth grew 6.2 percent more, Parkrose/Sumner 4.9 percent more, Polk and Smith 1.8 more, and East 172 nd 0.7 more. In 2003 a differe nce of 4.3 percent. The regional average finally topped four percent in 2003, which foreshadowed the boom a few years later. As Table 5 5 shows, Polk and Smith maintained a steady growth rate right around seven percent In 2004, Hollywood, MLK, Parkrose grew significantly less than they did in 2003. Two low income areas, 172 nd and Polk and Smith appreciated more in 2004 than in 2003 East 172 nd actually grew by more than five percent the highest increase for that study area in the years examined. In 200 3, Parkrose/Sumner's appreciation rate spiked to 13.7 percent from the previous year. The Hollywood/42 nd st ation still appreciated at a 9.6 percent
97 rate; its second best appreciation rate for this area up to that point. From 2003 to 2004 MLK/Killingswort h appreciated 18.6 percent -s econd only to Parkrose/Sumner 2004 could be considered the year before the housing market exploded. By 2005, the boom was underway. Regionally, the average Portland home appreciated 9.1 percent up from 4.6 in 2004. In 200 5 all areas appreciated in the study -Hollywood/42nd St. appreciated 4.2% more in 2005; MLK/Killingsworth 6.8 % more, Parkrose/Sumner 9.4 %, and Polk/Smith 2.7% and 172 nd St. 2.4%. These last two areas grew much slower then when the entire area was boomin g, this may point to the distance from the CBD. 2006 offered much of the same; with all case study with no LRT (Polk/ Smith) and with existing LRT (172 nd ) showed t he highest appreciation growth yet. The two areas closest to the CBD showed the most robust growth: this study shows a boom in appreciation rates in the year 2005 and after. All case study areas experienced nearly double digit growth rates or more in the y ears between 2005 and 2007. It can be said that the growth that the areas experienced were more related to the overall land valuation of the region that LRT station access. The literature review mentions that the implementation of new station could lead to higher than normal appreciation or speculative rate in the years immediately after. It was hardly the case in this study. The Parkrose/Sumner area compared to the other areas appreciated the second lowe st in both 2001 and 1998; in 2000, the year prior to implementation the area appreciated less than one percent. The results are intended to show what happened in the market in these very different areas of the same metropolitan area. Two tables are used i n the following discussion the first table examines yearly appreciation rates. The second table looks at the dollar per square foot
98 does not take into accoun t how far the area has come overall. The first table seeks to only compare appreciation rates year to year. The results show that having an existing LRT ultimately does not guarantee an accelerated home appreciation. The results will also show that distanc e to However, a new LRT stop does seem to trigger an accelerated home value growth, but in this study, that acceleration was inconsistent. A few different stories emerge -the general trend of Portland in these ten years was a slow appreciation market that really slowed down around 2000. By 2003, the Portland market had accelerated and by 2005, a boom in appreciation was well underway, as was much of the nation.
99 Tab le 5 1 Characteristics for Portland light rail transit l ines, 2008 MAX Line Blue Line Red Line Yellow Line Year opened 1986 2001 2004 Areas served Hillsboro Beaverton Downtown Gresham Beaverton Downtown Airport Downtown Portland Expo Center Stations 46 25 21 Station(s) Shared 22 22 12 Track Mileage 33 25.5 7.2 Transit Centers 9 6 1 Connections to other rail modes 2; S treetcar, C ommuter rail 1; S treetcar 1; S treetcar Source: TriMet System Information Map, 2008 Table 5 2 C ase s tudy t ransit s tops and l evels of s ervice Case study a rea Frequency/ L evel of service score Hours of s ervice/L evel of service sc ore Served by which l ine? Parkrose/Sumner 5 4:30am Red E 172 nd Avenue 5 4:30am Blue Polk Avenue and Smith Street n/a n/a n/a MLK/Killingsworth* n/a n/a n/a Hollywood/42 nd St. 5 4:30am Blue/Red Source: TriMet System Information Map, 2008
100 Table 5 3 Light rail transit scenario/case study area c haracteristics Case Study Type of LRT scenari o Median Family Income (1999 dollars) Median Home Value (1999 dollars) Percent of Housing Stock built before 1939 Percent using Public Transit (includin g taxicab) Distanc e from CBD (miles) Serve d by line(s) Percent of housing units SFR detache d Hollywood @ 42nd Zip Code: 97232 High Income Area with Existing LRT Transit $59,688 $233,40 0 56.9 20 2.52 Blue, Red 30.8 East 172nd Street Zip Code: 97233 Low Income Area with Existing LRT Transit $38,758 $136,20 0 3.3 12.2 8.96 Blue 51.6 MLK Blvd @ Killingsworth Zip Code: 97212 High Income Area with No LRT Transit $70,629 $232,90 0 73.9 12.3 2.72 None 72.1 Parkrose/ Sumner Zip Code: 97220 Low Income Area with New LRT Transit $46,580 $135,20 0 11.4 8.4 5.63 Red 69.4 Polk Ave. @ Smith St. Zip Code: 97203 Low Income Area with No LRT Transit $38,452 $121,20 0 29.6 9.2 5.92 None 68 Source: U.S Census B ureau, American Community Survey 2006 2007
101 Figure 5 1. Regional m ap of Portland Oregon Source: Regional Land Information System GIS.
102 Figure 5 2. Parcel Map of Smith Street and Polk Avenue Source: Regional Land Information System GIS.
103 Source: Regional Land Information System GIS. Figure 5 3. Parcel Map of Killingsworth Street /MLK Bo ulevard. So u r ce: Regional Land Information System GIS
104 Figure 5 4. Parcel Map of Parkrose/Sumner Area Source: Regional Land Information System GIS.
105 Figure 5 5. Parcel Map of Hollywood/42 nd Street Source: Regional Land Information System GIS
106 Figure 5 6. Parcel Map of East 172 nd Ave Source: Regional Land Information System GIS
107 Figure 5 7 Results, S ingle family residential parcels v a lue in dollars per square foot, per light rail transit s cenario in Portland, Oregon, 1997 2007
108 Figure 5 8. S ingle family residential appreciation p ercentage rate per square f oot in Portland, Oregon per findings Table 5 4 Dollar per square foot comparison, 1997 and 2007 per findings Scenario 1997 2007 T otal Appreciation (not adjusted for inflati on) Hollywood and 42 nd : High Income/Existing LRT $31.58 $75.54 239% Polk and Smith: Low Income/No LRT $18.10 $36.58 202% E 172 nd : Low Income Existing LRT $13.42 $24.47 182% MLK/Killingsworth: High Income/No LRT $19.20 $58.24 303% Parkrose/Sumner: Low Income/New LRT $14.15 $31.06 220% 0.0 5.0 10.0 15.0 20.0 25.0 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Hollywood/42nd High Income with Existing LRT MLK/ Killingsworth High Income with No LRT 172nd Low Income with Existing LRT Parkrose/Sumner Low Income with New LRT Polk and Smith Low Income with No LRT Average Metro Portland Home Appreciation
109 Table 5 5 SFR appreciation p ercentage rate per square f oot in Portland, Oregon, p er f indings Case Area/ Scenario 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Per Sq. Ft. Total Increase 1998 2007 Hollywood/42 nd : High Income with e xisting LRT 5.2% 6.9 4.0 6.0 13.9 9.6 8.3 12.5 12.1 13.3 239% MLK/ Killingsworth: High Income with n o LRT 7.4 11.2 4.3 5.6 11.8 18.6 9.1 15.9 16.9 17.5 303% East 172 nd Street: Low Income with e xisting LRT 3.1 5.9 3.5 1.5 2.2 2.2 7.4 9.8 12.1 15.1 182% Park rose/Sumner: Low Income with n ew LRT 0.3 6.8 0.9 2.3 7.2 20.9 3.5 12.9 10.9 18.2 220% Polk and Smith: Low Income with n o LRT 0.1 9.9 2.3 5.6 7.4 6.7 8.0 10.7 11.8 11.1 202% Average metro Portland home a ppreciation 7.2 4.5 2.1 3.6 2.3 4.0 4.6 9.1 15.0 14.4 n/a
110 CHAPTER 6 DISCUSSION Analyzing five study areas reveal ed very different developments for each case study. This section will interpret those findings and relate them to the literature review. Based on the literature review, it was expected that the areas near LRT would appreciate more or at least equally with areas that lacked LRT access. The literature review was not consistently illustrating the higher values conferred upon residential land. There were studies reviewed that indicated that proximity to LRT could have a negative effect on property value. However it was found that there were several explanations possible for why a LRT scenario area did or did no t appreciate. The individual station areas will be reviewed more in depth in the next section. Hollywood/42 nd Street: Many Factors Add to Promising R esults The Hollywood/42 nd station study area appreciated the second most out of all study areas. It ben efits from many of the bundle of factors mentioned in the literature review; t he area is closest to CBD, it has a more frequent LRT service, is closer to major retail, lacks negative externalities and has high transit ridership. Also, the area is directly l ocated on a major interstate highway. In terms of the factors influencing ridership, Hollywood/42 nd station area in a high income area, has multimodal conn ections, a moderate density of land uses and a high density of surrounding population around the s tation. Only thirty percent of the SFR housing was detached which indicates a presence of multi family uses (U.S Census Bureau, 2008) Accordingly, the station area has pedestrian amenities including sidewalks and a pedestria n crossover Interstate 84 to access the station. As demonstrated in the chapter previous, in terms of dollar per square foot the area was worth significantly more than the rest in 1997. Whereas MLK/ Killingsworth, the faster appreciator is still in the s tate of flux, Hollywood/42 nd St. already had a sizeable head start in terms of dollars per square feet value.
111 theories Hollywood/42 nd would probably appreciate reasonably well with or without LRT. The rent gradient from the Portland CBD to the periphery seemed to follow suit As shown in results, the study areas were valued less the further away from downtown Portland. There are other reasons that might explain this as well -t he frequency of se rvice cannot be understated, the schools in the area, as well as highway access. Hollywood/42 nd may be more desirable because it is more accessible by more ways to more destinations. If MAX never existed, this area would still be located on Interstate 84. The added mobility of being able to walk safely to another form of transit that duplicates a limited access highway into downtown may just be worth paying that extra dollar for. People buying a home in this area may spend less on transportation as part o f their household budget which may leave more money for housing. The next stop area Parkrose/Sumner does not benefit from the same location advantages that Hollywood/42 nd does: Hollywood is a mature, walkable, close to downtown, high income neighborhood easily accessed through the MAX lines. As shown in the results Hollywood/42 nd never had a year when it did not appreciate by at least four percent. The property appreciation in this study area was consistent with the literature review. The idea that thi s area benefitted from highway access must be considered as well. Parkrose/Sumner : New Station Yields Uneven R esults The most important aspect of Parkrose/Sumner study area was the implementation of a MAX station in 2001. This area is fragmented -Parkr ose on one side of the interstate highway and the Sumner neighborhood on the other. Add to that fact that an interstate highway divides the neighborhood and this fragmentation frames any discussion on this study/area. Referring to the literature review, t he Hiawatha Line in Minneapolis was a primary example, of land value
112 effects of implementation of new stations. The main difference is that was whole new LRT system with seven new stations this is a new station as part of an existing line extension. The H iawatha study, after the station construction, showed that homes outside the half mile buffer appreciated significantly higher than those within. The jury is still out on Parkrose/Sumner -all in all, the new LRT station may have had a positive effect on t his study area, albeit a staggered amount of time after LRT station completion. For Parkrose/Sumner to post the highest yearly value SFR growth is likely not a coincidence. Time will tell whether the good showings of 2005 (second most growth) and 2007 (18 .2 percent appreciation highest of all study areas for that year) were simply aberrations. The Parkrose/Sumner area compared to the other areas appreciated the second lowest in both 2001 and 1998; in 2000, the year prior to implementation the area appr eciated less than one percent By 2001, all case study areas appreciated faster than in 2000, with the exception of East 172nd Street. Parkrose/Sumner, the only stop in the study served only by th is new Red Line exclusively did not fully experience a rap id growth until 2003. Parkrose/ Sumner home values spiked following the completion of an LRT station in 2001. This year marked the completion of the Red Line, which connected the system to Portland International A irport (PDX). It took time for the improve ment to register -in 2003 Parkrose/Sumner's appreciation rate spiked 21 percent from the previous year. The bigger picture of the study area is uneven at best. In 2003, the Parkrose/Sumner houses appreciated 20.9 percent, which may lend credence to the LRT effect on single family residential homes. However, it cannot be said for sure that LRT caused the value explosion, as in the very next year Parkrose/Sumner s tudy area homes only appreciated 3.5 percent, which was well below other areas in that same year. By 2005, the appreciation rates were more in line with other areas in the study as well as the average in metro
113 as a whole. The 18.2 percent appreciation ra te from 2006 to 2007 was the highest that year out of any area, which may mean that as a lower income area Parkrose/Sumner and its new LRT was increasing the home values nearby. These healthy numbers help offset the However, it is unclear. Parkrose/Sumn er may still be adapting to its new transit investment and still suffers from its distance to the CBD. not as close as MLK and Hollywood, but still closer to the CBD then Polk/Smith and East 172 nd T he Parkrose/ Sumner study area lack s commercial uses (restaurants, grocers) to support the few hotels in the immediate area may provide clues as to the newness of the station as well as the highway. The development in this cas e study area seemed to have resulted from the interstate highway access rather than transit. T his area lacks the diverse land uses and densi ty of areas closer to the CBD. Recalling the literature review examples, t his area is too close to industrial, rai l lines and perhaps, f our lane highways to be very desirable in a desirable city like Portland The Parkrose/Sumner area lacks pedestrian amenities like sidewalks, the density and intensity of population and land use is less than say, Hollywood/42 nd Stree t. Another difference is that the station being in the middle of a large limited access highway lacks connectivity and is harder to access for residents than other study areas which further lowers the walkability. It is worth noting that the highway acces s may Polk and Smith: No T ransit Not Much A ppreciation Parkrose/ Sumner and Polk and Smith grew very similarly before 2001 when neither area had transit. They are both roughly six miles from the CBD. The two areas have gone in different directions since. Parkrose/Sumner got a new station while Polk and Smi th ranked last in median family income, last in median home value even though the area has the third oldest housing stock (U.S Census Bureau, 2008) P erhaps the most compelling story Polk and Smith
114 tell is the overall health of the Portland CBD, that even as far away as six miles the area still out performed the regional average seven out of ten years. Overall, Polk Street and Smith, a lower income area with no access to LRT, appreciated the least vigorously out of any case stu dy area. Polk and Smith is farther away from the CBD, lacks LRT, and is lower income when compared to the other study areas. The area is mainly comprised of SFR with no commercial corridor and lack of other types of uses. Polk and Smith appreciated the le ast twice in 1998 and 2007. The 11.8 the lowest of all stops that year. When comparing value per square foot, Polk and Smith went from $18.10 a square foot to $3 6.58, which was better than the comparable area, East 172 nd Street, an area that has long been served by LRT. Also interesting to note that the $18.10 per square foot 1997 value was the highest among all areas that are low income r square foot value of $19.20 this may indicate that a lack of LRT may have hindered the possible SFR appreciation in Polk and Smith. Referring to the Parkrose/Sumner stop, which was the new transit investment in a low income area, really experienced a gr owth in appreciation rate after 2001 more so than was nearly 14% yearly; only the no LRT high income case had a higher one. Parkrose/Sumner had appreciated a t an average of 3.28 percent before the stop was implemented. The Polk and Smith neighborhood appears to be typical middle class, with much of the housing units are SFR detached. The area also features bus service, prominent bike lanes and sidewalks in other words, a good candidate for LRT expansion. However, the area is isolated from the CBD, lacks office and retail land, and that probab ly plays a large role in the fact that the area performed less than the other areas. This area is close to the Williamette River actually
115 enforcing the isolation, thus proving that being restrained physically can have an adverse effect on home prices. Th is may contradict aspects of the literature review with regards to physical CBD prevents the higher value capture. Also, this area lacks highway access. Per haps if there was an interstate highway linking this area to the Vancouver area the n the results would tell a different story. However, the most may be the waterfront industrial land uses that c ompletely enclose the community on the north and the west. Heavy freight rail also encircles the area and connects more industrial land uses in extreme southwestern Vancouver, Washington area. Ironically the value perception of the community may be adve rsely related to the presence of heavy industrial rail. Overall Polk and Smith are located too far from the CBD capitalization effects. If you lived here you likely had to have an automobile, and as only nine percent of r e sidents reported using transit (U.S Census Bureau, 2008) The following study area East 172 nd Street had the third highest percentage of people using transit (12.2%), on a stop that has been there since 1986 and still appreci ated twenty percent less than Polk an d Smith in the overall picture. East 172 nd Street: LRT Access Does Not Automatically Equate A ppreciation Looking at aerials, you would think this is a great example of a transit dependent neighborhood -there is great sidewalk connectivity, transit oriented development near the station as well as a large park nearby. The area has been served by LRT since 1986, and it would follow that the case study area would appreciate at least as well as areas that were not served b y transit. However, that was not the case -East 172 nd Street, which is located nine miles east of the CBD near the end of Portland proper and more into the city of Gresham, was valued significantly lower initially than the other case study areas T his a rea appreciated the fewest in the ten year
116 period four times, including three years in a row, 2001 2003. In 2002, all study areas i ncreased in appreciation -with the exception of the East 172nd Avenue station area. The biggest detriment to 172 nd alue seems to be the lack intensity and density of land uses, and is too far away from the CB D to capture LRT access effect. O ne would suspect that this stop is too far away for the CBD to have a trickle down effect. It seems that even though this area is on a LRT stop for a while, the demand to live here is not like the other areas. I n terms of dollar per square foot value, East 172 nd Street appreciated 182 p ercent in the years between 1998 and 2007, which also was the least out of any case study area. It is hard to pinpoint the reason that this area appreciated the least, given that this area has had time to grow around their LRT station. Noted earlier, this area was the furthest from the CBD, the main center of economic activity. It was also noted earlier that Portland has a significantly vibrant downtown for a city its size, it could be more important to be closer to the center city. At first glance, t he East 172 nd stop disproved much of the motivations for research question presented in this study. But Von Thunen and Alonso would have recognized that East 172 nd lies on the outer rings. The neighborhood has had time to grow around the station and it has not. It has an array of pedestria n amenities; it has dense multifamily units that speak to some intensity around stat ions. It has a transit utilizing populace, has good frequency and level of service but in the end the lots are large which may hurt the area in terms of value per square f oot. The major detriment: t he area is nine miles from the CBD, and does not have much in the form of commercial uses. It also may suffer from lack of highway access, as Parkrose/Sumner and MLK study areas do Perhaps this stop is a victim of the succes s of the CBD; in other areas this type of neighborhood typically has big box commercial nearby whereas in Portland people still shop and go downtown. Perhaps 172 nd Street represents where we are headed a world
117 where people value living downtown more than really that not like in other cities where downtown (jobs, restaurants, shopping) actually located on the periphery. MLK Blvd. / Killingsworth Street Proximity to C entral B usiness D istrict (CBD) reaps valuati on benefits? Convers ely, the MLK/ Killingsworth area -much closer to the CBD, appreciates quite robustly in the dollar per square foot ten year study, and year to year, had the highest appreciation yearly rate five out of the ten years. The property parce ls were typically more uniform than the areas further out from the CBD, reflecting possibly the density of streets in the area. Also, it should be noted that the MLK Blvd i s a major arterial whereas the other case study areas without transit do not have a s major corridor. It should be presumed that upon examining the parcel map that many of the white par cels directly on MLK Blvd. are non residentia l. Interesting to note is that this area is a historically African American neighborhood -to the extent that such areas tend to be undervalued this area has become revitalized when compared to other case study areas, as well as Portland as a whole. MLK/ Killingsworth has avoided the slumps in appreciation that hit the other study areas, and as noted earlier has had the highest appreciation year to year five times out of ten years. When examining the results in the appreciation per square foot the Martin Luther King and Killingsworth Street study area appreciated a robust 303 percent from 1997 to 2007 (Table 6 2). This 303 percent represents the highest appreciation per square foot out of any case study area. Although this area is not served by LRT, it is exactly a mile away from the Yellow Line, which connects the CBD with the Expo Hall in North Portland. This are a also represents the convergence of two arterial roads and these factors alone make it desirable location choice. However, how exactly does this area perform better than any other in this study? The proximity
118 to the CBD is matched by the Hollywood/42 nd Av enue case area, but the Hollywood area is still worth $75.54 per square foot in 2007 compar The benefits of LRT access did not prevent from East 172 nd Street/ Burnside Avenue from ranking last in terms of overall per square foot appreciation. This area and the Polk and Smith area were close in 1998 in SFR value per square foot, and while all case studies appreciated East 172 nd appreciated the slowest. East 172 nd had by far the newest housing stock of the bunch and was by f ar the farthest away from downtown. East 172 nd was not the lowest income area of the three areas defined as low income and was also not the lowest home value yet it still appreciated the least add to that the fact that East 172 nd residents reported using transit nearly tied for second in the whole study. Home value and home appreciation depends on a bundle of factors, one of which is location and the a menities that location affords. Contrary to East 172 nd the MLK Blvd/ Killingworth study area lacked access to LRT. The area ended up appreciating more per square foot than any other study area this again totally contradicts the research question What MLK did have going for it was a higher income area, prox imity to downtown and a much older housing stock. This case study illustrates how a strong LRT downtown core can indirectly bestow benefits to areas that lack direct access to LRT. While MLK/Killingsworth lacked the LRT proximity they still used transit, still used sidewalks and bicycles that took advanta ge of a -advantages that far away East 172 nd did not or could not take. Parkrose/Sumner was the lone case study that allowed a picture of before and after. The research question sta ted that the new station opening would cause a jump in the home appreciation percentage. This was true the area enjoyed a study high 21 percent per square foot appreciation in 2003. However, the station opened in 2001, another case study area without
119 tran sit appreciated by nearly as much that same year, and the national trend that year was also appreciating by great rates. The appreciation rate fell nearly seventeen percent the following year, so the validity of LRT access equaling home appreciation is at best weak. Perhaps it takes a year or so for the LRT effects to arrive in home values. Parkrose/Sumner being a low income area, with an abundance of non friendly transit land uses (factories, multi lane highways, large interstate interchanges and flyover station access) makes one suspect that this area simply needs more time to mature around its new stop. Still Parkrose/ Sumner faired third best in terms of total appreciation, so if one wishes to make that correlation there is hope. Similar to previous studies, high income areas tended to fare better than low income areas wherever possible (Kent & Parilla, 2007) The national land boom made it difficult to separate what and what was not an LRT induced appreciation. Hollywood and 42 nd while benefitting from LRT access may very well have appreciated just as greatly if they did not have an LRT station, given the 303 percent dollar per square foot appreciation of MLK/Killingsworth. The purpose of this study was to illustrate bu t not predict the correlation between LRT station proximity and higher home appreciation values. New LRT did seem to ha ve a positive effect overall on t he before and after valuation effect on the one case study, the Parkrose / Sumner area. This news is en couraging to the many metro areas expanding LRT systems currently. Among the factors that correlated to home appreciation, age of housing stock and d istance to the CBD had the highest positive correlation s to home appreciation in this stud y To summarize, LRT access did not necessarily equate higher appreciation rates. The MLK/ Killingsworth parcels appreciated 303%, and while being close to the city's core the area lacked walkable access to LRT transit. This could be a result of the land being undervalued as people sought to move closer to downtown. This theory has some weight as the 172nd Street
120 stop area, while furthest from the CBD also appreciated the slowest at 182%. This would contradict any research that access to transit, alone is the largest dete rmining factor in housing appreciation rates. It is interesting is that the areas closer to the core have significantly higher home values to start with. It was more important that the study area be closer to downtown though not necessarily older housing stock. It was found that higher income neighborhoods performed well in the study regardless of whether or not they had transit. The unique factors of Portland may have influenced the study based on the East 172 nd Street stop there may be a heightened ben efit of living closer to the CBD. The suburbs in Portland do not have the same draw that they would in other places. The UGB is redefining what it means to be suburban in Portland. Portland has a large number of people that utilize the LRT. Since the lines all meet in the hub of downtown there is an added benefit to living closer to the CBD. With the Martin Luther King/ Killingsworth Street area the housing values belied what was seen on the aerial photographs. While to the eye, it appeared to be a neighborhood in transition, the housing appreciation rates clearly benefitted more from being three miles away from downtown more than they did not benefit from having LRT in a walkable distance. The five scenarios offered very different pictures of what affected home appreciation and home values. It turns out, that multiple factors dictate home value; people are making trade offs. For example, people who may want larger homes or yards but cannot afford to live that close to the CBD may trade the highe r transportation expenses of commuting. These same homes may not be expected to appreciate as quickly because less people are willing to do that compared to those who want less house and want to live closer to downtown.
121 CHAPTER 7 CONCLU SION S AND RECOMMENDATIONS FOR FURTHER RESEARCH The research question was answered, though not the expected answer The research question sought to illustrate a clear pattern of LRT access improving SFR appreciation rates in a place where LRT is such an integral par boom that was especially strong in Portland during the time of the study, it is clear that distance to the CBD is the strongest factor in home appreciation in this city. The five scenarios offered v ery different pictures of what effect LRT had on home appreciation and home values. In areas that were high income areas, LRT had a reinforcing effect on already high home values. In a high income area that lacked LRT access, the home values still apprec iated faster than any other case study. In a low income area without LRT, that area still appreciated higher over a ten year period than a similar area that had access. Similar to previous studies, high income areas tended to fare better than low income areas wherever possible. The national land boom made it difficult to separate what and what was not an LRT induced appreciation. Hollywood and 42 nd Street while benefitting from LRT access may very well have appreciated just as greatly if they did not h ave a LRT station, given the 303 percent dollar per square foot appreciation of MLK/Killingsworth. The news was not all bad, however. In the older neighborhoods LRT access had a positive effect on SFR. New LRT did seem to have a positive effect overall on the before and after valuation effect on Parkrose/Sumner. This news is encouraging to the cities expanding LRT systems currently. Among the factors that correlated to home appreciation, age of housing stock and distance to the CBD had the highest pos itive correlations to home appreciation in this study.
122 There are many different strategies that could be utilized for further research in this area. While examining the mile distance, it would be beneficial to use network distance rather than actual dis tance. Network distance takes into account that commuters may be confined by sidewalks, fences, and landscaping. A more in depth look at network distance may better capture the effects of proximity. Other research recommendations include analyzing the w hole of residential parcels, including multi family, attached housing and/or industrial zones would be beneficial for those interested in transit/transportation planning. Another opportunity for further study would be comparing appreciation rates in areas near LRT that exhibit very similar characteristics in age of housing accompanying street network, or lot sizes. These factors, if ex amined may be able to isolate the effects of LRT access more effectively One further recommendation for research would be to obtain data from a period of time when land values nationally were declining or appreciating slowly. This study examined property values during a boom and the beginning of a bust cycle. Perhaps if the study was conducted after the end of a bust, the near LRT properties may be shown to recover quicker. How LRT proximate home values resist a regional or national trend to depreciate could offer a concrete picture on the value added of an LRT station. A study of this nature could be conducted in an LRT city that is less susceptible to housing booms and busts Another aspect that could be examined is the effect of crime or the percept ion appreciation could the perception of crime keep homes near stations from appreciating more? The results of this study indicate that LRT does have a somewhat positive correlation on home appreciation but is hardly t he predictor of home values or home appreciation. One of this
123 embraced the system. Another purpose was also to illustrate but not predict the correlation betw een LRT station proximity and higher home appreciation values. The research question was LRT would have conferred economic benefits on all areas with LRT compared to those without LRT access. Home appreciation was a gauge on such benefits. Home apprecia tion offered a benefit of being a percentage and number that could be compared over time rather simply. If LRT could create a location advantage anywhere, it would definitely be apparent in Portland, gration and walkable blocks. Portland also has the most powerful regional land use government in the country. in transit and transportation planning. A first ste p would be to end the county by county approach to land use and zoning decisions, a second step would be to initiate some sort of UGB to promote infill and density. These first two steps would be a great start to enabling a healthy LRT. In the future, an honest discussion informing the citizenry about the true cost of highways of place. The social implications of transit should not be underestimated as we try to reverse this loss of place.
124 REFERENCE LIST Al Mosaind, M., Dueker, K., & Strathman, J. (1993). Light rail transit stations and property values; a hedonic price approach. Transportation Research Board. Washington, DC. Alonso, W. (1964). Location and Land Use: Toward a General Theory of Land Rent. Cambridge, MA.: Harvard University Press. Alonso, W., & Muth, R. (2007). Resources for the Future, a nd the Founding of Urban Economics. Journal of the History of Economic Thought 67 84. Altshuler, A. A., & Luberoff, D. E. (2003). The Changing Politics of Urban Public Investment. Brookings Institution Press. American Public Transportation Association. (2008). APTA Fact Book Banister, D., & Berechman, J. (2000). Transport Investment and Economic Development. Journal of Transport Geography, 10 (1). Barclay, B. (2008). Appraiser. (S. Anderson, Interviewer) Bartholomew, K. (1997). Freeways or Communities: It's Your Choice. Bowes, D., & Inlandfeldt, K. (2001). Identifying the impacts of rail transit station on property values. Journal of Urban Economics, 50 1 25. Calthorpe, P., & Fulton, W. (2001). Regional City: Planning for the End of Sprawl. Island Press: Washington, D.C. Cambridge Symantics, Inc. (1999). Quantifying the Impacts of a Light Rail Transit on Adjacent Property Values. Unpublished manuscript. Cervero, R. (1984). Light Rail Transit and Urban Development. Journal of the Ame rican Planning Association, Vol. 50, No. 2 133 147. Cervero, R., & Duncan, M. (2002). Land Value Impacts of Rail Transit Services in San Diego County. Cervero, R., & Duncan, M. (2001). Rail Transit's Value Added: Effect of Proximity to Light and Commut er Rail Transit on Commercial Land Values in Santa Clara County, California. Washington, D.C. Chandler, C., & Noel, L. (2004). Effects of Light Rail Transit Congestion. Chen, H., Rufolo, A., & Dueker, K. (1998). Measuring the Impact of Light Rail Systems on Single Family Home Values: A Hedonic Approach with GIS Application.
125 Debrezion, G., Pels, E., & Rietveld, P. (2004). The Impact of Railway Stations on Residential and Commercial Property Value. Amsterdam. Diaz, R. B. (2001). Impacts of Rail Transit on Property Values. Dueker, K., & Bianco, M. (1999). Light Rail Tra nsit Impacts in Portland: The F i r st Ten Years. Transportation Research Board. Washington, D.C. Federal Housing Administration (2008). Retrieved February 5, 2009, from Federal Hosuing Admini stration Appraisal Methods Web site: http:// www.sunnations.com/mortgagelibrary/fha_appraisals/fha_appraisal_methods.asp Federal Transit Administration. (2002). National Transit Database. Washin g ton, D.C.: Federal Transit Administration. Garrett, T. A. (2004). Light Rail Transit in America: Policy Issues and Prospects for Economic Development. Gatzlaff, D. H., & Smith, M. T. (1993). The Impact of the Miami Metrorail on the Value of Residences Near Station Locations. Land Economics Vol. 69 54 66. Giuliano, G., & Hanson, S. (2004). The Geography of Urban Transportation. New York: The Guilford Press. Grass, R. (1992). The Estimation of Residential Property Values Around Transit Station Sites in Washington, D.C. Journal of Economics and Fin ance, 16 139 146. Halifax Regional. (2002). Transit and Land Use Form. Halifax Regional Municipality, Regional Planning. Handy, S. (2005). Smart Growth and the Transportation Land Use Connection: What does the research tell us? International Regional S cience Review Vol. 28, No. 2 146 167. Hess, D. B., & Almeida, T. (2006). Impact of Proximity to Light Rail Transit on Station area Property Values in Buffalo, New York. Urban Studies Vol. 44 1041 1068. Holahan, C. (2008, September 9). The Trouble wit h Transportation. Retrieved February 20, 2009, from www.businessweek.com. Huang, H. (1996). The Land Use Impacts of Urban Rail Transit Systems. Journal of Planning Literature INRIX. (2006). National Traffic Scorecard. Jackson, M. (2008). Light rail ca n turn into money train. Retrieved March 5, 2009, from Denver Post: www.denverpost/breakingnews/ci_10850014
126 Jacobs Engineering. (2008). Connectivity Benefits of a Regional Transit System. South Florida East Coast Corridor Study Kain, J. F., & Zvi, L. (199 9). Secrets of Successes: Assessing the Large Increases in Transit Ridership Achieved by San Diego and Houston Transit Providers. Transportation Research Part A 601 624. Kent, A., & Parilla, J. (2007). Did the Hiawatha Light Rail Line Increase Single Fa mily Residentail Property Values. Kilpatrick, J. A. (2005). Application of Repeat Sales Analysis to Determine the Impact of a Contamination Event. Journal of Housing Research, 15 (2). Kim, K. (2008). Impact Analysis of the Hudson Bergen Light Rail on Property Values. Knight, R., & Trygg, L. (1977). Land Use Impacts of Mass Transit. U.S Department of Transportation. Kuby, M., Barranda, A., & Upchurch, C. (2004). Factors Influencing Light Rail Station Boardings in the United States. Transportation Rese arch Part A 223 247. Landis, J., Cervero, R., Guhathukurta, S., Loutzenheiser, D., & Zhang, M. (1995). Rail Transit Investments, Real Estate Values, and Land Use Change: A Comparative Analysis of Five California Rail Transit Systems Lewis Workman, S., & Brod, D. (1997). Measuring the Neighborhood Benefits of Rail Transit Accessibility. Transportation Research Record 1576 (pp. 147 153). Litman, T. (2008). Evaluating Rail Criticism. Mackett, R., & Sutcliffe, E. (2003). New Urban Rail Systems:A Policy Based Technique to Make Them More Successful. Centre for Transport Studies 151 164. Marshall, A. (2001). How Cities Work: Suburbs, Sprawl and the Roads Not Taken. University of Texas Press. Metro Data Resource Center. (2007, November). Geographic Infor mation Systems. Portland, Oregon. Muth, R. (1969). Cities and Housing. Chicago: University of Chicago Press. National Association of Realtors. (2008). Annual Average Median Prices for Single Family Homes Condos. Retrieved February 20, 2009, from National Association of Realtors: www.scribd.com/doc/3631369/US 19682008 Annual Average Median Prices for Single Family Homes
127 National Association of Realtors. (2009). Housing Bubble Graphs: Charts of inflation adjusted, historical home prices Retrieved June 200 9, from http://mysite.verizon.net/vzeqrguz/housingbubble Nelson, A. C. (1992). Effects of Elevated Heavy Rail Transit Stations on House Prices with Respect to Neighborhood Income. Transportation Research Record 1359 127 132. Nelson, A. (1999). Transit Stations and Commerical Property Values: A Case Study with Policy and Land Use Implications. Oregon Department of Revenue: Tax Statistics 2006 07 Fiscal Year. (2007, April). Retrieved January 12, 2009, from Oregon Department of R e venue: http://www.oregon. gov/DOR/STATS/docs/303 405 07/303 405 07.pdf Oregon Metro. (n.d.). Map of Portland Urban Growth Boundary. Parsons Brickenhoff Quade and Douglas, Cambridge Systemantics. (1996). Making the Land Use Transportation Air Quality Connection: Analysis of Alternatives. 1000 Friends of Oregon Parsons Brinckerhoff. (1999). Th e Effect of Rail Transit on Pro p e rty Values : A Summary of Studies. Portland Maps (2007). Retrieved 2008, from Portland Maps: www.portlandmaps.com Pucher, J., & Renne J. (2003). Socioeconomics of Urban T ravel: Evidence from the 2001 NHTS. Transportation Quarterly pp. 49 78. Puhskarev, B., & Z upan, J. (1982). Where Transit Works: Urban Densities for Public Transportation. Urban Transportation: Perspectives and Prospects, ed. by H.S. Levinson and R.A Weant Ryan, S. (1999). Property Values and Transprotation Facilites: Finding the Transportati on Land use Connection. Journal of Planning Literature 412 427. Transportation Research Board. (2002). The congestion mitigation and air quality improvement prgram: assessing ten years of experience (TRB Special Report No.264). Washington, D.C: National Academy Press Transportation Research Board. (2003). Transit Capacity and Quality of Service Manual, 2nd ed. Washington, D.C.: Transportation Research Board. TriMet. (2008). Retrieved January 2009, from TriMet Web site: http://trimet.org/pdfs/history/r ailfactsheet U.S Census Bureau. (2008). American Community Survey. U.S Census Bureau.
128 Vessali, K. V. (1996). Land Use Impacts of Rapid Transit: A Review of the Empircial Literature. Berkeley Planning Journal 71 105. Weinberger, R. (2000). Commercia l Pr operty Values and Proximity to Light Rail: Calculating Benefits with a Hedonic Price Model. Transportation Research Board 79th Meeting. Washington, D.C. Weinstein, B., & Clower, T. (2002). An Assessment of the DART LRT on Taxable Property Valuations and Transit Oriented Development.
129 BIOGRAPHICAL SKETCH Stephen Thomas Anderson was born in Newark, New Jersey as the e ldest of three children His early years w ere spent in Northern New Jersey .From 1986 to 1989, he lived a life among cotton fields and hog farms and attended a rural sc hool in Halifax, North Carolina The contrast between urban and rural life would stay in his mind as a child -indicated by his affection for and memorization of Place s Rated Almanac, a book that ranked M etropolitan S tatistical A rea s in different categories such as transportation, economics and education. In 1990, his family moved to Florida, first Tallahassee for a year, and then Bradenton. A very different world than rural eastern North Carolina, it was hard at first to adapt to a system of sprawling, cookie cutter suburbs. These subdivisions, he later realized tend to divide the neighbor from the neighborhood. He graduated from Southeast High School in 1998, and stud ied Political Science for two years at Stetson University in DeLand, Florida. Transferring to the University of Florida in 2000, he chose to study History with a strong interest in American History. Graduating in 2003 with ente urban regional planning in 2006. He has interned for Tindale Oliver & Associates, a transportation planning firm in Tampa, Florida as well as the St. Johns River Water Management District in Pa latka, Florida. The need for all transit to become more utilized was realized in the summer of 2008 when commuting from Palatka to Gainesville (50 miles) every day when gas prices were four dollars. I am interested in urban places that work, vibrant street s, multimodal options, G eographical I nformation S ystems (GIS) and other transit/ transportation planning issues. He currently works as a transportation planner/ GIS analyst at the South Florida Regional Transportation Authority in Pompano Beach. That age ncy manages Tri Rail, the only commuter rail currently operating in the state. He hopes to help solve t his through good transit /transportation planning.