Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effects of land use, transportation infrastructure and housing affordability on growth management… Allison, Mark B. 1997

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1998-0004.pdf [ 37.58MB ]
JSON: 831-1.0088415.json
JSON-LD: 831-1.0088415-ld.json
RDF/XML (Pretty): 831-1.0088415-rdf.xml
RDF/JSON: 831-1.0088415-rdf.json
Turtle: 831-1.0088415-turtle.txt
N-Triples: 831-1.0088415-rdf-ntriples.txt
Original Record: 831-1.0088415-source.json
Full Text

Full Text

THE EFFECTS OF LAND USE, TRANSPORTATION INFRASTRUCTURE AND HOUSING AFFORD ABILITY ON GROWTH MANAGEMENT IN THE GVRD: A STUDY OF HOUSEHOLD TRAVEL BEHAVIOUR AND LOCATION DECISIONS by MARK B. ALLISON B.Sc, University of Windsor, 1980 M.Sc, University of Waterloo, 1982 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (School of Community and Regional Planning) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1997 © Mark B. Allison, 1997 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. DE-6 (2/88) Abstract A great deal of planning literature in the last decade has been devoted to growth management and the concept of land use and transportation interactions. "New" approaches to planning, such as Transit Oriented Development (TOD) and Neo-Traditional Neighbourhood Design, are products of this evaluation of current development practices. The influence of housing affordability and accessibility, although intuitively related to the growth management problems of urban sprawl and automobile dependence, has often been overlooked. The purpose of this research is to bridge important gaps in our understanding of how residential land use and transportation infrastructure investments are shaping unsustainable growth and travel patterns in the GVRD, which is the main problem being addressed. The research objectives related to this problem are the correlation of observed trends in growth, housing and travel indicators, the determination of the importance of price and accessibility factors in household location decisions, and the analysis of the role that land use and transportation decisions have played in influencing housing costs and accessibility. To provide a context for understanding the scope of the problem and the relationships between the research results and proposed recommendations, the applicable literature, theory, and policies in the areas of growth management, land use, transportation and housing are given. Supporting research results include: a survey of senior stakeholders in the region on land use, transportation and housing issues; a synthesis of significant socioeconomic, growth, transportation and housing data; a summary of surveys outlining preferences for residential location and housing type; and an analysis of Place of Work data crosstabulated against Place of Residence and socioeconomic variables. The results show a strong dependency between location decisions and the cost and accessibility of housing, particularly for the critical group of younger households with children. Policy recommendations, based on the research and covering land use, transportation, housing, governance and education, are proposed to address the main sustairiability problems studied. The recommendations focus on promoting affordable, higher density communities, with a choice of transportation modes, as an attractive alternative to lower density, automobile-dependent suburbs. 11 Table of Contents Abstract ii List of Tables vi List of Figures vii Acknowledgments ix Prologue x 1. Introduction 1 1.1 Overview 1 1.2 Problem Statement 3 1.3 Research Questions 4 1.4 Study Area Definition 5 1.5 Thesis Structure 8 1.6 Research Methodology 11 1.6.1 Analytical Techniques 11 1.6.2 Special Considerations 16 2. Review of Applicable Literature, Theory and Policies 21 2.1 Growth Management in Cascadia 21 2.1.1 Historical Overview 21 2.1.2 Overview of B.C.'s Growth Management Legislation 25 2.1.3 Initial Experience with B.C.'s GSA 27 2.2 Land Use and Transportation 30 2.2.1 Overview 30 2.2.2 Impacts and Costs 34 2.2.3 Land Use Tools 39 2.2.4 Transportation Tools 42 2.2.5 Jobs/Housing Balance 46 2.2.6 Density and Land Use Mix 51 2.2.7 Summary 61 2.3 Land Economics, Housing and Location Choice 64 2.3.1 History and Overview 64 2.3.2 Bid-Rent Theory 67 2.3.3 Decision Tree Model 71 2.3.4 PITTS 72 2.3.5 Factors Affecting the Cost of Housing 73 2.3.6 Location of Employment 83 2.3.7 Summary 87 iii 2.4 Growth Management Policies 88 2.4.1 De Facto Policies 88 2.4.2 Official Policies 89 2.5 Land Use and Transportation Policies 95 2.5.1 De Facto Policies 95 2.5.2 Federal Policies 97 2.5.3 Provincial Policies 99 2.5.4 Regional Policies 104 2.5.5 Municipal Policies 110 2.5.6 Summary 114 2.6 Housing Policies 116 2.6.1 De Facto Policies 117 2.6.2 Federal Policies 120 2.6.3 Provincial Policies 122 2.6.4 Regional Policies 124 2.6.5 Municipal Policies 125 2.6.6 Summary 129 2.7 Summary of Applicable Literature, Theory and Policies 130 3. Research Findings 133 3.1 Survey on Land Use, Housing and Transportation Issues 133 3.1.1 Survey Description 133 3.1.2 Summary of Key Informant Responses 135 3.2 Population and Job Growth Trends 140 3.3 Transportation Trends 146 3.3.1 Personal Travel Trends 148 3.3.2 Transportation Irifrastructure Investments 156 3.4 Housing Type and Location Trends 161 3.4.1 Demographic and Socioeconomic Influences 162 3.4.2 Housing Starts and Types 165 3.4.3 Supply of Rental Housing 179 3.5 Housing Cost Trends ...181 3.5.1 Housing Prices 182 3.5.2 Mortgage Interest Rates 190 3.5.3 Rental Prices -191 3.6 Buyer and Location Preferences 194 3.6.1 Choices of Residential Location Survey 194 3.6.2 GVRD GOMD Telephone Survey 196 3.6.3 A Market Profile of Vancouver's Kitsilano Neighbourhood 207 3.6.4 AMarket Profile of Surrey's Clover Valley Station Subdivision 213 3.7 Place of Residence and Place of Work Analysis 222 3.7.1 Data Specification for Residence and Workplace Crosstabulations 222 3.7.2 Place of Work and Place of Residence Crosstabulations 225 3.7.3 Income Distribution by Place of Work and Place of Residence 233 iv 4. Summary, Discussion and Conclusions 239 5. Recommendations 247 5.1 Land Use 247 5.2 Transportation 250 5.3 Housing 251 5.4 Governance 253 5.5 Education and Awareness 255 Epilogue 259 Glossary 261 Bibliography 268 Appendix A - A Land Use Planner's Guide to Transportation Planning Methods 276 Appendix B - Key Informant Responses to Land Use, Transportation and Housing Survey 283 Appendix C - Place of Work versus Place of Residence Matrices for Baseline Case and (LT40, NMSD, Ground, Owned, Moved) Case 304 List of Tables Table 1 - Decision Tree Using a Housing "Bundle" 72 Table 2 - Supply of Land and Absorption Rates (1990) 73 Table 3 - Comparison of Vancouver and Langley District 110 Table 4 - Examples of Indirect Housing Policies 119 Table 5 - Numerical Targets for Growth Management 140 Table 6 - Analysis of Trends in Jobs/Housing Balance (1981-1991) 144 Table 7 - Changes in Travel Indicators (1985-1992) 148 Table 8 - Transport 2021 Transportation Targets (AM Peak Hour Period) 151 Table 9 - Inter-regional Workplace Share 153 Table 10 - Suburban Versus Regional Workplaces 154 Table 11 - Shaping Suburban Growth with Transportation Infrastructure (1976-1996).. 157 Table 12 - Factors Influencing Housing Trends 162 Table 13 - GOMD Survey Income, Age and Housing Type of Respondents 198 Table 14 - GOMD Survey Income, Age and Tenure Type of Respondents 199 Table 15 - GOMD Survey Level of Satisfaction with Current Housing 200 Table 16 - GOMD Survey First Housing Choice by Income and Age 201 Table 17 - GOMD Survey First Housing Choice by Subregion and Tenure 201 Table 18 - GOMD Survey Preferences by Subregion, Housing Type and Tenure 202 Table 19 - GOMD Survey Preferences by Subregion, Income and Age 203 Table 20 - Age Trends in Kitsilano 208 Table 21 - Income Trends in Kitsilano 209 Table 22 - New Housing Market in Kitsilano 210 Table 23 - Comparison of First-time and Previous Buyers in Clover Valley Station 217 Table 24 - Work Locations and Previous Residential Locations of Clover Valley Station Residents 220 Table 25 - Work Trip Matrix Format 223 Table 26 - Variables for Trip Matrix Crosstabulations 224 Table 27 - Income Distribution Matrix Format 225 Table 28 - Surrey POW Profiles: Homeowners, AH Ages, $60,000-$79,999 Personal Income, All Household Incomes, Moved in Last Five Years 229 Table 29 - Surrey POW Profiles: Homeowners, All Ages, $40,000-559,999 Personal Income, $40,000-559,999 Household Income, Moved in Last Five Years 229 Table 30 - Sample of Graduated Regional Subdivision Levies 249 v i List of Figures Figure 1 - Study Area Context Map 5 Figure 2 - Definition of Subregions 7 Figure 3 - Definition of Subareas 7 Figure 4 - Municipalities within the Study Area 8 Figure 5 - A Comparison of Subdivision Densities 57 Figure 6 - GVRD Land Price Gradients (1991) 76 Figure 7 - Normal Land Price Distribution Curve 77 Figure 8 - Balanced Land Price Distribution Curve 78 Figure 9 - Financial Analysis of a Current Development Project 82 Figure 10 - Income Class Polarization in Vancouver and the CMA 127 Figure 11 - Population Growth (1981-1991) 141 Figure 12 - Population Growth in the Study Area (1991-1996) 141 Figure 13 - Ratios of Employment to Employed Residents (1991) 143 Figure 14 - Trends in Employment/Employed Resident Ratios 144 Figure 15 - Transportation Infrastructure and Employment Concentrations 156 Figure 16 - Transportation Infrastructure and Commuting Distances 157 Figure 17 - How Transportation Planning Generates Traffic 159 Figure 18 - Housing Starts in the CMA (1975-1995) 161 Figure 19 - GVRD Population Trends (1976-1996) 162 Figure 20 - Household Demographic Trends (1971-1991) 163 Figure 21 - Income and Unemployment Trends (1976-1995) 163 Figure 22 - Mortgage Rate and Inflation Trends (1976-1995) 164 Figure 23 - Normalized Total CMA Housing Starts ...166 Figure 24 - Composition of CMA Housing Starts 167 Figure 25 - Share of CMA Housing Starts by Municipality 168 Figure 26 - Share of CMA Housing Starts (Vancouver, Surrey excluded) 169 Figure 27 - Location of CMA Apartment Starts (1975-1995) 172 Figure 28 - Location of CMA Townhouse Starts (1975-1995) 173 Figure 29 - Location of CMA Detached House Starts (1975-1995) 174 Figure 30 - Comparison of Housing Starts and Composition in Various Municipalities ...177 Figure 31 - CMA Private Rental Stock versus Population 179 Figure 32 - CMA New Apartment Prices (1991-1996) 183 Figure 33 - CMA New Townhouse Prices (1991-1996) 184 Figure 34 - CMA New Detached House Prices (1991-1995) 185 Figure 35 - GVREB MLS Apartment Prices (1991-1996) 186 Figure 36 - GVREB MLS Attached House Prices (1991-1996) 187 Figure 37 - FVREB MLS Condominium Prices (1991-1996) 188 Figure 38 - MLS Detached House Prices (1991-1996) 189 Figure 39 - Cost of Mortgages 190 Figure 40 - Average Two Bedroom Apartment Rents 191 Figure 41 - Mortgage Equivalents to Rents Without Taxes or Utilities 192 Figure 42 - MLS Prices for Vancouver and Fraser Valley SFDs 193 vii Figure 43 - Acceptance of Density by Different Household Types 205 Figure 44 - Effect of Price Differences on Housing Preferences 206 Figure 45 - Overview of Kitsilano Neighbourhood 207 Figure 46 - Location of Clover Valley Station Subdivision 214 Figure 47 - Age Profile of Buyers in Clover Valley Station 215 Figure 48 - Age Profile of Children in Clover Valley Station 216 Figure 49 - Household Mortgage Load in Clover Valley Station 218 Figure 50 - Proportion of Residents in (LT40, NMSD, Ground, Owned, Moved) Case ...227 Figure 51 - Income Distribution and POW for Surrey Residents 232 Figure 52 - Income Distribution and POW for Vancouver Residents 233 Figure 53 - POW and POR Income - Mission 234 Figure 54 - POW and POR Income - West Vancouver and UBC/UEL 235 Figure 55 - POW versus POR Income - Regional Core 236 Figure 56 - POW versus POR Income - Suburbs 237 viii Acknowledgments This work is dedicated to the hundreds of citizen advocates across Canada who are working on land use and transportation issues to promote social equity, environmental protection, and fiscal responsibility. In particular, I would like to thank past and present colleagues in NGOs such as Environmentalists Plan Transportation (EPT) and Transportation Alternatives in Toronto, Citizens for Safe Cycling and the Transportation Environment Action Plan (TEAP) in Ottawa, Better Environmentally Sound Transportation (BEST) and Cycling BC in Vancouver, Transport 2000 Ontario and Transport 2000 Canada. There is little doubt that our society would be worse off today without such citizen advocacy groups. A glance at any American city, characterized by empty cores gutted by freeways, is a memorial to the effectiveness of public involvement in Canada. Many initiatives now considered to be mainstream by professionals, such as bicycle facilities and neighbourhood traffic calming, originated from unpaid advocacy groups pushing a new agenda. Tragedies such as the Cassiar Connector freeway project and Knight Street truck route in Vancouver, or the urban sprawl common throughout the Fraser Valley, are reminders of what can happen when government and economic interests are too strong or the citizen watchdogs are too weak or unorganized. The encouragement and support of Dr. Tom Hutton, made an important contribution to the work. Useful feedback was also received from Dr. Mark Roseland of Simon Fraser University, Dr. Nancy Knight and Ralph Perkins of the GVRD, as well as the many who responded to requests for information and opinions on land use, transportation and housing issues. Special thanks go to Hugh Kellas and the GVRD Strategic Planning Department for providing much of the data used in this work and funding for the essential census data compiled by Statistics Canada's Ted Brown. Finally, I would like to extend my warmest thanks to Joell Vanderwagen, a long-time mentor and friend, who introduced me to the key roles that citizen advocacy and the planning process had to play in the promotion of a sustainable and equitable society. ix Prologue Early in the writing of this thesis, a lingering cold led to musings on the possibility that the bubonic plague was again rearing its ugly head. The thought then occurred that there were a lot of similarities between the plague and the scourge of urban sprawl. This scourge, inflamed by a growing psychological and physical dependence on the automobile, afflicts most major metropolitan areas around the world. A brief historical review of the plague may therefore be a suitable, if sobering, introduction to the thesis. "The Bubonic Plague1 The plague is one of the most devastating diseases that has ever afflicted mankind. It is a highly contagious fever caused by the bacillus Yersinia pestis, which is carried by fleas that infest rats. The plague, commonly called bubonic plague or the Black Death, has been known since ancient times, but the best documented instance was its deadly appearance in Europe in 1347. It raged throughout all of Europe, killing at least one-fourth of the population—probably 25 million people. Without understanding how it was spread, people had no defense against the disease. In general, the population of Europe did not recover to its size before the plague until the 16th century, and some towns never recovered. The immediate results of the plague—a general collapse of economies, a breakdown of class relationships, and a halt to wartime hostilities-forced a massive restructuring of society. It has had a lasting impact on art, literature, and religious thought." Millions of people around the world have also been killed and wounded, directly or indirectly, by the automobile. Over 500 deaths and almost 50,000 injuries result from car crashes every year in B.C. alone.2 There is also no doubt that our economic and cultural systems have succumbed to the automobile and that we have been forced to completely restructure North American society 'Excerpts from Compton's Interactive Encyclopedia, SoftKey Multimedia, 1996. 2 Motor Vehicle Branch, Ministry of Transportation and Highways, Province of British Columbia 1994 Traffic Collision Statistics (1994): p. 26. x and most of our communities to accommodate its needs. It could easily be argued that the need to protect access to the cheap petroleum sources required by an automobile-based economy and culture has been a major factor behind several large-scale armed conflicts this century, such the Suez Canal conflict (1956-57) and the Persian Gulf War (1991). We drive our children to school, ostensibly for their own safety. We allow the brown chemical soup that hangs above us to burn our eyes and afflict those with respiratory ailments. We watch our forests and pastoral farmland succumb to strip development and dysfunctional suburbs. We regret having less time to spend with families and friends as we commute further and work longer hours to pay for our cars and housing. We feel our blood pressure rise as we sit in traffic. Our architects now design houses around large garage entrances. We lament the loss of a sense of community and the feeling that we no longer know our neighbours. We complain bitterly about paying the increasingly high insurance premiums needed to repair or scrap the human and automotive bodies that have fallen victim to the traumatic events that we prefer to call "accidents" instead of crashes or collisions. How is it possible that, given all of these overwhelming indicators, the opinion polls still show that the greatest concerns of residents living in the Lower Mainland of B.C., the priorities that we insist our decision makers take action on, consistently include relief from traffic congestion and parking shortages? Like the plague, it seems that our society as a whole doesn't realize that the causes of the affliction are all around us. Once society recognizes their significance and accepts the need to change, remedies may be as easy to implement as the sanitary sewer systems that have made the bubonic plague little more than an unpleasant historical anecdote. The pages that follow will attempt to demonstrate how a study of our two most basic needs, shelter and access to necessities, can be used as a powerful mechanism to explore the magnitude and implications of the problems of urban sprawl and automobile dependence. Transportation, land use and housing policy directions can then be applied which, with sufficient political will and public awareness and support, could lead towards long-term solutions to these problems. xi 1. Introduction 1.1 Overview To loosely paraphrase Jane Jacobs, "This thesis is an attack on current urban planning in metropolitan regions."3 The negative effects of the urban sprawl and automobile dependence common throughout Canada and the U.S. in terms of social, environmental, and economic sustainability are well documented and the serious consequences are clear. Nevertheless, remarkably little has changed since Jacobs wrote the chapter with the tell-tale title of "Erosion of cities or attrition of automobiles" in her 1961 classic The Life and Death of Great American Cities. Her descriptions of the sprawling automotive wastelands that Detroit and Los Angeles had become are as applicable today as they were 35 years ago. Greater Vancouver is increasingly compared with L. A. Official policies on growth management appear to address the need for compact, complete communities oriented around walking, cycling, transit and goods movement. Observed trends, such as the increasingly distant location of new low-density development or the high level of per capita car use, indicate that this is not actually the case. An analysis of the policy instruments in place to support growth management policies show that these are having little effect and in some cases may be exacerbating the situation. The attitude of "we'll keep doing things the way we've always done them" seems to permeate the working culture of the current generation of planners and engineers at the municipal, regional and provincial levels. Municipalities complain of growing traffic and pollution, but little effort has been made by either the suburban municipalities, which act as bedroom communities, or the urban municipalities, which serve as employment centres, to address the issues that are at the heart of the problem: a lack of affordable housing, a better balance of jobs and housing within the region, 3 Jane Jacobs The Life and Death of Great American Cities (New York: Random House, 1961). 1 and the need for much higher overall densities. Tax payers view progressive planning initiatives by governments as an attempt to increase their tax burden or impose undesired changes on them. These factors have combined to create a powerful inertia to change among politicians, civil servants, and the public. Ways must be found to overcome this inertia before new planning principles and engineering standards, those which promote sustainable urban forms, will be accepted by society and implemented. This thesis is intended to challenge and influence the attitudes of professionals, decision makers, and the general public towards land use, transportation and particularly housing. The goal of the thesis is to contribute to creating a metropolitan region that is designed and built as a higher-density, people-oriented form instead of a lower-density, automobile-oriented form. The underlying hypothesis is that numerous government policies at all levels are effectively driving people to drive in order to access employment, amenities and, above all, affordable, family-oriented housing. These policies are often justified in the name of preserving personal freedom or, in the currently conservative fiscal climate, avoiding the appearance of providing subsidies for the alternatives. Nevertheless, enormous personal and public resources have been, and continue to be, expended to adapt our lifestyles and our neighbourhoods, cities, and metropolitan regions to accommodate a need for cars that we have created ourselves. The thesis will examine the available choice of housing types, locations, and costs, as well as the transportation modes that are available to move people between their chosen homes and their other activity centres, such as work, shopping, education, and recreation. A focal point of the research is the cost of ground-oriented family housing suitable for first-time buyers and its location with respect to the location of employment centres in the Greater Vancouver Regional District (GVRD). Socioeconomic factors will be examined to determine how these factors influence the lifestyle decisions of newer middle-class families, who form the basis for the perpetuation of low-density, car-dependent communities. It will be argued that, by not offering newly forming households an attractive alternative to the suburban lifestyle through sustainability-2 based land use and transportation planning, we are creating a region that not only has an enormous ecological footprint but lowers the quality of life for its residents as well.4 The thesis will be an examination of fundamental relationships between land use and transportation infrastructure as applied to housing decisions in a metropolitan context. Many planners, engineers, decision makers and members of the general public should intuitively recognize the existence of these relationships. Given the magnitude of the subject area, only primary indicators will be synthesized into a cohesive picture and an investigation of all possible indicators will not be attempted. These relationships will be translated into effective housing policy and policy instrument recommendations which address the interdependent issues of urban sprawl and growth management. An effective governance structure and a program of public education, which raises awareness of the issues and proposed solutions to a high level, will be outlined as integral components of a comprehensive strategy. "Regional and local governments should have housing as a centrepiece of their policy." ...Regional Planner 1.2 Problem Statement Land use patterns in the GVRD have resulted in prohibitively high housing prices near many of the region's major employment concentrations, particularly for the ground-oriented housing preferred by many new family households. This situation has created increasingly distant, low-density suburbs and exurbs which provide more affordable housing, but are inherently automobile-dependent. Households wishing to own ground-oriented housing appear to have little choice other than long-distance commuting, and large expenditures in personal and public transportation infrastructure have been made to accommodate trends towards increased travel. 4 Mathis Wackernagel and William Rees Our Ecological Footprint: Reducing Human Impact on the Earth (Gabriola Island, B.C.: New Society Publishers, 1996): p. 86, has an example of the Fraser Valley's "footprint." 3 Effective, ascertainable policies and policy instruments at the municipal, regional, and provincial levels are not in place to provide sufficient affordable housing near employment centres in order to counter urban sprawl and unsustainable travel patterns. It is highly unlikely that the GVRD can achieve its goal of creating complete, walking, cycling, and transit-oriented communities within a compact metropolitan region in the absence of such policies and policy instruments. 1.3 Research Questions The following questions will be addressed by this thesis research: • What are the trends in land use density and mix, jobs/housing balance, housing markets and transportation infrastructure that relate to GVRD housing affordability and travel patterns? • How important is the lack of suitable, affordable family-oriented housing close to jobs and amenities in household location decisions? In particular, what are the locational and travel influences of the critical "barometer" group of younger households with children seeking to own ground-oriented housing, and how do these differ from the average household? • Is the "push-pull" theory of land use and transportation modeling, i.e., integrating the "push" effect of high housing prices near employment concentrations and the "pull" effect of low housing prices in distant suburban areas, a more reasonable basis for strategic planning than the simple gravity model, which discounts the importance of housing prices on travel patterns? • In the context of the available literature, current provincial and regional growth management policies and the research findings, which new or existing policies and policy instruments have the greatest potential to promote affordable housing and a sustainable transportation system to support the goal of creating a compact metropolitan region in the GVRD? 4 1.4 Study Area Definition Before beginning, it is worthwhile to define the temporal and geographic study area and terms used to represent components of the study area. The study area includes most of what is known as the Lower Mainland of British Columbia (B.C.), primarily municipalities within the Greater Vancouver Regional District (GVRD) which had either more than 5,000 residents or more than 5,000 jobs in 1991. References will also be made to municipalities in the neighbouring Fraser Valley Regional District (FVRD), which is increasingly becoming an integral part of the GVRD's "commutershed." Regional Districts in B.C. are collections of municipalities and unincorporated areas which have limited jurisdiction over matters such as planning and shared infrastructure. For readers less familiar with B.C. or the GVRD, the following context map, including locations of places referred to later in the text, will be of use. Source: Base map from Maplnfo Corporation, overlays by author. Figure 1 - Study Area Context Map 5 Three major subregions will be referred to: the Regional Core, Suburbs, and Exurbs. The "Regional Core" is defined in this thesis as the central, higher-density GVRD municipalities which contain most of the major employment centres within the region, including Burnaby, New Westminster, North Vancouver City, Richmond, and Vancouver. The "Suburbs" are denned as the so-called "bedroom" communities surrounding the regional core, which may have major employment centres, but exhibit a severe imbalance between the number of employed residents and the number of employment opportunities available locally. Inner Suburbs and Outer Suburbs of Vancouver will also be referred to. "Inner Suburbs" include Burnaby, New Westminster, the North Shore municipalities and Richmond, while the "Outer Suburbs" include all other GVRD municipalities. "Exurbs" are defined here as those municipalities which are far from the regional core, either by physical distance or access time, and export significant numbers of workers into the suburbs and regional core. Exurbs often have a distinctive economic base underlying a "bedroom" function, usually agriculture. "Subareas" are collections of municipalities with common characteristics and clear geographic boundaries that are often used by the GVRD and some other agencies for data reporting purposes. For clarity of data representation, the municipality is the smallest geographical unit considered, which can result in the combination of urban and rural characteristics in the aggregated data. Delta, Langley District, Richmond and Surrey are examples of this urban/rural mixture. For historical reasons, the name "Langley Township" is often used for "Langley District" or the "District of Langley." Richmond in particular should be viewed as a special case. Northwest Richmond acts as part of the regional core, West-central Richmond is a classic suburb and eastern Richmond is clearly rural. Several municipalities have been amalgamated since the 1991 census, notably Abbotsford and Matsqui, and the GVRD has grown since then to become the equivalent of the 1991 Vancouver Census Metropolitan Area (CMA) with the addition of Langley City, Langley District, Maple Ridge and Pitt Meadows. A City in British Columbia is a municipality with more than 5,000 residents and higher overall density, while a District has more than 5,000 residents with an area greater than 800 hectares and a population density leass than 5/hectare. 6 Source: Base map from Statistics Canada, overlays by author. Figure 2 - Definition of Subregions Source: Base map from Statistics Canada, overlays by author. Figure 3 - Definition of Subareas In addition to the subareas shown, various reports often refer to the "Burrard Peninsula," which includes Vancouver, Burnaby, New Westminster and UBC's University Endowment Lands (UEL), while "South Fraser" consists of the South Region and the Langleys. iWest Vancouver r N Vancouver Dist TJ.E.L. s Vancouver Source: Base map from Statistics Canada, overlays by author. Figure 4 - Municipalities within the Study Area 1.5 Thesis Structure This section will describe how the various parts of the thesis address the denned research problem and research objectives. An overview of growth management, the relationships between land use and transportation, location theory and the land economics of affordable housing is the starting point. These sections combine a literature review of applicable planning work with an introduction to the theoretical concepts involved to provide a framework for understanding the nature of the problem. 8 The content and effectiveness of current growth management, transportation and land use and housing policies and policy instruments, taken primarily from "Cascadian" jurisdictions (B.C., Washington, and Oregon) will be then be surveyed. De facto policies, based on observed trends, will be given first in each of these sections, followed by the official policies. An emphasis will be placed on the policy climate that influences low-density urban sprawl and automobile dependency in the Lower Mainland of B.C. Research findings are then presented, which consist of a number of interrelated components: • A survey of key informants on their attitudes towards land use, transportation and housing affordability issues to set the context for observed trends in these areas; • A consolidation of growth trends in terms of population and employment; • A consolidation of transportation trends, including aggregate travel behaviour and a review of the types and impacts of recent infrastructure investments; • A consolidation and analysis of key housing trends, including the number of starts, types and locations of units being built, and the cost of owning and renting in strategically important locations of the Vancouver CMA; • A review of available opinion surveys of the stated and revealed preferences of potential home buyers, in particular first-time home buyers; • Market analyses of the two opposing choices in housing types and locations facing first-time home buyers for comparison; • An analysis of the relationships between Place of Work and Place of Residence data for various municipalities as a function of selected household socioeconomic variables; • The compilation of income profiles for individuals according to their Place of Residence and Place of Work and comparison of these against the identified travel patterns and the cost of housing to determine the relative strength of push-pull factors. A discussion of the research findings will then summarize key relationships and how these impact growth management goals, combined with conclusions. 9 In the final section, policies and policy instruments will be recommended which contain mechanisms for promoting sustainable land use, alternative transportation modes and affordable housing in support of the GVRD's growth management goals. How to Read This Thesis The length of this thesis is intended neither to impress nor intimidate the reader, rather to present all of the necessary facts, figures and background information in forms needed to support the analysis and conclusions made These forms include photos, maps, graphs, tables, equations, lists, quotes, footnotes and tevt Considerable effort has been put into formatting the tables and figures in ways that facilitate the comparison of data and the identification of important features As the work is intended to be accessible for both practitioners and the reasonably informed and interested layperson, a summary is usually presented in general terms either at the start of a section, directly before or after where more detailed data or findings are presented, and at the end of longer sections Readers have the option of examining a section or data in more detail or proceeding The approach chosen for selecting the background information presented was to provide all necessary and applicable data while referring the reader to more comprehensive works for details on specific subjects Readers who feel they have a high level of expertise in one or more areas of the available literature, theories or existing policies, may wish to skip these sections Information which is more technical, or which complements the research but is not essential for following the findings and conclusions, is enclosed in grey boxes such as the one that you are now reading A glossary of terms is included at the end of the thesis to provide convenient access to terms and acronyms used throughout the document Acronyms are usually defined once in the text, after their first usage Several appendices provide more detailed information that would be useful to many readers, but potentially disruptive if left in the main body of text 10 1.6 Research Methodology Given the enormous volume of readily-accessible background information that was available for the preparation of overview sections, this research methodology section will concentrate on describing the analytical techniques and special considerations that were applied in the preparation of "Research Findings" sections. 1.6.1 Analytical Techniques An exhaustive study of the locational and travel behaviour of all classes of household and housing types would have been far beyond the scope of a masters thesis. On the other hand, a study of carefully chosen "barometer" groups and areas allows locational decisions and travel habits to be studied on a micro scale (individuals and households) and related to cumulative impacts on land use and transportation at the macro scale (municipalities, subareas, subregions and regions). The study will therefore provide aggregated information for the baseline case of all residents and a framework for the closer investigation of subgroups found to be of interest. The presentation of every possible growth, economic and housing indicator would also likely obscure the main influences being exerted on selected barometer groups and the most important impacts of their behaviour on growth management goals. As a result, indicators are presented only where they can be related to the research questions and readily correlated with other indicators. As this research has a significant regional focus, there is a heavy dependence on secondary data sources, usually census or survey data collected by large organizations with different mandates and a wide variety of data definitions and collection methods. As a result, not all indicators of interest are available for all time periods and geographic borders. Although the data contains a few "gaps" in socioeconomic indicators, the independent collection of a complete set of primary data would be impossible to produce in a timely or cost effective manner. Cross-checks of the 11 available secondary data sources were made wherever possible to extract consistent conclusions. This "triangulation" approach is discussed later in this section.5 For readers with an interest in analytical techniques, these are now described. Certain techniques will be expanded on, where necessary, in conjunction with the presentation of findings. Key Informant Interviews A number of key informants in the areas of growth management and housing were interviewed informally during the initial phases of the research to establish: • That the problems identified were indeed significant and that a clear research gap existed; • Which data sources were readily available and which needed to be collected or modified; • The identities of appropriate individuals and agencies to contact for further information. All questions were open-ended, and were intended to provide a broad overview of the range of attitudes that might be expected from practicing professionals. Areas which appeared to have the greatest importance and diversity of opinion were incorporated into a survey of key-informants. Key Informant Surveys As a result of the preliminary key informant interviews, personal experience and a review of the literature, it was noted that there was a significant diversity of opinion on the issues within the professional community as to the importance of the issues being researched and the appropriate direction for policy. While this was originally not a focus of the research, an understanding of the attitudes of key government staff and decision makers on the issues being studied was considered to be essential. The survey proved to be a valuable complement to the research results by providing practical insights, background information and a guide to the acceptability of the final policy recommendations presented. 5 Therese L. Baker Doing Social Research (New York: McGraw-Hill, 1988): p. 260, "Deciding if Secondary Analysis is Appropriate for Your Topic." 12 Respondents were identified through earlier research into the issues and the recommendations of key informants. The responses are intended to identify the importance that the respondents assigned to the issues and the factors which were considered most likely to influence policy. The responses were summarized, but not coded for statistical analysis due to the limited sample size (18). The intent was to identify the "opposing camps" of the issues and gain insight into local and regional influences that do not appear in the global literature. Notable quotes appear in boxes throughout the thesis where applicable to provide the reader with a taste of the opinions held by practicing professionals. Focus groups were suggested as an alternative to the use of surveys. As the criteria for effective focus groups require that homogeneity and anonymity are preferred, discussion should be directed and debate should not be encouraged. It was decided that it would be difficult, given the senior level and highly diverse nature of the respondents, to assemble 8-10 such people in a room for 90 minutes and to keep the group from slipping into a heated debate over the issues. Growth, Transportation and Housing Trend Analysis Available data, generally from regional and federal government sources, was organized using computer databases and text tables for presentation in a number of ways, primarily with GIS maps, line graphs, bar charts, and tables. The goal of this analysis was to consolidate available trend information, present the most important information as it affects growth management, transportation infrastructure and housing affordability, and organize it in a way that removed the clutter of unrelated data and facilitated comparison. Readers should be able to confirm the premise of the problem statement by this point. Household and Buyer Surveys Existing survey data was available in either report or computer readable format. Raw data for the GVRD GOMD telephone study and developer surveys was available for crosstabulations and 13 detailed analyses of the attitudes and preferences of different types of household, which provided valuable insights. Data and results available only as printed reports were of significantly lower value, as report authors tended to present only the information that was requested by the commissioners of the report, as opposed to future researchers with other research questions. Where possible, applicable results from published results are analyzed and presented. Direct evidence on the housing preferences of younger, moderate-income, family households currently working in locations with high housing prices was needed to validate data from secondary sources. As a result, market analyses of two "barometer" communities were conducted: • A standard market analysis of the Kitsilano neighbourhood on Vancouver's expensive West Side, where low-rise condominium housing currently predominates new starts; • A household survey of a recently-completed, entry-level subdivision of detached houses (Clover Valley Station in Surrey). The subdivision selected was a classic "bedroom community," isolated from employment and commercial centres. It has received a large number of awards for being a model "neo-traditional" community, considered by many planners as an appropriate direction for higher-density, ground-oriented housing in the GVRD. Clover Valley Station marketing data was offered by the developer of the subdivision. It was decided that the time and expense of a specific survey was not justified, and that any missing information could be derived or inferred from the available data, which proved to be true. The data was then imported into SPSS statistical analysis software to generate buyer profiles for different types of household using crosstabulation techniques. Numerical Analysis of Census Place of Work Data The highlight of the research findings is the analysis of the special crosstabulation of 1991 Census Place of Work data against Place of Residence, individual, and household variables. This data, with over 800,000 elements, was provided by Statistics Canada in Excel spreadsheet software 14 format. Through Excel advanced filtering mechanisms and the "Visual Basic" programming language routines, important differences in the locational patterns of different types of individuals and households could be identified. Income profiles of employed residents in Place of Residence municipalities and employees in Place of Work municipalities were also derived from this data. GIS Representation of Data A Geographic Information System (GIS) provides a powerful way to visualize spatial correlations in data and identify "hot spots" using thematic maps and map overlays, making it an excellent tool for the investigation of urban sprawl and growth management. GIS maps and map overlays which support the presented background information and findings were therefore generated and appear throughout the thesis. Maplnfo was chosen as the GIS system, since suitable base maps for the study area were available and an inexpensive academic version was available for graduate student research. The basic steps needed to create the system included: • Identify and modify GIS base map. The GVRD created a digital map of municipalities and unincorporated "electoral areas," such as UBC. Additional layers were created to add labels for municipality names, subareas, local area maps and the location of major transportation infrastructure investments. All maps are oriented with true North at the top of the page. • Gather census, housing, travel data. The data parameters which were considered to have a potential impact on the policy areas studied were collected. The primary data sources for the GIS database were: • 1992 GVRD Travel Survey (Trip distances, vehicle ownership, mode splits, etc.); • 1991 100% and 20% census data for BC, published by municipality; • 1996 GVRD Key Facts, historical trend data used to validate other sources; • 1996 CMHC Rental and Housing Surveys, to determine rents and current price trends; • 1991 City of Vancouver Report on Selected Census Statistics. • Construct database and import to GIS. A database was constructed using an Excel spreadsheet with 42 data fields for each of the 25 municipalities that existed in the Lower Mainland in 1991. Jurisdictions appearing on the base map layer were combined with 15 parameters from the Excel database through the Maplnfo "Join" operation using the name of the municipality the database "key," or reference variable. Smaller jurisdictions with statistically insignificant populations were removed. • Determine correlations. The Maplnfo "Shade by Value" thematic map allows many relationships to be quickly determined and visualized. A knowledge of the problem domain to guide the scope of the queries was important, as the number of possible permutations and combinations of the various data fields is enormous. • Display using graphical overlays. Once correlations of interest had been determined, overlays were generated with class ranges that were considered to reflect the importance of the field values for government action. A greyscale, where black represented one extreme and white the other, was generally employed for simple maps. An example of an overlay is "Overlay a cross-hatch pattern of those areas where the average house price is less than four times the average GVRD household income on a map shaded by average distance to work." 1.6.2 Special Considerations As this thesis is "data intensive" and has a relatively large policy and geographic scope, this section will outline a number of considerations that academic readers may wish to keep in mind when reviewing the presented results: • Ecological Fallacy; • Survey Bias; • Data Validity; • Use of Data Triangulation; • Making Inferences on the General Population. 16 Ecological Fallacy Ecological Fallacy is the potential error of drawing conclusions on individuals or classes of individuals within a sample based on the characteristics of the entire sample. It is particularly important when aggregate data, such as the breakdown of incomes and work trips used frequently in this work, is applied to the behaviour of individual households, such as location decisions and travel habits.6 A similar related problem from the field of Geographic Information Systems (GIS) is the Modifiable Areal Unit Problem (MAUP) which relates to potential errors arising from the choice of geographic boundaries used for data collection. A good example is the use of municipal boundaries or subregions for the aggregation of most data used in this research. While housing prices are generally considered to be high in the City of Vancouver, these would be even more dramatic if the data was broken down into the east and west sides of the city. Similarly, the housing preferences of individual households can vary strongly within a municipality depending on factors such as income, age and tenure. There is also a certain amount of "noise," or interdependence of variables and cyclical fluctuations in the data. The number of housing starts may be closely related to the growth in population, but the influence of decreasing household sizes and economic uncertainty can distort this relationship. Many housing and transportation choices are a simple matter of personal taste or the effectiveness of marketing approaches, which are very hard to categorize. The varying mandates of data collecting agencies, combined with financial restrictions which limit the scope of data collection and sample sizes, are contributing factors. A simple lack of interest in these issues by the major data collection agencies, i.e., the federal and provincial governments, is a significant factor in the lack of complete data. Where possible, data which is incomplete, or subject to more than one interpretation, will be identified in the analysis. 6 Therese L. Baker Doing Social Research (New York: McGraw-Hill, 1988): pp. 99-100, "The Twin Traps: The Ecological Fallacy and Reductionism." 17 Survey Bias Household surveys used, such as the GVRD's GOMD telephone survey and those carried out by the developer in the Clover Valley Station subdivision, depend highly on the willingness of participants to provide responses, the design of the survey and the quality of the interviewer. The marketing representatives who conducted the interviews managed to survey most households in the subdivision, but did not push respondents to answer all questions and there are gaps in the survey data as a result. In a mailed survey, there would have been the risk that the profile of respondents would not have been the same as that of all households. Buyers with more time to spare, such as retirees, may have been more willing to provide more complete answers than busy young parents. Similarly, the key informant surveys seem to have been answered in greater depth by those with more time available (academics and mid-level government staff) than those working in management position. The profile of survey respondents in all cases was reviewed to confirm that a balanced and "representative response was in fact received. Data Validity A number of key informants approached for background information expressed concerns that the census and travel data used, although the best available (1991 and 1992 vintage respectively), had already reached their "best-before" dates. Nevertheless, ensuring that most of the data used for comparison purposes covered the same relative time periods allowed for consistent interpretation. Initial 1996 census data, recently released, has confirmed 1991 population trends. Essential data sources were examined to confirm that data definitions and collection methods were appropriate.7 For example, the Place of Residence versus Place of Work matrices from 1991 census data were essential components of the secondary data that was analyzed. Careful review of the 1991 census variable definitions and discussions with Statistics Canada analysts indicated that the coding of responses for the place of work sometimes required interpretation from data coders which may 7 Royce Singleton, Bruce C. Straits, Margaret M. Straits, Ronald McAllister Approaches to Social Research (New York: Oxford University Press, 1988), Chapter 12, Research Using Available Data, "Reliability and Validity: Authenticity and Accuracy," pp. 342-345. 18 have introduced some variations into the results. The data is nevertheless the most complete source of individual locational behaviour available, if only for two types of activity (live and work). The 1996 census data added travel mode as another variable. One hopes that more questions on travel behaviour and household lifestyles will be added to the census in the future. Sample sizes used for several regional surveys were also a major concern. In the case of the GOMD survey, some crosstabulations to identify trends within households with varying housing types, locations, and income classes will not be statistically significant. The survey of Clover Valley Station residents, who live together in houses of the same type and cost, would be more significant, but might not be applicable to residents of other neighbourhoods. Rather than excluding potentially significant results which involved small sample sizes, the actual data has been provided to allow readers to make their own judgments on statistical validity. Use of Data Triangulation Throughout the literature and in reference books, authors have made direct or indirect references to "triangulation," or the need to verify results by cross-checking information from one data source with others.8 Another common term for this technique is "multiple methods."9 This research approach relies heavily on using various data sources to "zero in" on the factors that influence the subject of interest in the study, in this case households and their locational influences. In addition to using different data sources for comparison, existing data sources were examined to identify information that could reinforce or refute the results. The integration of qualitative data, such as surveys using open-ended questions, combined with quantitative data, such as census data, helped to "plug the gaps."10 For example, most respondents in key informant interviews said that they intuitively knew that young households were moving out to the suburbs to find affordable housing, but could not 8 Ibid. A good introduction was found in Chapter 13, Multiple Methods, "Triangulation," pp. 360-362. 9 Julia Brannen (Ed.) Mixing Methods: Qualitative and Quantitative Research (Aldershot, England: Avebury, 1992): Chapter 3, Alan Bryman, "Quantitative and qualitative research: further reflections on their integration," p. 63. 1 0 Ibid, p. 60. 19 quantify how important the trend was. In the GVRD telephone survey on Ground-oriented Medium Density housing, factors such as "neighbourhood" and "schools" at first appear to be more important to most households than price in the locational decisions of residents. Examined more closely, price was much more important to younger households and closeness to work was indeed less important. Cross tabulations were performed where possible on available raw data to extract these subtle differences and provide independent confirmation of trends. "Macro" and "micro" level conclusions could generally be derived from the qualitative and quantitative data.11 Making Inferences on the General Population A search of the applicable literature produced a number of examples of the need to avoid generalized inferences from transportation and land use data. For example, research by Robert Cervero of the University of California (Berkeley) in the Bay Area of California indicated that there was the expected strong correlation between housing prices and long distance commuting. On closer examination, the direction of the commuting flow was found to be exactly the opposite to what is experienced in the GVRD.12 In the Bay Area, high housing prices in the suburbs are cited as the reason that many lower income employees in office parks are forced to commute back to the affordable, but crime-ridden, inner-city communities in Oakland. In the GVRD, of course, high housing prices in the Burrard Peninsula force many employees to commute from cheaper housing in the suburbs. It is important that general principles not be rigidly applied to case studies involving limited study areas or sample groups. 1 1 Ibid, p. 61. 1 2 Robert Cervero America's Suburban Centres: The Land Use-Transportation Link (Boston: Unwin Hyman, 1989), p. 48. 20 2. Review of Applicable Literature, Theory and Policies 2.1 Growth Management in Cascadia A number of urban areas in Oregon, Washington and British Columbia have been experiencing tremendous growth throughout the last few decades.13 While most people think first of the GVRD as the primary high growth region in B.C., Nanaimo, the Okanagan and the Victoria area have also been experiencing high growth rates.14 Regional districts and municipalities in B.C., until recently, have had few requirements to address growth management or to coordinate Official Community Plans (OCPs) with neighbouring municipalities. This has led to widespread cases of disjointed development, urban sprawl, and increased costs for infrastructure and community services in support of development.15 In an attempt to rectify this situation, the B.C. government legislated statutory changes in the 1995 Growth Strategies Statutes Amendments Act, often referred to as the Growth Strategies Act (GSA). Most of these changes were directed at the Municipal Act, requiring regional districts and municipalities to implement Regional Growth Strategies and Regional Context Statements.16 2.1.1 Historical Overview Throughout the recent literature on growth management in North America, authors refer to the legislation passed by the state of Oregon in 1973 as the landmark legislation to follow.17 Not only were these initiatives passed a full decade before those of any other U.S. jurisdiction, they are still used as a model for later legislation and are considered to be the most progressive and complete in North America.18 It is worth noting the historical context of this legislation. Oregon has a long 1 3 See, for example, 1996 Census Population Statistics. 1 4 Province of British Columbia, Ministry of Municipal Affairs Growth Strategies for the 1990s and Beyond (September 1994): p. 3. 1 5 Ibid. 1 6 Province of British Columbia Growth Strategies Statutes Amendments Act (Bill 11), 1995. 1 7 State of Oregon Land Conservation and Development Act (Senate Bill 100), 1973. 1 8 Jay M. Stein (ed.) Growth Management: The Planning Challenge for the 1990s (Newbury Park, California: Sage Publications, 1993): Chapter 1, "Growth Management and the Integrated Roles of State, Regional, and Local Governments", by John M. Degrove and Patricia M. Metzger. 21 history of land stewardship. Late in the last century, for example, the governor ordered that a wide strip along Oregon's entire coast outside of existing town sites to be preserved in perpetuity for use by the general public. The state's population doubled between 1950 and the early 1990s, a rate roughly double that of the U.S. as a whole. Many of the newcomers were Californians escaping from areas spoiled by rapid, uncontrolled development.19 Much of the growth was occurring in the Willamette Valley south of Portland, a valued rural escape for the metropolis and the state's most fertile agricultural area.20 Predating the legislation were decisions by the Oregon Supreme Court which recognized the precedence of comprehensive plans over local land use regulations.21 The "1000 Friends of Oregon," a citizen watchdog group formed to monitor the legislation, is often given credit for maintaining the strength and enforcement of the state law.22 British Columbia's Agricultural Land Reserve (ALR) legislation also originated in this period, for many of the same reasons but with a greatly reduced scope. The basic principles of the Oregon model are simple. The state defined a series of ascertainable growth management goals, such as affordable housing, shoreline protection and the preservation of farm land. Every city and county was required to prepare a plan consistent with these goals, as evaluated by a Land Conservation and Development Commission (LCDC). Well defined Urban Growth Boundaries (UGBs) were established, outside of which only controlled natural resource uses such as agriculture and renewable forestry were permitted. To offset the danger of increased urban housing prices due to a scarcity of available land, higher zoning densities were mandated. Oregon's legislation is highly distinctive in that: • Clear and detailed guidelines are presented for the interpretation of each of the nineteen goals; 1 9 Ibid. Chapter 4, "Growth Management in Oregon", by Deborah A. Howe. 2 0 Ibid. 2 1 Peter A. Buchsbaum and Larry J. Smith (eds.) State and Regional Comprehensive Planning: Implementing New Methods for Growth Management (Chicago: American Bar Association, 1993): Chapter 3, "Oregon Blazes a Trail," by Edward J. Sullivan. 2 2 John M. DeGrove Planning and Growth Management (Cambridge, Massachusetts: Lincoln Institute of Land Policy, 1992): Chapter 9, "Strengthening Metropolitan Governance in State Growth Management Systems." 22 • References are made to specific geographic areas considered to be of statewide importance; • An independent review commission examines the plans of subordinate jurisdictions for conformance with the statewide goals. As advanced as the Oregon legislation may seem, especially when compared with that of other jurisdictions, top-down growth management has nevertheless come under attack; from property rights activists and the development industry.23 Questions in a recent growth management survey prepared by Metro Portland's planning staff provide an indication that commitment to the UGB may be waning. The response which supports the gradual slippage of the UGB and placing apartments and townhouses only in specific areas seems to be the desired answer, being the only choice that does not have an "even if that means..." condition attached: 1. "The urban growth boundary should be held in place to preserve all of the land outside, even if that means increasing densities with more apartments and new houses on smaller lots. 2. The urban growth boundary is too tight and should be expanded to allow new houses to be built on the larger lots available five to ten years ago, even if that means that we have to develop significant amounts of land outside the UGB in the next 20 years. 3. The urban growth boundary should expand slightly and we should continue to encourage apartments and townhouses in specific areas with slightly smaller average housing lot sizes than we have now. " The State of Washington, which has a significantly denser freeway network and greater urban sprawl than Oregon, did not pass Growth Management Act (GMA) until 1990. While some of the concepts of Oregon's legislation are present, the UGBs defined are less stringent, allowing considerable room for business-as-usual style development for the foreseeable future, particularly in the Puget Sound area. It is interesting to note that two of Washington's fourteen goals are aimed more at protecting the rights of individuals than preventing urban sprawl:24 Editorial, "We need land-use rules," Salem Statesman Journal, 25 January 1990. State of Washington State Growth Management Act (RCW 36.70A.020), 1990. 23 • "Property Rights - Protect property from arbitrary or confiscatory actions; • Permits - Permits should be issued in a timely manner. " Washington's commitment to Growth Management is suspect. While the common call to reduce automobile dependence is present, the G M A states openly that alternative transportation modes should not be stressed outside of denser urban areas. A recent status report on the success of the G M A used a picture of a big box retailer in a small town to show how growth management was promoting business.25 Another picture from the same report shows a busy street, with a narrow sidewalk on only one side and an enormous parking lot on the other, as an example of a "walkable town centre." A large part of the $4 billion Puget Sound Ten-Year Transit System Plan is allocated to commuter rail, an expanded H O V network and only a token LRT hne while the improved local bus service needed to support higher density communities is overlooked. The 1992 Georgia Basin Initiative, a project of the British Columbia Round Table on the Environment and the Economy, sought to open a cross-border dialogue on sustainability issues and growth management.26 The Georgia Basin is roughly denned by the watersheds of the Strait of Georgia, Puget Sound, and the Strait of Juan de Fuca. The initiative resulted from a recognition that there were many social, environmental and economic values shared by the Province of British Columbia and the State of Washington that were being threatened by rapid growth rates. Many of the initiative's recommendations dealt with regional planning, compact community development and alternative transportation, and included: • Strongly encouraging urban containment, residential intensification and compact community development; • Developing provincial guidelines for urban settlement in the basin, building on experiences in B.C., Ontario, Washington and Oregon, which: 2 5 State of Washington, Department of Community, Trade and Economic Development Growth Management: It's Beginning to Take Shape (January 1997): pp. 12, 32. 2 6 British Columbia Round Table on the Environment and the Economy Georgia Basin Initiative: Creating a Sustainable Future (May 1993). 24 • Focus urban development within existing communities; • Provide for a range of housing (including affordable housing), community services and employment opportunities; • Minimize single occupancy vehicle use and encourage walking, cycling and the use of public transit. • Demonstrating leadership in transportation planning and management by: • Supporting T D M measures and user pay principles without inflicting undue hardship; • Shifting subsidies from private motor vehicles to encourage public transport; • Supporting the development and redevelopment of communities to reduce the need for vehicle transportation. • Recommending new models of governance which would integrate, at the regional level, comprehensive land use and transportation planning; • Executing a multi-faceted public education programs which increases awareness of the pressures that are imminently threatening sustainability and the quality of life in the basin, along with the need for coordinated action. The initiative has unfortunately been disbanded although some of its mandate has been absorbed by the Ministry of Municipal Affairs and Housing's Growth Management Office. 2.1.2 Overview of B. C.' s Growth Management Legislation Five years behind Washington and 22 years behind Oregon, B.C.'s GSA made statutory updates to the provinces Municipal Act which specify: • The intent and content of a Regional Growth Strategy; • The process for the development and ratification of a strategy; • The requirement for a region's member municipalities to address a strategy in their OCPs. 25 The focus of a strategy is to address a wide variety of the environmental and livability problems associated with uncontrolled urban development in the province. For example, strategies "should work towards" avoiding urban sprawl, preserving and creating open spaces for parks and recreation, reducing automobile use, promoting green modes of transportation, reducing pollution, and providing affordable housing.27 The plans to be developed must provide a 20 year strategy for the implementation of matters which are under the jurisdiction of a regional district. They should address the mechanisms by which a region will provide for projected needs in the areas of housing, transportation, economic development, parks, and regional services such as water and sewage.28 The process is initiated by a simple resolution of a regional district's governing body, a board of directors appointed from elected municipal officials. There is a requirement for extensive consultation with those "who the board considers will be affected," which includes the public, local governments (municipalities or other regional districts), first nations, and agencies of the provincial and federal governments.29 Several sections in the GSA detail how agreement on the plans is to be achieved, along with a complex arbitration procedure to be followed in the case of disagreements.30 The province, adjoining regional districts, and member municipalities must all be given the opportunity to review a plan before it is adopted by a regional board. Once a plan is adopted, there are several provisions in the GSA which outline the mechanisms by which the goals of the plan are to be implemented. The primary mechanism is the requirement that the OCPs of member municipalities include a Regional Context Statement which specifically states how the OCP supports the regional strategy and how the OCP will be made consistent with the strategy "over time."31 This time period is not specified. In theory, as municipal bylaws such as zoning must conform to a municipality's OCP, the goals of the growth strategy will eventually Province of British Columbia Growth Strategies Statutes Amendments Act (Bill 11), 1995: Section 942.11. Ibid. Section 942.12. Ibid. Section 942.17. Ibid. Sections 942.18, 942.19, 942.2, 942.21, 942.22, 942.23, 942.24. Ibid. Section 942.28. 26 be respected. The Act requires that an intergovernmental advisory committee be established to advise local governments on the strategy and to facilitate the coordination of provincial and local government activities.33 Local governments may enter into agreements for the coordination of activities relating to the implementation of a Regional Growth Strategy with other levels of governments and government agencies.34 Further details of the GSA will be presented in the sections on policy. 2.1.3 Initial Experience with B.C.'s GSA The Livable Region Strategy (LRS) was adopted in the fall of 1995 by the board of the GVRD as the regional district's official growth management plan. Although the strategy had been developed before the GSA came into effect, it was subsequently accepted by the provincial cabinet as a Regional Growth Strategy. Challenges to the strategy's policies appeared almost immediately at the local and provincial levels. The District of Langley 1995 Growth Management Plan recommended a growth rate of 3%, concentrated in designated areas. The actual rate of growth is 4-5% and the LRS set a target of only 1.5% for the district, citing a growing local jobs/housing imbalance as the reason for this limit.35 The new district council, which is strongly supported by the development industry and large property owners, dismissed their moderate district manager and director of planning immediately after being elected in November 1996 and rejected the District's growth management plan and the LRS in April 1997. New policies call for and end to phased development in semi-rural areas and support for dispersed growth throughout the district. An apparent loophole in the Act is that provisions contained in a Regional Growth Strategy only apply to communities that an OCP, which is not obligatory under section 944. In this case, under the added Section 942.33, the Minister of Urban Affairs may order a community to prepare an OCP within a defined time frame. 3 3 Province of British Columbia, Growth Strategies Statutes Amendments Act (Bill 11), 1995: Section 942.29. 3 4 Ibid. Section 942.3. 3 5 District of Langley, Growth and Planning Commission Growth Management Conclusions and Recommendations (November 1995). 27 The City of Richmond rejected provisions in the original LRS which limited growth in the municipality out of concerns for flooding as a result of earthquakes damaging protective dikes.36 The city threatened to "opt out" of the plan under GSA section 942.15, which states that a strategy may be adopted without a specific provision "on the basis that it is not binding on the jurisdiction of a local government that has refused to accept it," if the provision is "not essential to the Regional Growth Strategy."37 In the Regional District of Nanaimo, the 1997 PD3C award winning "Plan Nanaimo" applied the UGB concept to define a strip roughly twice as wide as the currently developed urban envelope.38 Most of the regional district's 80 km of coast is allocated to residential use under the plan, either as "rural residential" or urban areas. The plan makes only one reference to affordable housing, citing support for reducing actual construction costs as the mechanism to be employed. The province has indicated that, notwithstanding the intent of the Growth Strategies Act, it still has the final say in many regional matters, particularly in transportation. For example, the province has announced the construction of additional lanes to provide High-Occupancy Vehicle (HOV) capability on highways in the GVRD and has set the minimum occupancy of these lanes at two or more. This contradicts the GVRD's policy of converting existing lanes for H O V use wherever possible and requiring a minimum occupancy of three or more, in order to avoid the encouragement of Single Occupancy Vehicles (SOVs) by reducing overall congestion.39 The GSA requires a high degree of cooperation, consensus, and conciliation. Unlike Oregon's legislation, explicit compliance to the goals of the Act is not mandatory and there is no quasi-judicial body to provide an impartial review. No UGBs or density targets are provided by the province. Interpretation of the GSA is left to local regional districts and municipalities, which 3 6 Harold Munro "Richmond considers separation from the GVRD," Vancouver Sun, 28 October 1995. 3 7 Hugh Kellas of the GVRD Strategic Planning department has indicated that the provisions of the Act may be difficult to enforce, and that the primary motivation for implementation will continue to be moral-suation. 3 8 Development Services, Regional District of Nanaimo Growth Management Plan for the Regional District of Nanaimo (January 1997). The City of Surrey, a municipality synonymous with sprawl, subsequently won a national planning award for its new OCP. Such awards call into question the practice of rewarding plans without the need to demonstrate a reasonable track record of adhering to the plans. 3 9 Peter Boothroyd "Premier's incantation is so much black magic", Vancouver Sun, 14 September 1995. 28 often have considerably different views on the importance of, and acceptable approaches to, growth management. In the absence of public pressure to compel compliance local governments, the examples cited above indicate that the GSA may not be promoting the intended goals of the legislation.40 A fundamental difference between legislation in B.C. and that in Oregon and Washington is that there are currently no guidelines whatsoever for the interpretation of B.C.'s growth management goals, although the GSA makes a provision for these. The "matters" that Regional Growth Strategies "should work towards" are presented simply as a bulleted list in point form. The potential for inappropriate or inadequate responses to provincial goals is obvious and substantial, given that provincial approval of Regional Growth Strategies is not required. There are no implementation guidelines in the LRS to guide OCP Regional Context Statements in important matters such as transportation and housing, which has led to a great deal of frustration and expense in the development of new plans.41 New residential and "big-box" retail development in rural Abbotsford. Alan Artibise " Our new regional plan needs a plan - to police planners and politicians," Vancouver Sun, 09 November 1995. 4 1 The OCP for the unincorporated area that includes the UBC campus required a significant rewrite after the GVRD board rejected the draft OCP. Among other reasons for rejection, the board called for much higher levels of affordable housing and rental housing than existed in any other GVRD municipality at the time. 29 2.2 Land Use and Transportation 2.2.1 Overview While the question of the strength of land use and transportation interaction is often debated, the fact that there is a clear, symbiotic relationship will be assumed to be a fact in this thesis.42 This section presents examples of these interactions and reviews the social, environmental and economic impacts of past and present land use and transportation patterns. Some of the major land use and transportation planning tools used in various jurisdictions to shape the interactions and mitigate the impacts will then be identified. Finally, a discussion of the critical elements of land use density and mix that dictate many transportation and housing directions will be discussed. From the earliest towns and cities that sprung up along ancient trade routes to the early history of Vancouver's streetcar-driven development, transportation has manifested itself as a predominant factor in the ultimate shape of cities. Vancouver would perhaps not exist at all in its current form if the Canadian Pacific Railway (CPR) had not chosen it as the company's western terminus, with the line being completed in 1885. As Vancouver was being established as western Canada's major western seaport, Port Townsend and Seattle in the State of Washington competed for supremacy as the major seaport in the U.S. Northwest. Seattle was ultimately chosen as a railway terminus and has become a major metropolis while Port Townsend remains a quaint coastal town with well preserved architecture from the 1800s. Historically, the chronology of urban development patterns has depended strongly on both the health of the local economy and the transportation technology available at any given time to transcend distances. A 1993 SCARP thesis was devoted entirely to an authoritative literature review of land use and transportation interaction with applications to the GVRD. See Tony Parker Land Use and Automobile Dependence (UBC SCARP Masters Thesis, 1993). An earlier thesis looked at the impacts of two major transportation projects on the goals of the newly-emerging Livable Region Strategy. See Reginald Paul Faubert Coordination of Transportation and Land Use Planning: A Case Study of Greater Vancouver (UBC SCARP Masters Thesis, 1990). The conclusions of both studies are similar, noting that the causes of sprawl and automobile dependence are well documented, as is what works to promote effective land use and transportation. Both studies also conclude that the reluctance of decision makers, practicing professionals and the general public to make changes is pervasive. 30 In the bid-rent theories of land economics discussed later, the utility to businesses and individuals of accessibility to markets and goods has always been a major component of land values, for obvious reasons. Textbooks on urban economics usually devote large sections to the effects of transportation on the locational decisions of businesses and households.43 A case can be made for the proposition that, while transportation once clearly led in most major developments, the age of the automobile has now led to low-density, distributed development patterns. Of particular relevance in this thesis is the question of whether, given the increased mobility that automobiles provide, transportation still leads development or development now leads transportation in suburban areas. Discounting the important and growing influence of home-based businesses and telecommuting, the important questions to study are: • How important are transportation infrastructure investments in suburban municipalities in shaping rapidly growing metropolitan areas? • How much of this infrastructure exists to provide access "after the fact" to developments locating away from employment centres and established transportation infrastructure? Low-density suburban development and the automobile, as de facto standards in land use and transportation at the fringes of metropolitan regions, have exerted a strong influence on land uses in the rural and semi-rural parts of the region, which amounts to transportation access leading development. Resulting speculation and land-use conflicts have been shown to discourage agriculture near urban areas.44 Small developments in rural areas have had the effect of stalling necessary maintenance of farm infrastructure, as land owners wait for urban development to bring windfall profits for their properties.45 "Hobby farms" in the South and North Fraser subareas of the GVRD are excellent examples of this trend. These rural residential properties, zoned as five acre parcels, have proliferated in municipalities such as Langley District as wealthier urbanites See, for example, Edwin S. Mills and Bruce W. Hamilton Urban Economics (Harper Collins, 1989). 4 4 A. Nelson, "An Empirical Note on How Regional Urban Containment Policy Influences an Interaction Between Greenbelt and Exurban Land Markets," APA Journal, Vol. 54, Spring 1988. 4 5 B. Pond and M. Yeates, "Rural/Urban Land Conversion I: Estimating the Direct and Indirect Impacts," Urban Geography, Vol. 14, No. 4, 1993. 31 sought to escape to more pastoral settings. A wave of rezonings to low-density housing along the district's semi-rural north-western border had sent the clear signal that urbanization was on the way.46 The Langley Leadership Team (LLT) came to power in the district in 1996 with the strong support of developers and large property owners who sought similar rezoning along the south-western border, which has poor transportation access. Subsequently, in May 1997, Surrey's engineering department initiated a study of providing a private toll road to service leading to this area, in clear contradiction to the GVRD's Transport 2021 plan. In the Geography of Nowhere.47 James Kunstler's descriptions of suburban development patterns provides strong evidence that automobile mobility is changing land use patterns in many ways, often creating a vicious circle. For example, requirements for large amounts of parking in most new developments, combined with low levels of pedestrian amenities, has made shopping and working in suburban areas very convenient. This draws businesses away from the established urban cores or runs them out of business completely, making suburban shopping increasingly the only viable shopping option for most. Wide right-of-ways (ROWs) in suburban areas allow for future road widenings. As traffic levels and distances between amenities increase, few people wish to travel by foot or bike. With little need to accommodate pedestrians or cyclists, land use becomes increasingly automobile-oriented on the urban fringe. In Robert Cervero's seminal work on suburban centres in the U.S. and in other works on housing market analysis, land use and transportation equations can be skewed by several factors specific to the American context. Inner city crime, racial segregation, poverty and the presence of a federally funded freeway system which overlays most major metropolitan areas are most often cited.48 These factors are certainly present in the Canadian context, but do not feature as prominently in the literature. Another set of factors may be more applicable in the Canadian context. For instance, the impacts of growing concentrations of households of Chinese origin in Richmond and 4 6 During analysis of Place of Work data for this thesis, a surprisingly large subgroup of households earning more than $80,000 had moved to Surrey and Langley District in the five years before 1991. 4 7 James Howard Kunstler The Geography of Nowhere (New York: Simon and Schuster, 1993). 4 8 See, for example, Henry O. Pollakowski Urban Housing Markets and Residential Location (Lexington, Massachusetts: Lexington Books, 1982) or G. Thomas Kingsley and Margery Austin Turner (eds.) Housing Markets and Residential Mobility (Washington: The Urban Institute Press, 1993). 32 households of South Asian origin in Surrey on regional travel patterns would certainly merit further research. The land use shaping effects of freeways and major road projects in metropolitan areas cannot be understated. While the seeds had been sown earlier in the century, innumerable examples of low-density suburban communities sprung up around the United States after the Interstate Freeway system was constructed.49 A recent road map of Metro Toronto shows that the 401 expressway, intended in 1961 to be a four lane bypass highway on the rural outskirts of the city, is now a 16 lane expressway that lies almost exactly at the mid-point of urban development in the Greater Toronto Area (GTA). Not afraid of repeating enormous past mistakes, the province recently fast-tracked the completion of the 407 expressway through productive farmland just north of Toronto as another "bypass" solution. Support for public transit was cut at the same time in a close parallel to decisions currently being made in B . C . 5 0 On a recent visit to the GTA, the 401 corridor was already lined with "Commercial building for lease" signs and the 407 corridor was lined with "Commercial land for sale" signs, a remarkable indication of the speed at which market forces react to changes in the cost and accessibility of a cheaper supply of developable land.51 Toronto's 401 Expressway, completed in 1961 Toronto's recently completed 407 Expressway James Howard Kunstler The Geography of Nowhere (New York: Simon and Schuster, 1993) p. 107 - "(The 1956 Interstate Highway Act...) called for 41,000 miles of new expressways, with the federal government picking up 90 percent of the tab and the states 10 percent. The bill also subsidized the improvement (read, widening) of innumerable ordinary local roads to facilitate further urban sprawl." 5 0 Ontario ordered the Toronto Transit Commission (TTC) in 1996 to increase fare-box revenue from 70% of all costs, a level far above that of any other North American jurisdiction, to 75%, resulting in serious service cuts. 5 1 On the same visit, strip residential development on rural roads up to thirty kilometers from the urban fringe, with three and four garage houses becoming the standard for new developments. 33 2.2.2 Impacts and Costs Many sections could be filled with details of the impacts of transportation and land use patterns on communities. The list of the impacts of transportation infrastructure is lengthy and the monetized costs can be enormous. These costs are generally borne by people other than drivers and land owners:52 "As a rule, the environmental impacts (of transportation) fall on parties other than the provider and consumer of transportation services. Consequently, the impacts are also called externalities, and their costs are termed external costs. When environmental externalities benefit third parties they are desirable, but, with a few negligible exceptions, external impacts of transportation on the environment are negative. " Similarly, the negative impacts of land use are borne by people other that land owners:53 "Land use changes caused by transportation create a wealth of benefits, but they impose environmental and social liabilities at the same time. In general, the benefits tend to be captured by transportation users and land owners, while society as a whole bears many of the costs. A good example of a land use change which provides benefits to the user but imposes environmental and other external costs on society is urban sprawl. " A review of some of the social, environmental, and economic impacts of urban transportation and land use in metropolitan areas with high housing costs is useful. Peter Bein, Chris Johnstone and Todd Litman Monetization of Environmental Impacts of Roads (Planning Services Branch, British Columbia Ministry of Transportation and Highways, 1995) Executive Summary. 5 3 Ibid. Chapter 4, Environmental Impacts. 34 Social Negative social impacts of automobile dependence and urban sprawl include: • Longer commuting distances and increased traffic congestion; • Longer distances from home to services such as child care and medical centres; • Less green space or less accessibility to available green space; • Pollution-linked health problems; • Reduced affordability of housing in desired locations; • Stress on the provision of fire and police services due to increased distances; • Reduced sense of community, loss of neighbourhood character and cohesion. An example of one of the more subtle and increasingly costly impacts has been noise. In Vancouver, noise levels are doubling in intensity every six years, and the major cause of this increase has been due to the growth of motorized vehicle traffic.55 Besides lowering property values and reducing the quality of life for people not sitting inside the vehicles, increases in traffic noise have led to increases in hearing impairments, nervous tension and sleep disorders. The social impacts of land use-induced losses in housing affordability are central to this thesis. Nationally, home ownership rates for the lowest two income quintiles have substantially eroded in the decade between 1980 and 1990.56 The percentage of renters in the Vancouver C M A who can afford a starter home is the lowest in Canada, after Victoria.57 This rate is even lower for those aged between 20 and 40. Home ownership in the lowest income quintile dropped dramatically from 1/2 to 1/4 of these households in the decade between 1980 and 1990. Under federal regulations adopted in 1992, 250,000 households purchased CMHC insured homes with only a 5% down payment and 270,000 individuals have withdrawn $2.5 billion from RRSPs for down 5 4 Fraser Basin Management Program for the B.C. Ministry of Municipal Affairs Navigating for Sustainability: A Guide for Local Government Decision Makers (1995): p. 58. 5 5 City of Vancouver City Noise: Report of the Urban Noise Task Force (April 1997). 5 6 CMHC Habitat II Canadian National Report (1996): p. 18. 5 7 BC Ministry of Municipal Affairs and Housing British Columbia's Strategy for Affordable Housing (April 1996): p. 15, 17. 35 payments on new homes. The willingness to accept larger monthly interest payments and lower RRSP savings in order to own a home, combined with an increasing number of low-income renters, is a concern for the long-term social stability of society. In the former case, a growing segment of society will be less able to make contributions to their pension plans, and will depend more on home equity for retirement security. In the latter case, a major segment of society may not be able to build up equity in a home for retirement security at a time when the number of younger taxpayers available to provide them with financial assistance will be at a minimum.58 The presumed need to live in distant suburbs to attain ground-oriented housing, in addition to demanding an inordinate amount of commuting time, has created the general requirement for more than one car. According to the BC Automobile Association (BCAA), the average annual cost for operating a mid-sized car is approximately $6,000. This is equivalent to a $500 monthly expenditure, which would be the same as the payment on a $60,000 mortgage at an interest rate of 8% amortized over a 20 year period. The GVRD estimates that taxpayers subsidize each car on the road by over $2,600. This adds up to over $2.75 billion dollars per year in subsidies for GVRD car owners from general revenues. This sum is equal to the Insurance Corporation of B.C.'s (ICBC) annual budget and is greater than the combined annual budgets of the BC Ministry of Transportation and Highways (MoTH) , B C Transit and the GVRD. If the results of the GVRD study on Ground-Oriented, Medium-Density (GOMD) housing described later are valid, common knowledge of the financial impact of automobile ownership could inspire many households to choose a townhouse in the Burrard Peninsula over a detached house in the Fraser Valley. This assumes, of course, that adequate zoned land was available for such housing and that transit would be made available to remove the perceived need for at least one automobile per household. Unfortunately, in the current political and social climate, both of these assumptions are tenuous at best. If households owned one less car and lived closer to employment, there would be considerable time savings realized through shorter commutes. This According to demographic forecasts in David Baxter Homes in Metropolitan Vancouver's Future: Housing Demand by Structure Type, 1996 to 2021 (August 1996): p. 8, the number of additional people in under-45 age groups will be at an absolute minimum in the year 2010, at which time three times as many people in the 45+ age groups will be added to the Vancouver CMA. 36 would reduce stress levels and increase the time available for family and community activities in addition to increasing the money available for other discretionary items. Environmental The environmental impact of urban development in the Fraser Valley has been pervasive. As much as 70% of wetlands in the lower Fraser Valley have already been lost to urban development.59 A short list of negative environmental impacts of urban sprawl includes:60 • Loss of productive agricultural and forest land; • Degradation of natural habitat with loss of plant and animal species; • Increased energy consumption; • Poor water quality and quantity due to the inability of water treatment and provision systems to keep up with the pace of development; • Poor air quality due to excessive reliance on motor vehicles. Individual low-density developments, such as residential subdivisions and "business parks," when considered alone, do not require much incremental transportation infrastructure compared to the overall system capacity. The cumulative impact on the environment of low-density developments at the urban fringe can be overwhelming and lead to demands for road improvements such as has occurred in the Northeast Sector, i.e., development leading transportation. This lack of consideration for the cumulative effects of development is evident in the B.C. Environmental Assessment Act (BCEAA). While requiring that an Environmental Impact Assessment (EIA) be made on all LRT projects longer than 8 kilometres and highway projects longer than 20 kilometers, the B C E A A requires no EIA for the hundreds of kilometers of residential and arterial 5 9 B.C. Ministry of Environment, Lands and Parks Stewardship of the Water of British Columbia: A Vision for New Water Management Policy and Legislation (1993). 6 0 Fraser Basin Management Program for the B.C. Ministry of Municipal Affairs Navigating for Sustainability: A guide for Local Government Decision Makers (1995): p. 59. 37 streets that will be constructed in the next decade to service new developments. There is no requirement for an EIA of residential, commercial or any other non-industrial development, regardless of the size of the land area involved or the projected number of people who will need to access the development. This is in stark contrast to Washington State's G M A legislation, which requires a complete Environmental Impact Statement (EIS) of a wide range of urban development factors, including Land Use, Transportation and Housing. The EIS for the City of Seattle's master plan runs to several hundred pages.62 The existence of such an EIS does not guarantee that unsustainable elements of development will be avoided, as most planners familiar with Seattle could testify, but at least ensures that some of the "big picture" impacts of land use and transportation are considered at some point in the planning process. Economic The costs of providing infrastructure to service new low-density development has been a major motivation for pursuing growth management in Cascadia. As early as the 1970s, even the real estate industry realized that the costs of low-density, automobile-dependent suburban development placed inordinate financial burdens on individuals and governments, particularly at the local level, stating that these are:63 "...the most expensive form of residential development in terms of economic costs, environmental costs, natural resource consumption and many types of personal costs. This cost difference is particularly significant for that proportion of total costs which is likely to be borne by local governments." Province of British Columbia, Environmental Assessment Office Guide to the British Columbia Environmental Assessment Process (June 1995), Reviewable Projects Regulation, Part 7, Transportation Projects. 6 2 City of Seattle, Planning Department Towards a Sustainable Seattle: Final Environmental Impact Statement for the City of Seattle's Comprehensive Plan (March 1994). 6 3 From Mark Roseland Toward Sustainable Communities: A Resource Book for Municipal and Local Governments (1992) quoting Real Estate Research Corporation The Costs of Sprawl, Volume 1: Detailed Cost Analysis (Washington: U.S. Government Printing Office, 1974). 38 The time involved in commuting to affordable housing has been identified as a growing cause of worker stress and loss of productivity.64 Congestion delays caused by increasing volumes of commuter traffic is also causing hundreds of millions of dollars in losses to Lower Mainland businesses, which results in job losses and a decrease in the region's competitiveness. Congestion losses to business in the Greater Toronto Area (GTA) were estimated in 1991 to exceed one billion dollars per year.65 A value of the order of half a billion dollars, or $250/year for every resident, might therefore be a reasonable estimate for Metropolitan Vancouver, given that, while smaller, Greater Vancouver has much lower transit ridership and a less developed road network than exists in the GTA. 2.2.3 Land Use Tools The transportation tools discussed in the next section influence how many trips people make and which mode they choose. While these tools are important, they have the appearance of "add-ons' which mitigate the problems caused by a failure of provincial, regional and municipal planning agencies to implement effective land use policies and coordinate these with transportation planning. This section reviews land use initiatives that have been used to avoid the need to apply such mitigation measures in metropolitan regions. A list of land use measures compiled by Berkeley's Robert Cervero to reduce sprawl and travel requirements in suburban areas is comprehensive and can broken down into several broad policy areas as summarized below:66 • Regional government responses: • Tax base sharing. In Minneapolis-St. Paul, over one-quarter of the region's tax base is shared. Sharing the metropolitan tax base reduces the temptation for one 6 4 GVRD/MoTH Transport 2021 Technical Report 16: Economic Development Perspectives on Transportation Planning (October 1993): p. 8. 6 5 IBI Group for the Province of Ontario Greater Toronto Area Urban Concepts Study (Toronto, 1990). 6 6 Based on Robert Cervero America's Urban Centres: The Land Use-Transportation Link (Boston: Unwin Hyman, 1989): Chapter 8, "Linking land use and transportation, Overview of research findings." 39 municipality to make land use decisions which improve the local tax base while imposing negative impacts on other municipalities in the region. • Fair-share housing requirements. New Jersey and Oregon have implemented affordable housing quotas on municipalities to provide more opportunities for affordable housing near workplaces. Municipal regulatory responses: • Traditional zoning. Old-fashioned zoning to increase densities and allow a mix of uses provides the basic requirements for the provision of lower land costs and provide a range of services and employment opportunities close to where people live, reducing the need to travel. • Performance standard zoning. In the usual variant, mixed uses are promoted, provided that these uses do not cause a nuisance to other uses according to a set of pre-defined criteria. Cervero has extended this to include vehicular traffic as a nuisance which should be included in the criteria. Similar to legislated T D M measures, these criteria include minimum targets for walk/bike/transit modes and maximum number of trips by other modes. This approach has been used in Toronto, Bellevue (Washington) and Fort Collins (Colorado). • Inclusionary zoning. One way of looking at this type of zoning is the removal of existing regulatory barriers to mixed use and a variety of housing types. Another approach is to prevent exclusionary practices through an independent, third party review of zoning practices. Oregon's L C D C reviews all community plans and rejects those that are considered to be exclusionary. • Conditional or incentive zoning. Density bonuses or the relaxation of some regulations can be granted to developers who make their projects more pedestrian, bicycle or transit friendly, who promote mixed use, or who implement T D M programs for their clients. • Density bonuses. Density has been identified as perhaps the single most important requirement for the viability of alternative transportation and the availability of affordable housing. An entire section on density and mixed land use considerations appears later. 40 • Density transfer and zoning swaps. These paper tricks allow developers to take density from one site and transfer it to another to provide a higher density nodes which are more supportive of mixed uses and alternative transportation modes. Vancouver uses this tool, particularly for heritage preservation. • Increased awareness of site design factors. Often overlooked is how the physical layout of a development supports objectives such as alternative transportation and affordable, higher-density housing. This can range from the street layout, the location of services, the orientation of buildings and the exposure to sunlight. A significant component of site design is the accessibility of the site by walking, cycling and transit. • Promotion of parking reductions. The supply of parking, although a T D M measure in some ways, is also a land use issue. An excess of parking increases the cost of developments and reduces the appeal of alternative modes. One study showed that the likelihood of workers choosing transit over driving is much more related to the cost and supply of parking than higher transit service levels or lower transit fares.67 Moore and Thorsnes, in an American Planning Association (APA) report, recommend a less regulatory, more economic approach to influence land use patterns.68 Their suggested tools are summarized below: • Pricing commercial externalities. Similar to performance standards, businesses could be taxed not only on the value of their property, but also on community impacts such as increased client traffic, truck deliveries and visual encroachment. Such a concept would promote smaller businesses which support complete communities and strongly discourage the proliferation of big-box retailers such as Costco and the Home Depot. • Pricing neighbourhood externalities. Neighbourhoods and municipalities which resist development that supports complete, compact communities, such as local businesses, infill John Meyer and Jose Gomez-Ibanez Autos. Transit and Cities (Cambridge, Massachusetts: Harvard University Press, 1981). 6 8 Terry Moore and Paul Thorsnes The Transportation and Land Use Connection (American Planning Association, Planning Advisory Service Report Number 448/449, 1994) p. 60, in discounting non-economic approaches to alternative land use state simply "Zoning is a blunt regulatory instrument relative to pricing." 41 housing, or rental apartments, would be assessed a "homogeneity" tax by the municipality or regional government. This tax attempts to reflect how low-density or single-use design in a community forces people to travel, which impacts other neighbourhoods. This measure could also be viewed as a financial penalty on exclusionary zoning practices. Such taxes would likely have to be assessed according to property values in order to have the maximum effect, as neighbourhoods with higher property values seem to be the most resistant to land use changes. • Removing public service subsidies. The practice of subsidizing extensions of water and sewers supports lower density development, while user-pay systems encourage compactness. Development Cost Charges (DCCs) attempt to implement user-pay principles in the GVRD, although the equity of the various charge schedules is not always clear. Initiating "growth must pay for growth" policies sometimes amounts to asking new home owners to pay twice for their services, as they are required to pay for the full cost of their own services while also helping to pay off services used by the community's long-term residents. 2.2.4 Transportation Tools This section looks at how engineering professionals have typically address the transportation problems that are often induced by land-use practices. Needless to say, planning professional must simultaneously mitigate land use problems that are induced by transportation practices. The literature overflows with examples of concrete projects and trip reduction programs aimed alternatively at reducing congestion by adding vehicle capacity or reducing the number and length of trips by Single Occupancy Vehicles (SOV). Both supply-side and demand-side approaches appear to stem from a systematic application of Social Cost Benefit Analysis (SCBA). In SCBA, costs such as land acquisition, construction and environmental degradation are balanced against the benefits, which are usually the initial time savings for drivers and presumed environmental improvements from reduced congestion. Long-term congestion and induced congestion effects are externalized. The "Social" aspect of the analysis looks at general, non-monetized people factors such as neighbourhoods traffic impacts. 42 Solutions offered by transportation planners and engineers range from brute-force expansion of capacity through building new roads, widening existing roads and adding turn bays to more elegant solutions such as adding High Occupancy Vehicle (HOV) lanes, computerized Intelligent Transportation Systems (ITS) and commuter rail. These solutions are usually supported by demand forecasts generated by computer models, which discount the push-pull effects of housing prices on trip generation and travel distances. In spite of some remarkable successes that have been observed in maintaining traffic speeds,69 adding capacity is increasingly seen as a solution with diminishing returns.70 Appendix A contains a more detailed overview of transportation planning methods currently in use. While it would be refreshing to say that support for "let's build our way out of it" approach had finally been put to rest, recent examples prove otherwise. A study of the Puget Sound region in the State of Washington has shown that average automobile occupancy has decreased on several freeways since Seattle's massive H O V program was constructed.71 Both B.C.'s South Coast Transportation Systems Plan of 1995 and Central Puget Sound Regional'Transit Authority's Ten Year Regional Transit System Plan of 1996 rely heavily on supply-side solutions to congestion problems, such as new H O V lanes, Intelligent Transportation Systems (ITS) and commuter rail. Transportation Demand Management (TDM) programs, on the other hand, can be thought of as "demand-side" initiatives which keep people from making unnecessary car trips, thereby reducing the need for additional road infrastructure in the first place. There are strong analogies to the In Roland L Mitchelson and James S. Fisher "Long Distance Commuting and Income Changes in the Towns of Upstate New York," Economic Geography, Vol. 63, No. 1, 1987: pp. 48-65, it is noted that, while the average commuting time in New York state remained at 23 minutes between 1960 and 1980, the average commuting distance increased from 8 kilometers to almost 20 kilometers. Higher incomes, tax incentives, and highway investments are cited as the causes of the increased distances. The City of Vancouver has managed to maintain a peak-period travel time of 25 minutes between the CBD and Richmond since the 1960s. 7 0 Terry Moore and Paul Thorsnes The Transportation and Land Use Connection (American Planning Association, Planning Advisory Service Report Number 448/449, 1994) p. 3, states the situation bluntly: "The obvious solution, more highway capacity, has reached a point of greatly diminished marginal returns in developed metropolitan areas. Not only are cities running out of space for new lanes, but the continued addition of highway capacity may be paving the way to a larger, less-treatable gridlock." 7 1 Puget Sound Regional Council Puget Sound Trends, No. T3, October 1995, "Average Vehicle Occupancy (AVO) on the Region's Highways." 43 electrical energy sector, which has arguably been more proactive and aggressive in applying the conservation approach. Energy conservation has been prompted in large part by campaigns from environmental movements fighting nuclear plants and hydro dams. Although the journey to work now accounts for less than a quarter of all trips, most trip reduction programs, such as the obligatory employer-based programs in California and Washington, focus only the work trip.7 2 Voluntary programs, such as the GVRD's "Go Green" programs, are based on "moral-suasion." In a summary report from the U.S., seven distinct T D M strategies were identified:73 • Region-wide rideshare agencies. Voluntary programs to match drivers and passengers. • Developer requirements. Conditions in a development permit which limit the amount of traffic that a project can generate. This is similar to applying zoning-style performance standards to transportation access at a site. • Transportation fees. Payment required from developers to fund alternative modes. • Incentive ordinances (by-laws). Reduced parking requirements in exchange for payment-in-lieu or provision of measures to promote access to the development by alternative modes. • Transportation Management Organizations (TMOs). These are groups of employers and developers who form a non-profit organization to reduce traffic and support alternative modes. Participant benefit from reduced parking needs and local congestion. The Cambie Corridor Consortium is a good example of such a group now operating in Vancouver. • Employer rideshare legislation. Local requirements for employers to reduce SOV trips. • Comprehensive TDM legislation. Regionally administered requirements for major employers to implement and monitor SOV reduction programs, often linked with financial penalties for failure to achieve assigned targets. In GVRD Overview of TDM Research (November 1995): p. 17, 3/4 of all trips are not work related and are made in off-peak hours. Distance to work is nevertheless a heavily weighted factor in transportation and location decisions. This could be due to the psychological impact of rush-hour traffic, or a lack of awareness by households as to how much time they spend traveling to non-work destinations. 1 3 Carolyn P. Flynn and Lawrence Jesse Glazer, "Ten Cities' Strategies for Transportation Demand Management," Transportation Research Record 1212, pp. 11-23. 44 Of these strategies, regional ridesharing and developer requirements appear to be the most effective, TMOs and transportation fees have had some success, and the other strategies have had mixed success. Most U.S. strategies are aimed at "carrot" approaches to controlling SOV use and avoid "stick" measures such as parking charges, gas taxes, tolls and road pricing, which form a large part of the GVRD's proposed T D M strategy.74 Simulations using the GVRD EMME/2 transportation modeling program indicates that with a "dramatic" application of a package of carrots and sticks, such as the provision of widespread bus priority and the tripling of parking charges, are seen to have much greater impact than the sum of individual measures applied in isolation.75 This theoretical result of combining strong measures has been validated by empirical studies, notably in Singapore where transportation and land use is highly coordinated, particularly in the provision of higher density housing near transportation infrastructure.76 While supporting the combined package approach, the GVRD is nevertheless leaning towards delaying the implementation of sticks until the carrots have been implemented.77 It is important to note that the results of GVRD computer simulations of the complete package of proposed T D M measures, which showed a much more significant shift in modal split than any individual measure, effectively applies all of the measures together and not in a phased-in manner. T D M measures which reduce demand for SOV travel by reallocating existing road space for bike, bus or H O V lanes are rarely mentioned in the North American literature.78 The "take-a-lane" approach is common in Europe, perhaps due to the narrower ROWs which preclude street 7 4 GVRD/MoTH Transport 2021: A Medium-range Transportation Plan for Greater Vancouver, (October 1993): p. 27, Section 2.1, Medium-range Demand Management Package. 7 5 GVRD/MoTH Transport 2021 Technical Report 3: A Transportation Demand Management: A Forecast Modeling Approach (February 1993) Appendix A, TDM Impact on Screenline Demand. 7 6 V Setty Pendakur, Gopinath Menon and Joseph Yee, "TSM Innovations from Singapore, Lessons from Experience: 1974-1988," Transportation Research Board Annual Meetings, January 22-26, 1989. 7 7 GVRD/MOTH A Medium-range Transportation Plan for Greater Vancouver, (Transport 2021 Report, October 1993): pp. 30-32. Section 2.4, Staging Approach for Demand Management, calls for all carrots to be applied before any sticks are used. 7 8 e.g., U.S. Transportation Research Board National Cooperative Highway Research Program Synthesis 185: Preferential Lane Treatments for High-Occupancy Vehicles: A Synthesis of Highway Practice (Washington: National Academy Press, 1993). In the single paragraph devoted to non-engineering aspects of traffic problems, the report states "Federal, state or provincial, and regional policies to promote clean air and reduce pollution represents another reason why HOV facilities are pursued. Concerns for energy and its efficient use have motivated enhanced HOV lane consideration, including take-a-lane approaches. During 1991 for example, converting an existing mixed-flow lane to HOV use was seriously considered in Seattle, northern New Jersey, and California." No lanes were actually taken from general purpose traffic for HOV use. 45 widening. Both the GVRD's Transport 2021 and Vancouver's CityPlan support this reallocation approach to T D M in principle. In practice, system capacity has been increased to accommodate alternative modes to the SOV. 7 9 The need for regional coordination of land use and transportation, as well as the application of T D M carrots and sticks, has been raised by several authors. In terms of temporal coordination, the phased approach appears to be chosen all too often for expedience, such as availability of funding sources.80 Fear of public backlash to rapid or sweeping changes is likely a factor. More straightforward programs, such as making automobile costs more closely related to distances travelled or reallocating currently externalized costs of automobiles from drivers and applying these funds to alternative modes, have received remarkably little attention in the transportation engineering literature or support from decision makers. Economists and environmentalists have taken the lead in examining the substantial externalized costs of road infrastructure, the subsidization of driving and the potential for road transportation investments to exacerbate traffic and congestion.81 This polarization of thought indicates the need for an expanded dialogue between engineers, land use planners, economists, environmentalists and policy analysts at all levels. 2.2.5 Jobs/Housing Balance All other things being equal, if the number of employed residents in a community matched the number of jobs, and the skills profile of the residents matched what was required by the GVRD/MoTH A Medium-range Transportation Plan for Greater Vancouver, (Transport 2021 Report, October 1993): p. 47, City of Vancouver CityPlan: Directions for Vancouver (1995): p. 32. 8 0 Strategic Planning Department, GVRD "Transportation Governance and Funding Workshop Notes," May 1997, confirms that while road projects are usually funded from general revenues, major transit projects are usually debt-financed. The report also notes that transit funding in particular "has often been reverse-engineered to fit the available funds rather than being based on substantive principles." 8 1 See, for example Richard Arnott and Kenneth Small, "The Economics of Traffic Congestion," in American Scientist, Volume 82, September 1994, pp. 446-455 or James J. MacKenzie, Roger C. Dower and Donald D.T. Chen, "The Going Rate: What it Really Costs to Drive," World Resources Institute Report, June 1992. 46 employers, there would presumably be less need to travel routinely outside of a community for common activities. This would also reduce the need for applying stop-gap solutions to keep traffic moving and accommodate development pressures. As will be seen later in the section on land economics and factors influencing housing location, such a balance is far from the case in most metropolitan regions, and some would say that achieving even a rough balance is not a realistic goal.82 Nevertheless, a goal of working towards a balance of jobs and housing on a subarea basis could create a more positive planning mindset than the attitude of "there's not much we can do, people are going to do whatever they want." A reasonable balance of jobs and housing at the local level, combined with minimal rates of in- and out-commuting, is a necessary, if not sufficient, condition for creating complete communities. The terms "local," "community" and "complete," although widely used by planners, are often used in different ways. The glossary of terms used by the American Society of Planning Officials defines none of these terms. The BC Municipal Act, GSA and GVRD Livable Region Strategy also avoid any clear, ascertainable definitions. A formal definition of "community" is "A locality inhabited by a group of people who share a government and often have a common cultural or historical heritage."83 The most applicable formal definition of "local" is "Pertaining to a city, town or small district rather than an entire state or country, e.g. local transportation."84 To overcome this lack of precision, the terms local and community in this thesis refer to areas roughly the size of the subareas defined earlier. These are areas large enough, if properly designed, to provide large amounts of housing and employment and yet compact enough to be conveniently accessed by walking, cycling or transit. The Northeast Sector, Richmond, Buraaby/New Westminster or the City of Vancouver would be examples of such "local" areas or "communities." The corollary term "complete" will mean providing a wide range of jobs, services, amenities, affordable housing and alternative transportation opportunities that satisfy most of the routine needs of local residents, most of the time. See research findings section on key informant survey. Webster's Dictionary s.v. "conununity." Ibid. s.v. "local." 47 Robert Cervero has also written extensively on the effects of imbalances between jobs and housing on transportation patterns.85 He concludes that imbalances in suburban areas have been driven primarily by ad hoc market forces and municipalities that make decisions with little regard for regional consequences.86 Even in the presence of regional plans developed through consensus, such as the LRS, municipalities often disagree with their designated roles and find subtle or overt ways to subvert the process. An example of such subversion was the District of Langley informing the GVRD in 1995 that it intended to allow growth at a 3% annual rate, "provided that Langley can achieve a balance between labour force and employment at this growth rate."87 The District claimed that this rate was consistent with the LRS, although the GVRD had limited growth in the District at 1.5% to prevent residential sprawl. The District's report stated that the GVRD should be notified that this growth was supported by the public, presumably based on the strength of a 1995 survey in which 80% agreed with the statement: "Langley is part of the larger metropolitan region and should accept its fair share of new residential growth."88 63% of respondents in the same survey had also indicated that they wanted no growth or slow growth in response to the question: "The District has been growing at 4% to 5% each year for the last 20 years. What growth option is best for Langley's future?" Municipalities appear to understand the importance of seeking a balance of jobs and housing, but are often undeterred from their chosen course of action when there is little likelihood of this happening, such as is being observed in the GVRD's outer suburbs and exurbs. Cervero describes five economic and demographic forces that exert strong influences on where jobs and housing are located: Robert Cervero "Jobs-Housing Balancing and Regional Mobility," APA Journal, Spring 1989: pp. 136-150 and Robert Cervero America's Suburban Centres: The Land Use-Transportation Link (Boston: Unwin Hyman, 1989). 8 6 Ibid. p. 48. 8 1 District of Langley, Growth and Planning Commission Growth Management: Conclusions and Recommendations (November 1995): p. 3. 8 8 District of Langley, Growth and Planning Commission Growth Management: Public Input, Community Development Survey (1995). 48 • Fiscal Zoning. Municipalities seek out, or at least tolerate, commercial and industrial uses for land due to the higher tax revenue generating potential of these activities. The Vancouver CBD is an example of this. High tax revenues from businesses help to cross-subsidize property taxes in the lower density residential areas of a city. Agglomeration economies such as the CBD may be important for the overall economic health of the region, but the impacts of commuting should be mitigated by providing a range of suitable housing for employees near these employment concentrations. • Growth Restrictions. This is often synonymous with exclusionary zoning. Simple restrictions on density, resulting in high land costs, can effectively preclude most new housing initiatives. • Worker Earnings/Household Cost Mismatches. In this interesting example of the differences between American and Canadian land use patterns mentioned earlier, Cervero uses the example of how suburban homes are generally more expensive than inner city properties, forcing lower paid workers in low density suburban commercial parks to commute into the city for cheaper housing. While the GVRD exhibits some of these trends, such as lower paid high-tech and service employees working in Richmond commuting to abundant apartments in the Burrard peninsula, the trend in ground-oriented housing is decidedly in the opposite direction. • Two Wage-Earner Households. With the continued trend towards working couples, there is a tendency to locate the household residence somewhere between the two workplaces.89 • Job Turnover. Employees are changing jobs more frequently, leading to a higher probability that at least one wage-earner will be employed in a municipality other than the Place of Residence municipality. A number of causes are identified, such as the growing trend towards short-term contract work and decreasing long-term loyalty between employers and employees. Moore and Thorsnes claim in their APA report that most commutes to work, even in places like Los Angeles, take less than 30 minutes, indicating that most people actually do live relatively close to their work.90 Comprehensive studies done in Toronto91 and Vancouver92 appear to 8 9 Pat Bell "Family income study shows that we're not any further ahead," Vancouver Sun, 23 August 1997, reports that in spite of a huge influx of two-income households, there has been a nearly complete stagnation of housing comes over the last twenty years. 9 0 Terry Moore and Paul Thorsnes The Transportation and Land Use Connection (American Planning Association, Planning Advisory Service Report Number 448/449, 1994) p. 81. 9 1 D3I Group for the Government of Ontario Greater Toronto Area Urban Concepts Study (Toronto, 1990). 49 contradict this conclusion. The average one-way commuting time in the G T A was approaching one hour in 1991 and the average work trip in the Vancouver C M A was over 15 km in 1992, equivalent to the distance from Vancouver's CBD to New Westminster's downtown. Few would call this amount of time or distance "living close to work." Even if this amount of travel was considered to be low, such travel distances are a good indication that an inordinate amount of street and highway capacity has been made available to maintain speeds and minimize congestion in the absence of appropriate land use decisions. Although not always well publicized, the cities of Toronto and Vancouver have officially embraced the important link between additional housing close to employment and reduced trip generation.93 In a seminal planning article with an unusually solid empirical foundation, Greg Stewart, supervisor of transportation policy and research in the City of Toronto planning department and David Nowlan, a professor of economics at the University of Toronto, made the discovery in 1991 that an average of 1.2 downtown workers were living in every new downtown housing unit.94 The transportation implication was that, while 85,000 new downtown jobs were created in the 1975-1989 period, only 50,000 new peak-hour trips were generated. The facts speak for themselves in Vancouver as well. Almost 60% of West End residents work in the downtown peninsula and 70% of peak-hour trips made by these workers are by foot, bike, or transit, a vastly higher percentage than for the region the region as a whole. The Moore and Thorsnes again took an economic approach, arguing that if people really wanted to live close to their work, this would be reflected through market mechanisms by bidding up the prices of housing available near employment. This is, of course, exactly what is being observed in places like Toronto and Vancouver, metropolitan areas featuring livable cores, and constitutes the main impetus for this thesis. Housing prices in New York City and San Francisco, two of the few 9 2 GVRD, Strategic Planning Department 1992 Greater Vancouver Travel Survey: Report 3, Travel and Demographic Characteristics (1994): p. 8. 9 3 City of Toronto, Planning and Development Department CityPlan '91 Report: Transportation and the Land-Use Nexus (1991): p. 201, "The Transportation Benefits of Increasing Central Area Housing." City of Vancouver, Planning Department Central Area Plan: Goals and Land Use Policy (1991) p. 18, "Housing Policy," reflects Toronto's approach to focus housing near the downtown. 9 4 David M. Nowlan and Greg Stewart, "Downtown Population Growth and Commuting Trips: Recent Experience in Toronto," APA Journal, Spring 1991: pp. 165-182. 50 remaining livable metropolitan cores in the U.S., are very high. Nevertheless, Moore and Thorsnes suggest that the mobility provided by the automobile makes most people trade-off proximity to work for proximity to shopping, recreation and friends. Indeed, many prominent authorities on the subject of land use and urban design continue to insist that neither sprawl nor automobile use are significant problems, given the right technology.95 "Oddly, it seems easier to grasp the real costs of sprawl than to understand the benefits of intensification, which is simply the reverse of sprawl." ...Nowlan and Stewart 2.2.6 Density and Land Use Mix Two recent studies of west coast metropolitan areas, with many similarities to Vancouver, have provided strong empirical evidence in support of the importance of land use density and mix on how, and how often, people travel for different trip purposes. The common theme was that the land use factors of density and mix of land uses exert a strong influence on the overall demand for transportation, modal splits and infrastructure investments. By definition, these higher densities, mixes of use, and reduced travel needs create more compact metropolitan regions with reduced automobile dependence, making the research results of particular interest to this thesis. Suburban residential subdivisions obviously display extremely low density and a lower mix of uses. An examination of the strength of these two factors is important for the understanding of the land use, transportation and housing dynamics that are required to achieve a compact metropolitan region with complete communities. 9 5 In Moshe Safdie The City After The Automobile: An Architect's Vision (Stoddart, 1997), the internationally acclaimed architect promotes a vision of perpetuating North American's low-density cities, but with a fleet of Le Corbusier's automated electric vehicles replacing private automobiles. In Amory B. Lovins and L. Hunter Lovins, "Reinventing the Wheels," The Atlantic Monthly, January 1995, the popular alternative-energy expert embarked upon a campaign to promote "hypercars," low-weight personal vehicles fueled by small gas-electric hybrid engines, as the solution to pollution and urban congestion. 51 Kenworthy and Newman, in their classic 1989 work Cities and Automobile Dependence: An International Sourcebook, showed a clear exponentially decreasing relationship between per capita gasoline consumption and higher urban densities in 32 metropolitan areas around the world.96 The data is powerful enough to demonstrate convincingly that, regardless where and urban centre is located and in spite of Le Corbusier's dreams of stacked urban freeways, traffic congestion and livability factors at higher densities eventually lead to the need for alternatives to the automobile. What is less clear is how often the transition is instigated by the increasing economic viability of alternative modes and how often congestion and pollution problems force metropolitan decision makers to provide alternatives. BC Transit has also noted strong relationships between increased density and increased transit use, both for increased employment density and population density.97 Data plots show Vancouver in a class of its own at the top end of the scale for density and transit usage with South Fraser subarea municipalities at the bottom end. Density was also identified by almost every respondent to the land use, transportation and housing survey conducted for this thesis as the most important factor in promoting alternative transportation and affordable housing. The two areas investigated in the studies, the Bay Area of California and the Puget Sound area of Washington, share many similarities with the GVRD. All three are large, relatively modern metropolitan areas that have a dense core with a high proportion of jobs and housing surrounded by a low density suburban fringe. The Bay Area has around 5 million people, while Puget Sound and the GVRD have around 2 million people each. Geographic constraints in all three places, such as water bodies and steep elevations, exert a strong influence over growth patterns. As a result, the provision of transportation infrastructure can be expensive and housing prices are high in the metropolitan cores, particularly for ground-oriented family housing in established neighbourhoods. Peter Newman and Jeffrey Kenworthy Cities and Automobile Dependence: An International Sourcebook (Aldershot, England: Gower Publishing Company, 1989): p. 49. 9 7 BC Transit Transit and Land Use Planning (1994): p. 6, as derived from the GVRD 1992 Travel Survey. 52 A significant difference between the three study areas is the relatively low cost of operating automobiles in the American centres in addition to the existence of an extensive, federally-funded, freeway system. The freeway system may also be responsible for a higher level of suburban employment in the U.S. Transit ridership is much higher in the GVRD, and mortgage payments are not deductible from taxable income in Canada as they are in the U.S. In a 1994 University of Washington doctoral dissertation by Lawrence Frank,98 extensive analyses were carried out to correlate the effects of density on travel distance and modal choice in the Puget Sound area. It was shown that higher density in both primarily residential areas and primarily employment areas is conducive to shorter trip distances and increases in the modal shares of walking, cycling, and public transit. Perhaps the most important conclusion of the analysis is that a critical density appeared to be necessary before a noticeable transition started from automobile predominance to the increased use of alternative modes. The critical employment density for a shift from the SOV mode for work trips appeared at 20 to 30 jobs/acre and the critical residential population density for decreased SOV travel appears at about 10 residents/acre, although this point is less sharply defined. Higher density at both ends of a trip was understandably identified as a major contributor to lower automobile ownership and use. A mix of uses within the higher density locations was also seen to contribute to trip reduction. For housing, a mix of land uses such as grocery stores, professional services and restaurants reduced travel outside of the immediate neighbourhood. For workplaces, a mix of business and employee services such as dry cleaners and print shops reduced the need for regional travel. Frank concludes that a balance of jobs and housing does in fact reduce trip distance and travel times considerably. Of some interest is that, while the results were "controlled" for the effects of non-land use variables such as age, income and household type, housing factors such as price were not considered to have an impact on results. This would appear to be a dramatic oversight, Lawrence D. Frank An Analysis of Relationships Between Urban Form (Density, Mix, and Jobs:Housing Balance) and Travel Behaviour (Mode Choice, Trip Generation, and Travel Time) (Washington State DOT & USDOT, July 1994). 53 given that basic urban land economics makes clear the importance of these two factors in locational decisions, i.e., the economic incentives to live in one place and work or shop in another. In a 1996 University of California at Berkeley masters thesis by Kara Kockelman, several dimensions of "accessibility" are added to Franks' basic density and mix indicators." Access to "Opportunity Sites" (destinations) was seen to be one of the most important "Amenities" or "Disamenities" that influence where people locate in the Bay Area. An interesting result of the study was that access to wide range of opportunity sites, including those that would be rarely visited, had a much higher importance for potential home owners than might be expected. This observation lends support to the hypothesis that people need to feel that they are able to conveniently access a high proportion of possible trip destinations by alternatives to the automobile if they are to accept less automobile-dependent residential locations and lifestyles. The distribution of housing prices was again not considered to be an important factor. Other research papers by Kockelman have examined the influences of housing prices on travel behaviour in the Bay Area. 1 0 0 Housing prices are described as a proxy for all of the usual amenities associated with a home, such as floor space and neighbourhood, with accessibility to opportunity sites seen as just another amenity. One interesting result concerned the added utility of a home being located where only one car was considered necessary instead of two. The difference in monthly mortgage payments for a home in an area where only one car was considered necessary was found to be considerably less than the monthly cost of owning an extra car. This indicates that the full monetary value of reducing car ownership may not be understood by many households. While these studies confirmed house price as a primary indicator of the desirability of a location in terms of accessibility in the Bay Area, many of the correlations are considered to be weak. Income is considered in the transportation mode decisions of households, but not in their location decisions, and then only as an average for the aggregated population. As 9 9 Kara Kockelman Travel Behaviour as a Function of Accessibility, Land Use Mixing, and Land Use Balance: Evidence from the San Francisco Bay Area (University of California, Department of City and Regional Planning Masters Thesis, 1996) 1 0 0 Kara Kockelman, "The Value of Travel Savings As Reflected in Housing Prices" and "Housing Price as a Function of Accessibility," University of California, Department of City and Regional Planning Term Papers, 1995. 54 such, the studies did not look at the subtle differences that may exist between the behaviour of different types of household. Kockelman makes the intuitively obvious, but nevertheless important, observation that the value of land is highly correlated to accessibility considerations. Land outside of metropolitan commutersheds, which has no recreational or agricultural value, generally sells for a small fraction of what urban land sells for. Accessibility considerations contributing to price fluctuations within urban areas are complex and discussed further in the section on land economics. Density and Land Use Mix - Theory and Numbers While many readers will be familiar with the importance of density and land use mix in community design and growth management, it is not always easy to see exactly how different housing forms consume land, or how all of the land use "mix factors" can be assembled to identify where a balance or imbalance is present The following example seeks to provide clarification Consider a conventional subdivision built with 30' by 90' lots with 30' ROW and no lane versus a 50' by 150' with 45' ROW and 20' lane The former housing unit has a physical footprint of 3150 sq ft and the latter 9000 sq ft , a ratio of almost I 3 As these subdivisions extend out from activity centres and the source of services such as sewers, water, hydro and gas, then the best-case cost of providing infrastructure for these subdivisions would also vary by a factor of 1 3. depending on the mix of lot sizes and street widths A comparison between entry level homes targeted for middle classes in Mexican and Canada or the U S demonstrates the vast differences in density mindsets of North Americans 1 0 1 Typical new Mexican subdivisions, dominated by two story townhouses, have an average FSR of I 0 and Measurements and layouts compiled during field work by the author at several locations in Mexico City, in several cities around Guanajuato State, Mexico and in Surrey, B.C. during January 1997. 55 a useable floorspace of approximately 1200 sq ft without basements The smallest two story house in the award winning Clover Valley Station subdivision near Vancouver, which features small lots and neo-traditional design, has 2400 sq ft of usable floor space and an FSR of 0 5 with a full basement By simple calculation, the Mexican subdivision is four times as dense, without considering street widths While Mexican residential streets were 20' to 24' wide with no back lanes, streets in Clover Valley Station range from 33' to 40' wide with an additional 22' wide rear lane Another simple calculation shows that while street space in the Mexican subdivision accounts foi 1/6 of the land available, at least 1/3 of the Canadian subdivision is dedicated to asphalt Considering these extreme differences in density, perhaps it's not really surprising that Mexico City's 20 million residents arc able to live in an area approximately the same area as Greater Vancouver's two million residents, with only a marginally higher total number of cars Typical New Subdivision m Mexico FY TH CY TH FY Legend S • Sidewalk, CY - Courtyard, 1H = Townhouse 56 I'ypicalNew "Small Lot" Subdivision in (he Iraser Valley Lane Street Figure 5 - A Comparison of Subdivision Densities Some of the differences in densities and design can be attributed to cultural factors, but it is important to note that Mexicans of all income classes live quite close to one another, and this "culture" of compactness is ultimately what North Americans living north of the Rio Grande wil have to re-lcarn if growth management is to have a reasonable chance of success Unfortunately, services provided within automobile-oriented subdivisions are just the tip of the iceberg The narrower street that was appropriate for the subdivision must expand to a four lane arterial to collect all the cars from the subdivisions and then six lane arterials or freeways as major activity centres are approached At most times of the day the streets of the subdivisions remain quiet, although damage from the elements, such as frost action, continues through the years Similaily. services such as water and electricity need additional facilities such as repeaters and pumps to get their products to users as the distance increases BC Hydro notes that the denser development called for in the GVRD Livable Region Strategy will delay the need for a number of substations in the Fraser Valley that would be needed otherwise if current development trends 57 weie to continue * The Hydro report also notes that a duplex requires one-quarter of the "linear! infrastructure" (e g , power lines, sewer lines, etc ) of a detached house and that a rowhouse uses only one-third of the heating energy of a detached house m BC Transit has concluded that it i cannot economically support feeder routes in the low density suburbs due to the large distances involved 1 0 4 Nevertheless, political decisions in the 1980s to provide minimal transit service levels in most GVRD suburban areas has drained capacity on main routes in urban areas, leading to severe overcrowding and a loss of potential ridership on the urban routes The wide distribution of activity centres in suburbs requires more capacity for access roads and large expanses of parking at the centres which lie empty much of the day This further expands the radius of the metropolis The distance that heavy delivery trucks must travel to distribute goods also increased As the damage done to pavement surfaces by trucks is orders of magnitude \ higher that of a car, arterial street maintenance costs escalate quickly On the macro level, if development is allowed to proceed at low density, infrastructure costs across the region will climb rapidly, and individual taxpayers will face higher costs in property taxes, transportation costs, and travel time Land Use Mix Land use "mix" is a measure of the actual number and variety of land uses in a subarea, which can1 be compared to the number and variety needed for a complete community Several mathematical i definitions of "mix" were found in the liteiature, none of which claimed to be true repiesentations of the real world, and for which only limited empirical confirmation was offered The concept is nevertheless well understood by organizations with specific mandates Most cities have a recipe book which specify the amount of public amenity ingredients that it feels is necessary for complete neighbourhoods Vancouver, for example, has set 2 75 acres of park per 1000 people, 2 29 sq ft 1 0 2 BC Hydro Bringing Electricity to the Livable Region (1994): pp. 52-53. 1 0 3 Ibid, p. 47. KM B C T r a n s j t Transit and Land Use Planning (1994): p. 10, shows that the average suburban residential density of 4-6 units/acre supports only hourly service during the daytime and that this service will be much more heavily subsidized than urban service. 58 of community centre per person and one elementary school per 200 children as minimum city-wide standards Developers and retail chains have well-established rules of thumb as to how many people are required in a neighbourhood, community or regional market area to support different types of sales or service outlets A consolidation of the principle elements that would be involved in creating an analytical numerical model of land use mixes is presented here There are a number of ratios corresponding to activity centres that people need to access which need to be "balanced" locally to reduce the need to travel The term "local." as defined above, is important In a metropolitan region, there is generally a built-in "balance" of uses since most common services, amenities, etc needed within the region would be assumed to exist within the region Examples of how activity centre balance ratios could be defined which measure the local balance, or mix, of land uses include E, = Employment Ratio, (Employment available in income class il^hnr^. (Employed residents in income class iKi*..*.. . ...there should he a balance of jobs and housing, as represented by the number of employed residents, in each income class within a community. For the reasons indicated earlier, (his can never be achieved perfectly, but shouldfall within a range, 0.S to 1.2 being reasonable. H, =• Housing Ratio, _ (Number of suitable housing units affordable to income class Osub.iggi (Employment available in income class iXubnrea ...there should be a stock of suitable housing available locally which matches the income profile of those working within a community. 59 P - Personal Services and Recreation Ratio = tNurrfepf^sonaJ.Sery^^ (Number of Personal Services and Recreation/ Population^ B - Business Services Ratio -(Number of Business Services / Ej^lojiment^^, (Number of Business Services / Employment^ ... (here should he sufficient series (o supply most business needs of local employers within a community. i Some menaces are dearly m „re -local" than others A Vancouver apartment dweller should no. • have ,o travel to a big-box retailer in Richmond for a jug of milk and a majo, financial firm in Vancouver's CBD should no, have to travel to a neighbourhood drug store to buy stationary in small quannties. Rvery neighbourhood may no. need an appliance store, bu, there should be a « hardware store nearby where a washer can be bough, ,o fix a leaky tap These "micro" and § "macro" relationships would need to be further refined in a complete model The important concept is that most commonly needed services, amenities, c.c. should be available locally at an I appropriate scale To determine whether or no. a subarea has the healthy mix of uses needed for a eomplele community, the ratio for each factors cou,d be wcghted according to its demonstrated tmportanc* to growth management and averaged in an equation of the form 60 IF •= Imbalance Factor = n N n N i(WE * (S (|1- Et I) * ^ ) ) + (WH*(Z (|1- Ht I) * ^ ) ) + (WP Ii - ^ 1) + (W„* I1 - B \ ) 1=1 i V ei/ 1=1 iV elf W j , Wu, W|>, W» are weighting factors related the importance of the ratio to growth management objectives, 11 - Rj is the absolute, or positive, value of one minus ratio R, N , eii is the number in the region's employed labour force in income class i, NJI is the size of region's employed labour force, n is the number of defined income classes in the subarea Taking the absolute value ensures that all differences from one are counted as contributions to the IF and that positive and negative values don't cancel each other In a perfect world, all ratios should ideally be closely balanced and have a value near one, resulting in an IF closer to zero A 's subarea would then have a balance of all the land uses identified as needed in a complete community Subareas with II- significantly greater than zero would have a serious imbalance in one or more of the defined land use parameters, and policies could be applied to address these On the macro level, if the IF indicates that significant imbalances in land uses in a subarea exist, people will need to travel to other subareas to access the employment, housing, personal services and recreation that they need The results section will indirectly examine two of these ratios, the i (im)balance of the numbers of workeis and residents in various income classes in selected parts of 2.2.7 Summary The preceding sections have presented an overview of the current thinking in the area of land use and transportation interaction. Numerous interrelated factors were presented to explain what 61 planning and engineering professionals consider to be the underlying causes of urban sprawl and automobile dependence. A survey of existing policy instruments suggested which responses have gained acceptance in different jurisdictions. What became clear in the preparation of these land use and transportation sections was that the nature of urban sprawl and automobile dependence problems, their costs to individuals and society, and workable solutions are well understood. Stated succinctly, the most realistic and proven solutions to these problems are higher densities with mixed land uses and a substantial reallocations of funding from automobile infrastructure to alternative transportation modes. The number of individuals and organizations who have both an understanding of the overall situation and the ability to push an alternative agenda in the face of strong professional and public resistance appears to be quite limited. This is particularly true given the myriad of competing entities in metropolitan areas, with various geographic and jurisdictional mandates, that make effective land use and transportation coordination a daunting task. The only approach in North America which seems to have effectively promoted such coordination has been the comprehensive "top-down" approach to planning at the state or provincial level, such as the Oregon model described earlier in the section on growth management. The success of the top-down approach in Oregon seems to result from a commitment by most stakeholders to ensure that land use, transportation and housing should all act as means to the end of more livable communities and long-term sustainability. Many of the other state and provincial approaches appear hesitant and reactionary in comparison, i.e., how to deal with the short-term growth problems as conveniently as possible. A number of respondents to the key informant survey were adamantly opposed to any kind of intervention by higher levels of government, perhaps reflecting a current suspicion of big government throughout North America. Many presuppose that, while little practical action has actually taken place, local solutions to regional problems are preferable to provincially imposed solutions. The fact remains that neighbourhoods and municipalities stubbornly resist land use and transportation changes, such as higher density in established neighbourhoods or lower parking 62 standards, and this ultimately leads to a lower quality of life for all residents in a region. Pro-development councils in the GVRD are now challenging even the generalized land use and transportation goals of the LRS, perhaps out of fear that they may someday be enforced. The Oregon model uses two deceptively simple mechanisms to overcome these problems, problems driven by land economics and the protection of perceived self-interests, while maintaining a high level of community support. Firstly, the statewide goals were developed in direct consultation with the state's citizens, establishing a strong, overarching mandate and bypassing the rhetoric of potentially parochial individual, neighbourhood, and municipal interests. Secondly, the interpretation of the guidelines developed to achieve the goals is overseen by an •independent review board, again ensuring that common goals are not misinterpreted by local officials.105 Not only does the state require effective citizen involvement in all planning processes, it provided startup grants at the time the original legislation was passed to create a strong citizen watchdog group, the 1000 Friends of Oregon, as the keeper of the vision.106 The 1000 Friends of Oregon have recently championed the "Land Use, Transportation, Air Quality (LUTRAQ) Connection" reports, which formed the basis for canceling new freeway development in the Portland area and promoting new Transit Oriented Development.107 State of Oregon, Department of Land Conservation and Development Oregon's Statewide Planning Goals and Guidelines (1995). 1 0 6 State of Oregon, Citizen Involvement Advisory Committee How To Put The People In Planning (July 1992). 1 0 7 1000 Friends of Oregon The Land Use, Transportation, Air Quality (LUTRAQ) Connection (Volumes I-IV, 1991-1992). 63 2.3 Land Economics, Housing and Location Choice 2.3.1 History and Overview A historical perspective is a good way to start an overview of what influences housing costs and where people live in relation to their jobs and other needs. The field of land economics, originally applied to agricultural land, is well developed. Early writings assumed that only two factors affected the price of land: fertility and distance to markets.108 The price or rent paid for farm land was related simply to the possible yield and market value of the crops grown, less the cost of getting the harvest to market. In agrarian societies, the value of urban land was less important, as it accounted for only a small part of the total land mass available and far fewer people lived in larger towns or cities. Early in this century, with the exodus from rural to urban areas beginning to grow, theories of urban land economics began to be formulated, with the value of commercial property being the primary interest.109 The location of commercial land in terms of proximity to supplies, amenities and markets was still a main focus of the theory. Links to transportation were seen to be the overriding factor in location decisions. The concept of "friction," or the difficulty in getting from one place to another, became a cornerstone of all theories and is still the term used to describe travel time in transportation models such as the GVRD's EMME/2 model. When the economics of residential land emerged in the 1920s, the emphasis on transportation continued,110 but had broadened to include access to activity centres other than work, such as shopping and recreation. By the early 1960s, many of today's theories on residential land values and location preferences had been established, based on the total utility of a range of property characteristics to an potential buyer or renter.111 1 0 8 David Ricardo On the Principles of Political Economy and Taxation (1821) cited in William Alonso The Economics of Urban Land: Toward a General Theory of Land Rent (Cambridge, Massachusetts: Harvard University Press, 1964). 1 0 9 Richard M. Hurd Principles of City Land Values (New York: The Record and Guide, 1903). 1 1 0 Writers from Robert M. Haig, "Towards an Understanding of the Metropolis," Quarterly Journal of Economics, Vol. 40, May 1926, and Richard U. Radcliffe Unban Land Economics (New York: McGraw-Hill, 1949) indicated that convenience in terms of accessibility to amenities is the primary component of residential rents. It is interesting to note that the terms "accessibilty" and "activity centres," have enjoyed a recent revival in the literature as "new" planning paradigms, paralleling "neo-traditional" design. 1 1 1 William Alonso The Economics of Urban Land: Toward a General Theory of Land Rent (Cambridge, Massachusetts: Harvard University Press, 1964), based on his doctoral thesis A Model of Urban Land Market: 64 A legacy of this later work was the incorportation of neighbourhood factors, such as crime, prestige and appearance, and the value of commuting time to individuals, into the models. A GVRD report on housing influences from the 1970s contains many statements that could easily be mistaken for statements from a 1990s report.112 Large increases in housing starts were then being seen the Northeast Sector and South Fraser subareas, due to the large amount of relatively cheap land available in these locations. Concerns were expressed over the displacement of low income housing due to redevelopment, the fact that housing prices were growing faster than incomes, particularly for younger families, and the tendency for new housing to be located in places without public transit. The last concern merits clarification. "Without public transit" in the 1973 context meant no transit of any kind was available in most suburban GVRD municipalities. Growing up in a automobile-oriented culture without transit may still be shaping the opinions of suburban decision makers today. Although the cost of providing development services to new subdivisions was raised as a concern, it is significant to note that the impact of housing locations on transportation needs, environmental quality or the then newly established A L R land was not mentioned. Two quotes from the report, although they may appear to be statements of the obvious, indicate the timelessness of land economics issues as applied to housing: 1. "For those families entering the single family dwelling market for the first time, rising house prices are a considerable encumbrance. Those families are forced to pay today's prices for single family accommodation and may pay two or three times as much as their neighbour. The effect of rising house prices is distributed inequitably throughout the population; those who own a house benefit from the increase in house prices, and those who wish to purchase their first house are penalized. " 1 1 3 Location and Densities of Dwellings and Businesses (University of Pennsylvania, 1960). Alonso's work is often cited in the literature, and his "unified land theories," analogous to Newton's unified field theories, seem to have stood the test of time remarkably well. Alonso admits that no economic model can be more than a gross simplification of reality and can only apply to the aggregate population as opposed to individuals. Similar results were reported at almost the same time in Lowden Wingo Jr. Transportation and Urban Land (Washington, D .C: Resources for the Future, 1961). 1 1 2 GVRD The Housing Issue (May 1973). 1 1 3 Ibid. p. 17. 65 2. "We like to think that we choose the environment in which we locate our homes andfamilies. In respect to the tightness of the market and the limits to housing unit choice, how much real choice do we really have in this regard? "II4 Land, the most important single component of housing prices in the GVRD, is not treated as other commodities by government regulations. The old adage "they ain't making any more land," with ecologically important exceptions such as marsh and tidal basin reclamation, summarizes the argument for special treatment in the case of land. If a pure supply and demand system applied to land capital, it has long been argued that land rents would eventually consume the value of all other goods and services produced by the economy for the sole benefit of a few land owners.115 More recently, it has been observed that land price increases have effectively offset some of the wealth and leisure time gains anticipated earlier this century as a result of increasing productivity, as more work time is required to pay for housing and more commute time is required to access work. 1 1 6 Governments in peacetime generally do not control what a company can produce with its capital resources, provided that the product and the production method is legal. If there is a market for a commodity, chances are good that someone will exploit the opportunity. Land use zoning has the effect of limiting what the market would normally do with land capital to maximize profits. Needless to say, some land uses, such as apartment buildings that block sunlight or factories that create noxious odours, may constitute a nuisance situation, but these torts could theoretically be remedied in the courts. Many zoning restrictions currently being applied reflect personal interests and community attitudes more than market economics. The consequences of restrictive zoning practices are clear. A 1990 report prepared for the GVRD warned:117 "The declining supply (of zoned and serviced land) will likely result in: 1 1 4 Ibid. p. 28. 1 1 5 David Ricardo Principles of Political Economy and Taxation (1821) and George Henry Progress and Poverty (1879) quoted in Edwin S. Mills and Bruce W. Hamilton Urban Economics (Harper Collins Publishers, 1989). 1 1 6 Mark Roseland "Linking Affordable Housing and Environmental Protection: The Community Land Trust as a Sustainable Urban Development Institution," Canadian Journal of Urban Research, Vol. 1, No. 2, December 1992: pp. 162-180. 1 1 7 Coriolis Consulting Corporation for the GVRD Evaluation of Greater Vancouver's Land Supply (May 1990). 66 • Higher prices for existing and new single family homes; • More rapid single family residential development in municipalities such as Langley and Maple Ridge and the Fraser Valley; • Increased pressure to develop ALR lands, particularly in municipalities such as Delta and Richmond." Arguments surrounding the equity of various land distribution mechanisms will be left to later discussion and the more complex mathematics that have been developed to support the theories will be reduced to simpler forms. It should be noted that the theories are rarely accompanied by supportive empirical data. The following sections will focus on the more tangible determinants of land values, primarily residential land in metropolitan regions, and mathematical models which represent tradeoffs between these factors. "Lifestyle choices are the most critical factor (in housing). Economics appears to drive undesirable consumptive practices." ...Real Estate Representative 2.3.2 Bid-Rent Theory Much of what has come to be known as urban land economics could be described as quantified common sense. All businesses want to find a location that will make them the most money, all households want to live in places where they can enjoy the highest possible quality of life within the constraints of their income and other competing uses for this income. The quantification of the common sense variables that go into decisions on which price to charge or pay for a property falls into the field known as bid-rent theory. The term "rent" is used as a basis for calculations, as owning or renting a property equate to the monthly cost of amortizing a total cost over a relatively long time period. 67 Decisions on where to locate almost always involve tradeoffs between competing factors. Price, for example, is rarely considered as the only factor, or even the major factor. The cheapest commercial lease rates might reflect a location with low sales volumes while the cheapest residential land price may end up being swamp land in Florida. The relative value of these factors to a business or household is called the "utility" of the factor. The procedure of assigning weights to these factors to determine the optimal combination of tradeoffs is called "maximizing utility." The utility based model predicts that, just as transportation mode choice is based on factors such as convenience, time and money, locational choices of businesses and households are influenced by factors such as accessibility ("location, location, location"), amenities, price and operating costs. The model can be applied on an aggregate basis to determine the probable prices of properties in a location or on a disaggregate basis to determine the probability than an individual household with given characteristics will locate in a given area. This thesis is primarily concerned with the disaggregated, or "micro," choices that individual households make in response to options in the housing market, which combine to create the aggregate, or "macro," situation. Once the utility of various locations have been estimated, the expected value of a property can be determined. Conversely, the likelihood that a household or business with certain features will locate in an area can be calculated using a probability function. It is here that the "push-pull" factors that are missing in most transportation models can be found. Those factors that increase the utility of a site are positive while those that decrease it are negative. Price is the most obvious component, with high prices driving households from locations near workplaces and low prices drawing them to others. Transportation models generally only consider travel time and costs.118 The Mathematics of Housing Utility and Locational Probability The following two steps can be applied to mathematically represent a household's location decision in terms of measurable parameters Appendix A contains an overview of transportation planning methods. 68 1 Determine the utility of all significant factors at a given housing location A basic example of the utility of housing, (/h, for a given household with income / and number of members m can be represented crudely as If no" • a,h*(lj ' tj) a:hX" • a?*(sh m) a4h*(ch i) etc. /„/' is the average travel time to access employment by the / workers living in the household, tj is the average travel time to access amenities such as shopping and recreation, sh is the size of a house, which could easily be expanded to include variables such as number of bedrooms against the number of children, m is the number of household members, c' is the monthly cost of housing being considered, which should include all relevant costs, such as the mortgage payment, taxes and maintenance, / is the average income of the household Heie, as income increases, the importance of the cost of the housing to the household decreases, etc. are all of the other factors left out of this simple model a*, a/\ a A etc, are proportionality constants, which can be thought of as weighting factors reflecting the relative importance of a factor to a household. a0h represents all of the hard to measure intangibles involved in the decision, i e the non-monetary and non-temporal utility of the real and perceived benefits of a potential choice of housing Examples of these intangibles include safety, neighbourhood "feel,"' density, prestige and community amenities such as parks, schools and community centres ln practice, this constant could be difficult to determine empirically 2 Determine the probability that a household with specific characteristics and a primary Place of Work (POW) will choose one type of housing and Place of Residence (POR) over all of the 69 available possibilities, given the relative utility of each choice, based on a "Multinomial Logit Model"119 This probability can be expressed as Where type household jtype,household POR,POW ~ ~U !•(>'< rou I hiiutthoiJ i l'(Hi is the Utility of housing type and Place of Residence (POW) to a specific household with a given Place of Work (POW), represents a housing type with a defined set of characteristics, i e . apartment, townhouse or detached house, new or old, etc , represents a household with a defined set of characteristics, such as income, number of workers, number of children, etc , repiesents the different housing type choices, represents the different Place of Residence choices A probabilistic model reflects the fact that not all households, even those with very similar characteristics, will necessarily make the same housing choices As a result, there would be a statistical distribution around the most probable choice There is an analogy to the field of traffic engineering Route determination for drivers would appear on the surface to be a simple matter of finding the quickest route from A to B Empirical evidence, expressed in Dial's Algorithm, shows that there is a significant variation of the actual routes that people take between two points when there are options available 1 2 0 While the majority of drivers take the fastest route, substantial numbers of drivers often prefer a variety of secondary routes 1 1 9 See, for example, Henry O. Pollakowski Urban Housing Markets and Residential Location (Lexington, Massachusetts: Lexington Books, 1982): Chapter 5, "The Multinomial Logit Model of Residential Choice." 1 2 0 Robert B. Dial "A Probabilistic Multipath Traffic Assignment Model Which Obviates Path Enumeration,' Transportation Results, Vol. 5, 1971, pp. 83-111. 70 As complex as this model may appear, it is still clearly a simplification of reality For example, the model assumes all relationships to be linear, which is usually not the case, particularly over a wide range of household types and locational factors Some factois which might appear to be obvious, such as travel time to work, seem to have little statistical significance for some income groups Others, such as travel costs, appear to be highly inelastic for higher income gioups The value of a0" is a proxy for all the variables that are either unknown or cannot be measured objectively, making it subject to suspicion 1 low does one measure a bias towards living in East Vancouver versus West Vancouver7 Another difficulty with the model is the collection of iclevant data to identify statistically significant relationships As mentioned earlier in the methodology section, many data sources come from different agencies with different mandates, study areas, definitions and collection times, making correlation difficult Nevertheless, the model serves to indicate how various factors can influence the decisions of consumers and, if properly calibrated, could be used to indicate the potential impacts of policy decisions on the locational decisions of households 2.3.3 Decision Tree Model A decision tree, commonly used in applications such as business and engineering, seeks to mimic the flow of logic that many decision makers follow without going into the mathematical detail that a deterministic probability-based model attempts. It assumes that decisions are made in a linear fashion using "bundles" of parameters.121 Parameters of greatest importance to the decision maker are dealt with first, and less important factors later. 1 2 1 Henry O. Pollakowski Urban Housing Markets and Residential Location (Lexington, Massachusetts: Lexington Books, 1982): p. 70. 71 Table 1 - Decision Tree Using a Housing "Bundle" a) Tenure b) Location c) Type d) Cost e) Size Owned Vancouver Apartment <$ 150,000 800 sq.ft. or OA- or or or Rented Inner Suburbs Townhouse <300,000 800-1200 sq.ft. or or or or Outer Suburbs House <$450,000 >1200 sq.ft. or Exurbs In this example, the household would first decide whether or not it wanted to own. Those wishing to buy would identify preferred locations within the region and then decide on a type of housing. The decision maker would then make trade-offs between the cost and size of the unit. One hypothesis of this thesis is that most young family households actually decide on Tenure, Type (Ground-Oriented versus non-Ground Oriented), Cost and Size together, then make tradeoffs between Type (Detached House versus Townhouse) and location. 2.3.4 PITTS In discussions with Dr. Ann McAfee, Director of City Plans for the City of Vancouver, suggests that a simple, five-step hybrid of decision trees and bid-rent theory replicates how locational decision are made by most GVRD households, the '"PlTTS" model:122 • Principle - The asking price of a house; • Interest - Market rates and trend versus income and employment outlook; • Taxes - DCCs, Sales taxes, land transfer taxes, and property taxes; ! Personal communication (January 1996). 72 • Transportation - Ease of access to activity centres; • Services - Neighbourhood amenities and personal services. Crime and intangibles such as prestige do not appear in this classification explicitly, but could be considered as "disamenities" under "neighbourhood amenities." "The media tends to report negative stories which contribute to a public perception that has painted the picture of transit and cities as crime-filled. This is a powerful motivator for persons to leave for the suburbs." .. .Director of Planning 2.3.5 Factors Affecting the Cost of Housing First and foremost, as discussed earlier, the availability of zoned and serviced land is the most variable factor in the land component of housing costs, and often the most important in urban areas. A 1990 study, based on the OCPs that existed at that time and an analysis of the potential for servicing, indicates that the limited supply of serviced, zoned land for single family lots, combined with an abundance of land available for multi-family units, may have unduly influenced housing prices and led to the dominance of apartment units observed in the early 1990s.123 Table 2 - Supply of Land and Absorption Rates (1990) Housing Type Short-term Capacity Long-term Capacity Annual Absorption Rate Supply (Years) Single Family Lot 45,100 59,530 7,250 14 Multi-family Units 181,760 25,600 8,950 23 Source: Coriolis Consulting Corporation for the GVRD Coriolis Consulting Corporation for the GVRD Evaluation of Greater Vancouver's Land Supply (May 1990). 73 It should be noted that about half of the short-term single family lot capacity was in Surrey, Langley District and Maple Ridge, while an alarming two-thirds of the long-term capacity was in concentrated in only two exurbs: Langley District and Maple Ridge. Vancouver, Burnaby and New Westminster had negligible remaining capacity for single family lots. Surrey and Vancouver accounted for over half of the available short-term capacity for multi-family units, which is almost seven times the available long-term capacity. A "pro-forma" or "residual" financial analysis, is commonly used in the real estate industry to determine the economic feasibility of a development project or the likely selling price for a property. Assuming a competitive housing market, the market value of housing units fall out of this analysis, along with the costs involved and probable profit margins. This section will identify the main factors which contribute to the final cost of producing a housing unit. A numerical example of a financial analysis for a typical multifamily housing project on Vancouver's West Side will then be presented which provides realistic current values for the various costs. Zoning and Density The crucial factor in the financial analysis is the Floor Space Ratio (FSR), the ratio of livable floor space in a building to the surface area of the lot, particularly in urban areas. This value is usually exclusive of parking, which is generally below grade in multifamily buildings in areas of high land value and at-grade where land prices are lower. Site layout, along with setbacks, height restrictions and other zoning requirements, may influence the number of "Square Feet Buildable" (SFB) calculation often used by developers, but FSR is usually the initial determinant of how much space in housing units is ultimately available for sale in a project. Permitted uses for a site are also an important part of the zoning component. Restrictions on uses, such as commercial activities and secondary suites, influence the value of a property. In particular, units in mixed commercial/residential buildings are often less expensive than equivalent units in purely residential buildings. 74 Land costs Land costs are based on factors such as the utility of a building site, as described above in bid-rent theory, the FSR and permitted uses. Within a given geographic area, the utility is often the same in terms of access to activity centres, neighbourhood amenities and other livability factors such as crime levels. This latter factor can be extremely important in major U.S. cities to the point where the relative value of land in suburban and urban areas can be the mirror image of similar locations in Canadian cities. Site specific influences such as high traffic levels and the presence of large parking lots, while potentially advantageous for commercial property, can act to lower residential property values substantially. Homes in the Clover Valley Station subdivision in Surrey, to be examined later, have an average cost of around $100/SFB, whereas new condominium apartments on the West Side of Vancouver are currently selling from $260/SFB to $290/SFB.1 2 4 These costs are inclusive of any exterior space, which varies from a yard in the case of Clover Valley Station to a balcony, patio or no exterior space at all in the case of condominium units. Construction costs are clearly a relatively small proportion of the cost of development in an expensive city such as Vancouver. Regional Land Price Variations It is possible to construct a "contour map" of land prices per square foot for different types of land use in the region, where lines represent constant price The "gradient" of the contour map indicates how quickly prices fall off and the "peaks" represent the most expensive, and presumably desirable, places to live The example below, for commercial property, shows that higher prices aie generally concentrated near Vancouver's CBD and drop off relatively quickly p s Based on recent market analyses by author, described later. 1 2 5 From Graeme Wynn and Timothy Oke (eds.) Vancouver and its Region (Vancouver: UBC Press, 1992): p. 225, Figure 19, GVRD: Land Cost Gradients (in thousands of dollars per acre). Price data was measured at contour intersections with a straight line along the most gradual price gradient (between Vancouver CBD and North Delta) and converted to $/sq.ft and km units. Remarkably little data of this type is readily available, although it could be "reversed engineered" from a residual analysis of residential properties in different parts of the region. 75 1991 G V R D Land Price Gradients $30.00 i $25J»-$20.00 -$i5.oo -/ $10.00-/ $5.00 -1 1 1 1 $0.00 1 1 1 1 -25 -20 -15 -10 -5 0 5 10 15 20 25 Distance from CBD (km) Figure 6 - GVRD Land Price Gradients (1991) The curve resembles a bell curve for a normal distribution, which displays the characteristics of exponentially decreasing land values, as is common in urban areas The general form of this type of distribution can be repiesented as Prices - Price,,,, x e Where Pricei„st Priceni.LN Distance Sigma is the price per square foot buildable at a given distance from a central point (taken here as the CBD), is the maximum price per square foot buildable, is the distance from the location of the maximum price, is a distance on each side of the central point from which the area under the curve is approximately 2/3 of the total area In this case, it can be thought of conceptually as a measure of the spread of land prices as distance increases If the distance is short, land prices are highly peaked and fall off rapidly further out from the CBD If the distance is far, land prices are relatively flat and developing further from the CBD is less attractive 76 Using a value of 12.5 km for Sigma, which represents the distance from Vancouver's CBD to the Fraser River, provides a reasonable first approximation of the actual price distribution in the following figure The plus and minus signs for distance scales are used here to represent two radial directions from Vancouver's CBD A value of 25 km for Sigma, which is approximately the distance from the CBD to the western border of Langley District, would represent the achievement of a more balanced distribution of land prices, particularly in regional core municipalities If a constant total value for all land in the region is assumed, the maximum price of land would decreases from approximately $100/sq ft. to near $60/sq ft This would make the price of land in the regional core substantially lower while making the price of land in the outer suburbs moderately higher 77 I Mjplc Ridpc Woslmtnsler DisInU -2S -2d 10 -5 0 5 10 Dists imr fmm ( KI) (km) IS — f — 20 25 Figure 8 - Balanced Land Price Distribution Curve It could be said that this distribution would better reflect the value to a suburb of being part of a metropolitan legion with superior infrastructure and economic opportunities Overlooking such massive benefits may have led the mayor of Surrey to claim recently that regional land use and transportation planning has stifled development in his municipality The fact that Surrey in its current form would likely not exist without the presence of Vancouver's economic engine and provincially-financed tiansportation infrastructure was clearly discounted While the model presented was simplified somewhat, it again demonstrates the power of having analytical tools to monitor the prevailing land economic "micro-climates" within the region that drive residential development and housing prices A better understanding of these trends would allow more proactive intervention in support of growth management Construction Costs and DCCs The cost of construction can vary greatly, depending on the site, the type of construction, and the quality. Sites with poor drainage, concrete structures and high-quality finishes can add anywhere from $10/SFB to $50/SFB to the cost of housing. $55/SFB to $90/SFB is the normal range, 78 exclusive of parking facilities. These costs are usually broken down into "hard" costs and "soft" costs, reflecting items such as materials and labour or items such as management and fees. Where the cost of most infrastructure for new developments was once paid for from general revenues, the current practice is to allocate these costs "up-front" to developers who then pass them on to new home owners. These are known as Development Cost Charges (DCCs) and have two components in the GVRD: the municipal DCCs which apply to amenities such as schools, libraries, and parks, and regional DCCs which cover services such as water and sewers. Total DCCs range from $10,000 to $15,000 and are generally charged by the unit or according to the unit type and size. If charged by the unit, a smaller apartment may face an additional cost of up to $20/SFB. It is interesting to note that the proposed City of Vancouver rate structure for new regional sewer DCCs charges a townhouse almost the same as a detached house and an apartment of any type or size would pay 2/3 of the DCC that a detached house would.1 2 6 Development and Building Standards Alternative Development Standards (ADS) are increasingly becoming examined as a way of bringing down the cost of housing. The most prominent standard is the amount of space required for road and lane right of ways, and consequently the amount of land that is available for housing within a property. Excessive minimum setbacks from streets will also limit the amount of land that is available for housing. Traditional Right Of Ways (ROW) can range up to 66' (20m) or more for residential streets. Ratios of 30% or more of urban land dedicated to streets and lanes are not unusual in many municipalities. A reduction to 20% to 25% could result in a 10% to 15% reduction in land costs. The role of land savings of this magnitude was confirmed in a 1992 CMHC report.127 The District of North Vancouver has proposed a new regional standard as low as 13m and the City of Portland, Oregon has created "skinny" street standards in predominantly residential areas with only one travel lane and parking on one side of the street only. 1 2 6 City of Vancouver, Standing Committee on City Services and Budgets, "Regional Development Cost Charges," Administrative Report, 12 September 1996. 1 2 7 IBI Group for CMHC, Canadian Homebuilders Association and the University of Western Ontario Achieving Infrastructure Efficiency (Ottawa: CMHC, 1992): p. 20. 79 The current practice of using separate space in utility trenches for water, sewer, electrical and gas under the street ROW requires extra land. The importance of integrating infrastructure for services could result in life cycle savings of over $10,000, according to CMHC. 1 2 8 Another set of standards applies to building practices. In the City of Vancouver, for example, the additional cost of providing compulsory minimum levels of parking and sprinkler fire extinguisher systems adds $20,000-30,000 per built unit. For smaller units, these two requirements alone could add up to $30/SFB to the total cost of a unit. A Financial Analysis for a Residential Development Project The following analysis will provide a current example of how the costs of a residential project are reflected in the final price of a new unit offered to potential buyeis An understanding of the magnitude of the cost components will provide a context for some of the policy recommendations made later The example applies to a recent mixed retail and apartment condominium project in the Kitsilano neighbourhood of Vancouver This type of development being strongly promoted by Vancouver's planning department as a favoured solution to complete communities and more affordable housing u < ) The analytical principles used would apply to any residential project, whether apartment or ground-oriented A profile of the types of buyers in this neighbourhood will be presented in the results section for comparison against the profile of buyers in new suburban subdivisions The site chosen for analysis is on West 4th Avenue in Vancouver, is zoned C2, and lies along a stretch of major arterial street which is undergoing an extensive development phase, in spite of 1 2 8 CMHC Infrastructure Costs Associated with Conventional and Alternative Development Standards (Ottawa: CMHC, 1995): p. 30. 1 2 9 Condominium market data from MacDonald Realtors, Grey stone Properties Limited, United Pacific Management Corporation, Intergulf Development Corporation QXitsilano), Bonaventure Projects, Real Estate Board of Greater Vancouver, CMHC December 1996 New Housing Report. Local land value estimates from Prospero International Realty. Zoning and building requirements from City of Vancouver, "Zoning and Development By-law, C-2 District Schedule" and "Parking By-law No. 6059." Construction estimates from Ledcor Industries, a prominent Vancouver construction firm. 80 increasingly heavy traffic volumes The total lot size is approximately 8000 sq ft with a total FSR of 3 00 Some of the more important C2 zoning conditions to consider include • A maximum FSR of 2 5 can be applied for residential purposes. • A height restriction of 43 feet exists, which could be relaxed due to the grade of the land. • The first storey of the building to a depth of 10 7 metres, exclusive of entrances, must be used for non-residential purposes, • A minimum requirement for one off-street parking stall is required for every 753 sq ft of gross floor area and a maximum of 2 2 stalls per unit The FSR is exclusive of • Parking stalls built at or below grade, • Open balconies and decks to a maximum of 8% of residential floor area, • Patios and roof gardens, • Residential storage space above grade up to 40 sq ft per unit, • Amenity areas such as day care, recreation facilities, and meeting rooms to a maximum of 10% of the total building floor area The bottom line, or the price of land per square foot marketable, would be much lower in a suburban area The price of land per square foot marketable for condominiums in Kitsilano ($90-100/SFB) turns out to be comparable to the total price per square foot of a detached house and lot on the outskirts of Surrey 81 Site: Area 8330 sq.ft. ResidentialCommercialCombined FSR 2.5 0.5 3 Gross floor area 20825 4165 24990 sq.ft. Marketable area ratio 85% 85% Net marketable floor area 17701 3540 21242 sq.ft. Parking stalls 36 0 36 Revenue: Price $250 $300 per sq.ft. Marketing/Commission 5.00% 5.00% Sales Marketing i Net Revenue $4,425,313 $1,062,075 $5,487,388 -$221,266 -$53,104 -$274,369 $4,204,047 $1,008,971 $5,213,018 s Costs: Annual interest rate > Term of loan 6.00% 1 Hard costs (unit) $75 65 Parking cost $35 N/A Parking space size 350 N/A Soft costs (ratio) 20% 20% Profit allowance 15% 15% Years per sq.ft. per sq.ft. sq.ft. of Hard Costs of Gross Costs Hard costs $1,327,594 $230,116 $1,557,710 Soft costs $265,519 $46,023 $311,542 Parking costs $441,000 $0 $441,000 Financing $122,047 $16,568 $138,615 Profit $630,607 $151,346 $781,953 Total costs (less land) $3,230,820 Residual available for land Price of land per square foot marketable $1,982,198 $93 Figure 9 - Financial Analysis of a Current Development Project 82 2.3.6 Location of Employment Factors affecting the location of employment, while not the primary focus of the thesis, is nevertheless closely related to the location of housing and the demand for transportation infrastructure. If employment in the region was locating in greater numbers sufficiently close to affordable housing, there would be little need for a thesis which investigates mechanisms for providing affordable housing closer to employment. Earlier models of land economics focused theory on a dense CBD at or near a transportation and goods distribution node. The CBD was surrounded by commercial and residential areas. Vancouver was certainly no exception to this historical model.130 The city nucleated around the transportation nodes of the port and the terminus of the transcontinental railway. Most people walked to work or took an early, horse-drawn version of a trolley. The horses were replaced near the turn of the century by electric streetcars, and housing patterns changed as "streetcar suburbs" along the radial arms of the streetcar network were formed. Most people still lived within walking distance of the streetcar, and their work destinations were to the CBD or the commercial districts that lined the streetcar routes. Goods were mostly moved by horse drawn wagons and congestion could be a significant problem, as many old photographs of major downtown areas throughout North America show. With the advent of motorized goods movement in North America, business patterns changed dramatically.131 No longer tied to locations close to the CBD or rail lines, many businesses chose locations that traded off land prices with accessibility for customers and workers. Until after the Second World War, worker mobility was quite limited, as private automobiles were not common. Starting in the 1950s, accessibility problems for customers and workers were effectively removed 1 3 0 Yvonne Brodda A Trip Around Vancouver: The Early Days of the Transit System background Report, Vancouver Draft Transportation Plan, 1996). 1 3 1 Edwin S. Mills and Bruce W. Hamilton Urban Economics (Harper Collins Publishers, 1989): Chapter 12, "Trends in Urban Transportation." 83 as barriers to the location of business, and the current paradigm of business location theory began. The effects have been dramatic, particularly in the U.S. where many CBDs have been dismantled and reverted to dysfunctional inner cities surrounded by suburban business parks. During the 1980s, metropolitan planners began experimenting with the concept of promoting regional town centres around a central regional core in order to promote distributed growth and provide employment closer to residential areas. Metro Toronto and the GVRD were notable supporters of the concept, which contributed greatly to the urban form currently observed in these two regions.132 Vancouver's CBD as late as 1981 has been described as "classic pole model of an agglomeration economy," as confirmed by a survey of companies that showed 71% of corporate respondents citing "Proximity to Business Contacts" as the main reason for locating in the CBD while 'Labour Force Accessibility" was important to only 33% of the respondents.133 Suburban corporate centres were seen to be more viable in the U.S. due to the existance of an extensive, federally-funded freeway system in that country. The demographics of the Vancouver workplace in the early 1980s is worth noting, with highly-paid executives and management personnel being almost exclusively men while lowly-paid support staff were predominantly women. A 1985 study of the location decisions of two CBD head offices, BC Hydro's decision to remain in the CBD and BC Tel's decision to move to Burnaby, shows that one of the greatest social benefits of BC Tel's move was to cut employee travel time by over 1/3, and by over 2/5 for the support staff, who found it difficult to afford living near the C B D . 1 3 4 Employee surveys of the two companies, performed several years after the BC Tel move, showed that BC Hydro's employee residences were distributed throughout the region while BC Tel's employees were concentrated in the area immediately surrounding the new head office. The same survey showed that employee satisfaction with the workplace was most Thomas Hutton and H. Craig Davis "The Role of Office Location in Regional Town Centre Planning and the Metropolitan Multinucleation: The Case of Vancouver," Canadian Journal of Regional Science, Vol. 8, No. 1, Spring 1985. 1 3 3 Thomas Hutton and David Ley "Location, Linkages and Labour: The Downtown Complex of Corporate Activities in a Medium Size City, Vancouver, British Columbia," Economic Geography, Vol. 63, April 1987. 1 3 4 David Ley "Downtown or the Suburbs? A Comparative Study of Two Vancouver Head Offices," The Canadian Geographer, Vol. 29, No. 1, 1985: pp. 30-43. 84 directly related to commuting time, with BC Hydro employees who lived in the West End and walked to work showing the greatest satisfaction with their workplace location. It should be noted that Burnaby's housing prices were much lower in the early 1980s and the observed results of the relocation might have been substantially different if it had been made in 1996. A 1995 study in the San Francisco Bay Area concluded that the majority of factors most influential in the locational decisions of larger employers were not dependent on market forces, but were shaped by public policy: including housing costs, land costs, travel times and operating costs.135 "Pull" factors were considered to be generally more important than "push" factors. Of 17 possible push-pull factors, the top pull factors that drew businesses to new locations were: • Cost and suitability of land and space; • Availability of professional and skilled labour; • Image/prestige of location; • Transportation/commuting time; • Local attitudes towards business. Urifortunately, transportation and commuting time were not disaggregated in the study, so it is not clear whether respondents were mostly concerned about the time that their workers needed to get to work or the business cost of moving goods.136 It could be implied from the concern over availability of labour that establishing in a location that was more accessible for available professional and skilled labour was an important factor. The most important "push" factors driving businesses from their current locations were: • Cost and suitability of land and space; • Local attitudes towards business; 1 3 5 Gruen, Gruen and Associates "Push me, pull me: corporate locational decisions," Urban Land, Vol. 44, No. 2, February 1985. 1 3 6 In. Thomas Hutton The Transformation of Canada's Pacific Metropolis: A Study of Vancouver (Montreal: Institute for Research on Public Policy, 1997): Chapter 5, "Structural Change and the Post-Industrial City-Region: Implications of the Vancouver Experience," indicates that post-industrial cities are increasingly competing with each other to attract highly skilled workers with amenities and a high quality of life. 85 • Housing costs; • Transportation/commuting time. The most important common factor is, not surprisingly, the cost and suitability of land and space, an element over which local governments have ultimate control. Transportation/commuting time and local attitudes towards business are also common factors. Of particular note is that housing costs are a significant "push" factor, which could again be linked to concerns over the availability of labour. The City of San Francisco has housing prices of the same order of magnitude as those in the City of Vancouver, with an average detached house price of $316,073 (U.S.) and a median price of $294,800 in 1990.137 One of the major concerns of U.S. businesses locating in suburban business parks is the shortage of lower-income labour due to difficulties in accessing these locations by those without cars who are forced by economic reasons to live in inner-city areas.138 Comprehensive empirical data on the distribution of various types of employment, and the income profile of employees working within municipalities, is limited. Most census data which includes personal and household information is focused on the "night time population," i.e., where people reside not where they work. There is little recent information, other than qualitative observations, of the trends in GVRD employer locations or of the factors that influence their decisions. A 1995 GVRD publication on the subject provided an estimate of employment distribution by Standard Industrial Classification (SIC) code, based on extrapolating responses received from a survey of employers in the region.139 It would be very interesting to correlate this information with various indicators to determine empirically what influenced the location of these employers. Perhaps the most simple proxy to measure the attractiveness of an area to business, the ratio of available employment to employed residents presented later in the section on growth trends, shows that the residential location of the region's labour pool is irrelevant to the majority of GVRD employers. 1 3 7 U.S. census data, 1990. 1 3 8 G. Thomas Kingsley and Margery Austin Turner Housing Markets and Residential Mobility Washington: The Urban Institute Press, 1993) contains a number of essays about this problem. "Antipoverty Strategy Where the Rubber Meets the Road: Transporting Workers to Jobs," by Mark Alan Hughes, discusses the problems which many American employers encounter in attracting lower-income workers to the suburbs. 1 3 9 GVRD, Strategic Planning Department Estimation of the 1994 Spatial Distribution of Employment in Greater Vancouver: Methodology and Results (July 1995). 86 2.3.7 Summary Households and businesses, either consciously or unconsciously, are constantly making tradeoffs in their location decisions in order to maximize the utility of a property or unit. Neighbourhood, costs, and proximity to amenities are three of the most common factors considered. These tradeoffs are reflected in the price that developers are willing to pay for land at any given location and the distance that workers are willing to travel to access employment, services and other needs. Most employers do not appear be basing their location decisions primarily on either the absolute cost of a location or the residential location of the region's workforce. The price and supply of housing units is dictated predominantly by the cost of land, which is strongly influenced by market demand and the supply of appropriately zoned and serviced land. In Vancouver, the price of land is very high near employment concentrations and drops off gradually towards residential suburbs. The land price gradient can be mapped by calculating probable land prices using a residual analysis. Development and regulatory costs, including DCCs and building standards, can contribute significantly to the final cost of new housing units. A considerable number of publications feature complex mathematical models of bid-rent theories. Nevertheless, surprisingly little empirical work appears to have been carried out to determine the many subtle interrelationships present or the strength of the various determinants involved in locational decisions. Housing prices vary highly according to local and regional conditions and directly influence household location decisions, which then influence travel habits. "If you improve a piece of highway, you almost always get a short-term reduction in congestion, but you also get a long-term increase in traffic volume and in congestion, because people take into account and reevaluate their location decisions with respect to their place of work and their place of residence.140 ...Stephen Putnam, Professor of City and Regional Planning and Director, Urban Simulation Laboratory, University of Pennsylvania Quoted in Chris Lazarus "Integrated Modelling" Earthword, Vol. 4, 1992, Transportation Planning. 87 2.4 Growth Management Policies 2.4.1 De Facto Policies Since 1973, the Agricultural Land Reserve has borne the brunt of the responsibility for growth management in the Lower Mainland. The ALR was followed in 1994 by the Forest Land Reserve (FLR), which is playing an increasingly significant role in limiting urban sprawl, particularly in the North Shore subarea and on the southeast coast of Vancouver Island. While it has been said that the A L R was intended primarily as a measure to protect farms,141 the entity that farms were being protected from was primarily urban sprawl. Hence the roles of the A L R and an Oregon-style Urban Growth Boundary (UGB) can be thought of as similar. Unfortunately, the ALR does not include the many low-density, semi-rural throughout the region in critical growth areas such as Surrey, Langley, District, Pitt Meadows and Maple Ridge. Large inventories of cheap land for residential construction are available in all of these places as a result, providing 'Tootholds" for further development in the Fraser Valley. Although the integrity of the A L R has remained essentially intact since 1973, development pressures have increasingly infringed on farming activities at the edges. In addition to attempts to use the A L R for such dubious purposes such as golf courses and greenhouses, there have been requests to withdraw land considered as "non-prime" for low-density development, which have largely been denied.142 The Promontory subdivision and proposed Ryder Lake projects in the District of Chilliwack are good examples of where the Ministry of Agriculture has recommended release of A L R land to protect prime Fraser Valley "bottom" farmland.143 Agricultural Land Commission (ALC) Vice Chair Julie Glover, speaking at the UBC School of Community and Regional Planning in March, 1997. 1 4 2 Province of British Columbia, Agricultural Land Commission ALC Annual Report: April 1, 1995-March 31 1996 (1996): p. 10, indicates that while the overall amount of land in the ALR has remained constant, 11,500 acres of prime farmland has been removed and replaced predominantly with farmland classified as second class. 1 4 3 From discussions in March, 1996, with Bill Weismueller, Chilliwack District, Ministry of Agriculture, recorded in Mark Allison, Jennifer Keesmaat, Edward Kozak "Growth Management in the District of Chilliwack," UBC School of Community and Regional Planning, Project Report, April 1996. 88 The anti-planning legacy of the Social Credit government in the 1980s is still felt, when the planning role of regional districts was eliminated and the Municipal Act was modified to replace planning guidelines with a new sections under the title of "Management of Development."145 Many municipal planning departments in B.C. continue to use the name "Development Services. As will be discussed in the sections on transportation and land use policy, the province often appears to fuel low-density growth on the fringe of the region by financing transportation infrastructure projects which support suburban lifestyles and long-distance commuting. 2.4.2 Official Policies The 1995 Growth Strategies Statutes Amendment Act (GSA) sought to establish regional growth management strategies and ensure conformity between regional and local policies to promote development that is "socially, economically and environmentally healthy and that makes efficient use of public facilities and services, land and other resources."146 The statute officially restored the strategic planning function of regional districts, which had been removed during the early 1980s by the provincial government. The GVRD, acting unofficially with the consensus of its member municipalities, initiated the "Creating Our Future" regional planning exercise in the 1980s which eventually led to the Livable Region Strategy (LRS), the province's first Regional Growth Strategy. The GSA outlines the following land use, housing and transportation concerns that should be addressed in a Regional Growth Strategy:147 • "Avoiding urban sprawl and ensuring that development takes place where adequate facilities exist or can be provided in a timely, economic and efficient manner; • Settlement patterns that minimize the use of automobiles and encourage walking, bicycling, and the efficient use of public transit; 1 4 4 Strategic Planning Department, GVRD The History, Status and Prospects of Regional Planning in Greater Vancouver (July 1994): p. 4. 1 4 5 Province of British Columbia, Municipal Act, Part 19. 1 4 6 GSA Section 942.11(1). 1 4 7 GSA Section 942.11(2). 89 • The efficient movement of goods and people while making effective use of transportation and utility corridors; • Protecting environmentally sensitive areas; • Maintaining the integrity of a secure and productive resource base, including the agricultural and forest land reserves; • Reducing and preventing air, land and water pollution; • Adequate, affordable, and appropriate housing; • Adequate inventories of suitable land and resources for future settlement; • Protecting the quality and quantity of ground water and surface water; • Settlement patterns that minimize the risks associated with natural hazards; • Preserving, creating, and linking urban and rural open space including parks and recreation areas; • Planning for energy supplies and promoting efficient use, conservation and alternative forms of energy. " The first three items presented deal directly with urban sprawl and transportation issues, while many of the remaining items deal directly or indirectly with these issues as well as "settlement patterns" and housing. The GSA mandates a broad consultative process and strongly supports a consensus based approach to conflict resolution.148 Once approved by member municipalities and adopted by the board of a regional district, the GSA requires that the Official Community Plans (OCPs) of municipalities within a regional district include a Regional Context Statement which aligns their policies with the Regional Growth Strategy. The process for adoption and implementation of a strategy is highly consensual, with complex arbitration procedures, which has been cited as a potential flaw of the legislation.149 In general, all neighbouring municipalities and the board of the 1 4 8 GSA additions to the Municipal Act Sections 942.2 through 942.24 deal with the dispute resolution mechanism. 1 4 9 Artibise, Alan (Professor, School of Community and Regional Planning, UBC) "Our new regional plan needs a plan - to police planners & politicians", Vancouver Sun, 09 November 1995 90 regional district must approve of a municipality's Regional Context Statement before it may come into effect. The GVRD board has set a deadline of February 1998 for the preparation of Regional Context Statements consistent with the LRS. The Municipal Act, through wording describing the OCP as "a general statement of the broad objectives" of a community, invites municipalities to avoid the statement of clear, ascertainable goals.150 The GSA reinforces this lack of clear direction by stating that Regional Growth Strategies "should work towards" sustainable settlement patterns and that OCPs need only "be made consistent over time" with Regional Growth Strategies. Courts in BC have taken the stand that zoning bylaws must clearly contradict an OCP before they can be quashed, which in practice means that overly general OCPs may not ensure that community and regional growth management goals will be honoured.151 Municipalities and regional districts that produce Regional Growth Strategies and Regional Context Statements that are not clear or ascertainable make it difficult for their neighbours to counter the negative effects of poor settlement patterns in a timely manner. Although the following example, a short excerpt from Oregon's Statewide Planning Goal 10 Guidelines (Housing), relates to a subject area covered in the later section on housing policies, it demonstrates the level of the specificity that is lacking in B.C.'s G S A : 1 5 2 "Provide for the housing needs of the state. Buildable lands for residential use shall be inventoried and plans shall encourage the availability of adequate numbers of housing units at price ranges and rent levels which are commensurate with the financial capabilities of Oregon households and allow for flexibility of housing location, type and density. " In contrast, B.C.'s GSA states simply that "adequate, affordable, and appropriate housing" as "a matter that should be dealt with" in Regional Growth Strategies. Oregon's L C D C provided an 1 5 0 B.C. Municipal Act, Section 945. 1 5 1 See, for example, Rogers versus Saanich, BC Municipal and Planning Law Reports, Volume 22. 1 5 2 State of Oregon, Department of Land Conservation and Development Oregon's Statewide Planning Goals and Guidelines, 1995 Edition (1995). The LCDC has the power to revise these goals and guidelines. Eight new goals have been added since Senate Bill 100 took effect in 1973. 91 expanded clarification of the statewide planning goal on housing to require that each community within a region consider the broader housing needs of the region to arrive at a fair allocation of housing types (i.e., single family versus multifamily housing). Based on this L C D C clarification, Portland's Metropolitan Housing Rule then required local plans to: • "Provide adequate land zoned for needed housing types; • Ensure that land within the Metropolitan Portland (Metro) urban growth boundary may accommodate the region's projected population growth; • Provide greater certainty to the development process; and • Reduce housing costs." The Housing Rule also requires that at least half of all new residential construction in Metro Portland be multi-family housing and specifies minimum housing densities. The L C D C ensures that the letter as well as the spirit of the law are observed by regional and local governments in their plans. The result of this more ascertainable and quantifiable approach to the goals of growth management has been dramatically increased the number of multi-family units, increased affordability and the achievement of density targets.153 As will be seen in the section on housing policies, B.C. has given municipalities a number of tools in the Municipal Act for the provision of more affordable housing, such as density bonusing and density transfer, but these powers are entirely discretionary and have not been widely applied. "There is an awareness of the need to manage "growth" but only the power to manage development is actually exercised. Growth itself is almost unmanageable." ...Urban Policy Analyst 1 5 3 1000 Friends of Oregon Managing Growth to Promote Affordable Housing: Revisiting Oregon's Goal 10, Executive Summary (September 1991): p. 5. 92 Under the GSA, the GVRD can ensure that member municipalities include a Regional Context Statement in their OCPs which conform to the LRS. The legislation is in its infancy and the mechanics of such coordination and arbitration can lead to compromises. The OCP prepared in 1996 for UBC's unincorporated University Endowment Lands (UEL) was the first test case for reviewing a Regional Context Statement's conformance to the GVRD's LRS. The OCP called for a large proportion of high-end market condominiums to be built near the City of Vancouver's western boundary with a fairly traditional transportation plan. The city, which holds a large number of seats on the regional district board, feared increased traffic levels and was strongly opposed to the O C P . 1 S 4 The GVRD board subsequently rejected the OCP, setting much higher requirements for providing affordable housing geared to people working and studying on campus (50%) and mandating a comprehensive transportation plan to reduce current traffic levels. While this show of resolve by the regional board was commendable, it is not clear that the OCPs of more powerful municipalities in the region will be treated in a similar manner. In particular, it will be of considerable interest to see if the City of Vancouver, with the nation's highest real estate costs for a major city, and a large surplus of jobs over housing, will include targets for sufficient suitable and affordable housing to accomodate a high percentage of those employed in the city in its Regional Context Statement. In the GSA, the province specifically rejected the concept of an arms-length provincial planning review body such as the Ontario Municipal Board (OMB) or Oregon's Land Conservation and Development Commission (LCDC) to ensure that a Regional Growth Strategies and Regional Context Statements follow the spirit of the G S A . 1 5 5 Dispute resolution is focussed on resolving disagreements between municipalities within a regional district, not in ensuring conformance with provincial goals. Theoretically, if the Minister of Municipal Affairs and Housing ordered the directors of a regional district experiencing runaway growth to prepare a Regional Growth Strategy, and the board approved a strategy that said "nothing is required," the minister would be 1 5 4 City of Vancouver, Planning Department, "Official Community Plan for Part of Electoral Area A UBC," Policy Report to Vancouver City Council, September 10, 1996. 1 5 5 Province of British Columbia, Ministry of Municipal Affairs Updating British Columbia's Planning System (September 1994): p. 11. 93 unable to intervene. The GSA would first need to be amended or policy guidelines regarding the strategy development process or content would need to be mandated, as the GSA currently allows. Such policy guidelines are generally not enforceable, whereas regulations would be. In any event, no policy guidelines are currently written. It is also worth noting that, although regional districts and their member municipalities must abide by the terms of a Regional Growth Strategy, the province itself is not bound to respect it. The GSA does provide for the formation of an intergovernmental advisory committee to coordinate the actions of local and provincial governments in the area of growth management, but the advisory committee was given no statutory powers.156 A relevant example of unilateral provincial action came in September 1996 when M o T H declared that the minimum occupancy of vehicles using newly opened H O V lanes on the Barnet-Hastings "People Mover" project would be set at two, contrary to the GVRD policy requiring three or more occupants.157 Automobile traffic on this corridor increased 50% within one year of operation as a result of the increased capacity.158 Similarly, the province has invested over $200 million dollars in the "West Coast Express" commuter rail project to the North Fraser Valley bedroom communities of Pitt Meadows, Maple Ridge and Mission in 1995, infrastructure which had been criticized by the G V R D . 1 5 9 GSA Sections 942.29 "Intergovernmental advisory committees" and 942.3 "Implementation agreements." GVRD/MoTH Transport 2021: Medium Range Plan for Greater Vancouver (October 1993): pp. 45-47. Glenn Bonn "Barnet Highway traffic climbs, but moves faster," Vancouver Sun, 31 March 1997. Ibid. pp. 55-57. 94 2.5 Land Use and Transportation Policies Transportation and land use policies at many levels of government impact the Lower Mainland and exert a strong influence on urban form, the locational decisions of households and the choice in transportation mode made by individuals. The level of coordination between the various levels of government, and indeed within levels of government, has historically ranged between minimal and low, although there have been recent signs of increased cooperation.160 2.5.1 De Facto Policies A review of the land use and transportation decisions that have been made in B.C. results in two clear observations: • A form of laissez-faire economics, operating within a myriad of restrictive building regulations and zoning bylaws, is the basis for municipal land use policies; • Private automobiles and commercial airlines, supported by highly subsidized infrastructure, are the basis for transportation policies. The survey of key stakeholders was unanimous in citing high land prices as a result of low density zoning as the major cause of the high housing prices leading to urban sprawl. Nevertheless, there is a near-universal resistance to increased density from members of the community, regardless of the municipality. Municipalities with a surplus of businesses and high density apartments are reluctant to alter these land use patterns, in part due to their contributions to cross-subsidizing property taxes in Single Family Detached home (SFD) areas. Vancouver is a notable example where high density multi-family development is restricted primarily to the dense downtown core, 1 6 0 The GVRD and the Province signed an agreement in May 1997 to undertake a review of transportation funding and governance. 95 released industrial lands and major arterial streets. Meanwhile, 30% of the population lives in the 70% of the city reserved for SFDs and enjoys low property taxes.161 Although almost every official policy refers vaguely to the need for a "mix" of land uses and housing types, municipal planners can be reluctant to cooperate when attempts are made to implement new uses.162 A glance at a regional zoning map shows that land uses are highly segregated between commercial and residential uses and that potential locations for mixed use development are limited.163 Affordable housing is generally equated with social housing, and there are no serious initiatives in the region aimed at providing a continuum of housing types and prices that provide reasonable housing options for people working within the community. Concerning transportation, in an estimate generally considered to be conservative, the GVRD estimates that every automobile on the road, with an average occupancy of only 1.12, was subsidized by over $2600 in 1993. If transit passengers were subsidized at this level, the government would be paying each passenger approximately $600 per year instead of requiring the equivalent of $600 per year for monthly transit passes. Provincial government policy is to keep insurance rates low, in part by having part-time "recreation" drivers cross-subsidize full-time commuting drivers.164 Most municipal zoning bylaws require that ample minimum parking levels for most foreseeable occasions is available at all destinations. Although most policies stress that walking, cycling, and transit are to be given priority over private automobiles, transportation budgets overwhelmingly support the provision of street capacity for general purpose traffic. Design standards for streets and buildings, such as street widths and parking requirements, continue to focus on the automobile. Pedestrians are forbidden 1 6 1 From 1997 municipal property tax reports, Vancouver, with an exceptionally high level of services, is in the middle range of GVRD municipalities for absolute property taxes, but has by far the lowest rate if based on assessed value. 1 6 2 Frances Bula "Developer to quit future projects in city, blaming over-regulation," Vancouver Sun, 31 July 1997. 1 6 3 GVRD Strategic Planning Department Greater Vancouver Land Use Map (1996) and Greater Vancouver Economic Development Opportunities Map (1995) provide an excellent overview of zoning and lan use patterns throughout the region. 1 6 4 Glenn Bonn "$300-million loss anticipated this year, ICBC reports," Vancouver Sun, 21 March 1997. 96 from crossing at dozens of high-traffic intersections throughout the City of Vancouver. Off-peak transit service and transit outside of major destinations is limited and only a minority of bus stops are equipped with a bench or shelter. As an automobile is required in many parts of the region for trips other than the commute to work, and the fixed costs of driving are much larger than the operating costs, few are inspired to incur the expense of both car ownership and transit fares. "Affordable housing is an extremely important issue. One of the biggest obstacles that I face in encouraging the use of more sustainable transportation modes has to do with the argument that people feel they "must" move to the distant suburbs in order to find affordable housing." . . .TDM Specialist 2.5.2 Federal Policies The U.S., through support for the interstate highway system and legislation such as the Integrated Surface Transportation Efficiency Act (ISTEA, pronounced "ice tea"), has a direct role in transportation in urban areas. The deduction of mortgage interest from income for tax purposes in the U.S. can be seen as a massive subsidy to promote suburban home ownership. In contrast, Canadian federal policies are rarely seen as having an impact on land use or transportation. Nevertheless, federal tax regulations and policies that support the air and road modes while gradually eliminating support for the rail mode have had several significant impacts on local land use and transportation. Firstly, long-distance travel has become increasingly more convenient by combining car and airplane trips than by combining public transit and bus or rail trips. While air navigation services have until quite recently been provided for nominal fees, passenger rail service and subsidies have been systematically eliminated.165 Many airports and railway stations have been relocated to places where cheaper land was available, regardless of accessibility considerations. The choices 1 6 5 Recently privatized, the air navigation company "NavCan" will receive the new, taxpayer financed Canadian Automated Air Traffic Control System (CAATS) for a small payment. 97 of Mirabel and Pickering as second airport sites for Montreal and Toronto are perhaps the most spectacular and costly examples to date. As the federal government provided no support for rapid transit to these sites, including Vancouver, many are poorly served by public transit and have become almost entirely automobile dependent.166 International bus and train passengers face considerable delays at the border due to thorough customs inspections, and the frequency of service to most secondary destinations from airports or train stations is minimal.167 Secondly, the trucking industry, through strong lobbying, has succeeded in achieving a massive expansion of the road network while only paying a fraction of the taxes required to repair the damage that trucks inflict on the roads.168 Due to the heavy subsidies that the trucking industry receives, and the federal requirement that railways pay the full cost of government provided services, transport companies have increasingly shifted their shipments from rail to road. As there are few practical ways to force trucks and cars to use separate facilities, the additional capacity needed to support efficient goods movement by truck has been made available for cars as well. The onslaught of additional trucks in urban areas has caused a dramatic decrease in the quality of life in many residential neighbourhoods. Major truck routes in Vancouver, such as the Knight Street/Clark Drive corridor and Hastings Street, have become highly undesirable places to live. Rental vacancies in the East Hastings area are four times of those in other parts of the city and rents are only two thirds of the city-wide average.169 Thirdly, federal taxation policies contribute to low-density land use and automobile dependence. Employers can deduct the cost of providing free parking to their employees and customers, while See Vancouver International Airport Authority 1995-2015 Draft Master Plan (July 1995): Chapter 9, Ground Transportation. With 4,100 parking spots currently available, Vancouver's new airport facilities will include 3600 additional parking spots and no new transit facilities. The only public transit access to the airport currently is the #100 bus which runs at 30 minute intervals in peak periods. 1 6 7 There is currently one train per day between Seattle and Vancouver. The train departs Vancouver in the early evening, reaching Seattle after 2300. The return train leaves Seattle at 0700. Travel time is approximately five hours, compared to two hours by car and 45 minutes by airplane. Passengers arriving in Vancouver can stand for up to an hour in a lineup at Canada Customs with their luggage. 1 6 8 See for example,Government of Canada Getting There: The Report of The Royal Commision on National Passenger Transportation (Ottawa: Minster of Supply and Services, 1991): p. 8,138, 210. 1 6 9 CMHC Vancouver CMA Rental Market Report (1996). 98 the provision of transit passes or credits to employees for walking and cycling are considered to be taxable benefits. Public transit and support for alternative transportation modes were excluded from eligibility in the shared-cost infrastructure programs initiated after the 1993 federal election, while spending on roads and sewers was encouraged.170 The effect of federal policies are subtle, but can strongly influence the levelness of the land use and transportation "playing fields." 2.5.3 Provincial Policies The B.C. Transportation Finance Authority (BCTFA), as the primary funding agency for major transportation projects in the province, coordinated two major transportation programs with far reaching implications on the Lower Mainland since 1995, "Going Places" and BC Transit's Ten-Year Development Plan. These are transportation policies with a high land use content. Going Places: Transportation for British Columbians The "Going Places" document embodies the BC government's current policy on transportation. The role of transportation in the province is stated as:171 • Provide people with access to goods, services, jobs, and recreation; • Maintain the competitiveness of industries; • Promote economic growth; • Shape growth patterns and land use; • Improve air quality and the overall quality of life. Note that land use, livability and environmental concerns all appear after mobility and economic concerns, perhaps reflecting the fact that a large number of constituencies in the province are far 1 7 0 B.C. Premier Glen Clark negotiated the first exception to the program's scope in April 1997, to allow federal funds to be used for planning work on the future Broadway-Lougheed LRT line. 1 7 1 Province of British Columbia, Transportation Finance Authority Going Places : Transportation for British Columbians (September 1995). 99 removed from major urban centres and are relatively unconcerned about issues such as sprawl, congestion, pollution and housing affordability. A number of elements in the policy relate specifically to transportation in the Lower Mainland. Some of which have already been introduced. Specifically, the policy calls for: • Initiation of commuter rail in the Fraser Valley; • The creation of an H O V network through construction of additional lanes; • Expansion of BC Transit service (see next section), including the construction of an LRT line following the Broadway/Lougheed corridor from Vancouver to Port Coquitlam; • The relief of congestion on inter-regional routes; • A limited bicycle infrastructure program, with costs shared 50/50 with municipalities; • Coordination of land use with transportation infrastructure;172 • New road capacity for movement of people and goods. Many of these initiatives reflect those contained in the regional Transport 2021 plans discussed later. Coordination with land use is again near the bottom of the list. There are some notable differences from the region's policies that have important impacts on growth management: • The Broadway - Lougheed corridor is to be the initial LRT corridor constructed instead of the New Westminster - Coquitlam corridor. This decision will substantially delay the introduction of rapid transit needed to shape growth patterns in the fast-growing Northeast Sector; • Commuter rail from the Vancouver CBD to Mission was fast-tracked, with no stops provided in the regional core, at an initial capital cost of $180 million. The Livable Region Strategy sought minimal growth in the North Fraser subarea. Only 6000 regular passengers are accommodated, each of the five trains makes a single, two-hour return trip per day and the service reported an operating loss of over $30 million in 1996-97, the first year of operation; 1 7 2 Blair Redlin, President and CEO of the T F A in an address to UBC School of Community and Regional Planning students in the Transportation Planning course on 28 November 1995, stated that there would be no LRT development until municipalities along proposed corridors adopt zoning bylaws compatible with the Growth Strategies Act, as embodied in the GVRD's Livable Region Strategy. 100 • Emphasis on improving road system performance, reducing congestion, and expanding capacity over the discouragement of automobile use through financial "sticks." No new tolls, taxes or plans to tie insurance premiums to distances traveled have been announced; • Pollution concerns are addressed primarily by stricter emission controls on new vehicles,173 as opposed to aggressively promoting shifts to alternative modes; • Construction of additional roadway capacity for H O V lanes on major regional routes instead of re-allocating of existing roadway capacity;174 • Addition of new H O V lanes to proceed before RapidBus service available to regional centres. "What do we do with sterile, land gobbling single family house subdivisions in Maple Ridge, Mission, Langley and Abbotsford? We put in West Coast Express and H O V lanes along the 401 that all of us as taxpayers must finance in order to subsidize the unsustainable, which ironically, will now actually encourage more of the same - since more people will now see it as feasible to live out there and work in the City." ... Government M L A The policy does reaffirm the commitment to coordinated land use through cooperation with local governments, but neither the details of the mechanisms to be used to achieve this goal, nor a timeline to have these mechanisms in place, have been provided. BC Transit JO Year Development Plan BC Transit's development plan was developed in conjunction with Going Places to provide more detailed information on transit's future role. Improved transit is cited throughout official policy documents as the priority alternative transportation mode and has been identified as the only mode that can realistically support desired land use patterns and to challenge the automobile in the short- and medium-term time frame.175 The LRT component has the most important long-term 1 1 3 California-style emission controls were announced in December 1995 by Moe Sihota, BC Environment Minister. 1 7 4 GVRD/MoTH Transport 2021: A Long-range Transportation Plan for Greater Vancouver (September 1993): "High-occupancy vehicle (HOV) and bus priorities," p. 33. 1 1 5 Assistant City Engineer Peter Judd, the head of Vancouver's Transportation Planning Project, responded to a city councillor's request during hearings on the plan from a Vancouver to outline all of the alternatives to reduce car use stated: "There is simply no alternative to better transit service." 101 role to play in the shaping of regional land use, and has resulted in two major background 176 177 178 reports. ' The plan has three main goals: • Increase the number and proportion of people who use public transit; • Shape urban growth and help reduce sprawl; • Ensure people are well served by transit, especially those who do not drive. The plan stresses that future success is dependent on improvements to the road infrastructure for transit vehicles and the provision of new LRT lines within ten years. It is expected that bus priority measures such as H O V lanes on major arterials and queue jumper lanes at major bottlenecks such as bridgeheads will be constructed, which is consistent with Going Places and the GVRD's Transport 2021 plans.179 Simple technological improvements, such as bus activated traffic signals and improved schedule and route information services, are also to be provided, although details are not available. The simplest measure of all, legislated priority for buses merging back into traffic after picking up passengers, used in many jurisdictions for over a decade, is described as an "innovative measure." While overall ridership appears to have stabilized at 130 million (+/- 5 million) revenue passengers, it has not kept pace with population growth, with per capita rider ridership decreasing 12% in the period between 1986-1991.180 It is difficult to identify a single factor, although car ownership, significantly decreased overall urban density, and a reduction in provincial transit subsidies have contributed to the decline.181 The Ten-Year Development Plan forecasts an increase of over five million revenue passengers during each year of the plan, which is slightly above the rate of increase in the overall population. It should be noted that the addition of new n 6 B £ T Y a ^ L o n g Range planning Department Summary of Intermediate Capacity Transit System Studies in Greater Vancouver (March 1995). 1 7 7 Province of British Columbia, Crown Corporations Secretariat Multiple Account Evaluation of Rapid Transit Options in Greater Vancouver (May 1995). 1 7 8 BC Transit In Transit: A Ten-Year Development Plan (Fall 1995). 1 7 9 Vancouver's Transportation Plan, approved in May 1997, made the assertion that there were no additional locations in the city that would benefit from bus-only lanes for at least the next six years. 1 8 0 Canadian Urban Transit Association Transit Operating Data (1986, 1991). 1 8 1 Jeffrey Patterson, "Urban Public Transit and Sustainable Cities", Sustainable Cities, Spring/Summer 1993. 102 road capacity for HOVs on several major congested commuter routes and reduced air pollution concerns due to tighter emission controls may have a significant impact on the ability of BC transit to attract new riders before new LRT routes are in service well into the next century.182 Multiple Account Evolution of Provincial Transportation Investments The B C T F A has established three basic principles that should applied when policy is formulated:183 • Make better use of existing facilities and manage demand better in order to reduce the need for new facilities; • Be more strategic in the investments made in transportation infrastructure to ensure that the high priority projects are undertaken in the most effective way; • Be more innovative in the way facilities and services are provided in order to reduce costs to users and taxpayers. The B C T F A has required all major new transportation projects undergo a evaluation according to the province's Multiple Account Evaluation Guidelines, presumably with these three principles in mind. 1 8 4 M o T H staff have confided that this requirement has often been deferred until after a final decision on a project was already made. The guidelines are intended to ensure that non-monetary factors, the "accounts," are considered equally in the decision-making process. As stated earlier, the method previously employed was a modified cost-benefit analysis, which considered the major benefit to be time savings for drivers and the major non-monetary cost to be social equity. The guidelines are intended to provide a framework to aid governments in the systematic identification and evaluation of the social, environmental and economic impacts of alternative courses of action. Examples of "accounts" used for evaluation for the LRT system included:185 1 8 2 Glenn Bonn, "Tough low-emission standards unveiled for new cars", Vancouver Sun, 08 December 1995. 1 8 3 Province of British Columbia, Transportation Finance Authority 94/95 Annual Report (1995). 1 8 4 Province of British Columbia, Crown Corporations Secretariat Multiple Account Evaluation Guidelines (1995). 1 8 5 Province of British Columbia, Crown Corporations Secretariat Multiple Account Evaluation of Rapid Transit Options in Greater Vancouver (May 1995). 103 • Financial. Net Present Value of costs and revenues using an 8% discount rate. • Customer service. Ridership increases and time savings. • Environment. Air pollution and other benefits from reduced car use. • Urban development. Contributions to regional land use goals. Some of the criteria developed to evaluate these accounts appear to be highly qualitative and subjective. In particular, the "urban development" account used to justify LRT on the Broadway-Lougheed corridor over alternatives such as RapidBus on the Richmond corridor, seems to be little more than a guesstimate.186 Nevertheless, as experience with the guidelines increases within various government ministries, and policy analysts become more familiar with the environment and urban development accounts, better land use and transportation coordination may result. 2.5.4 Regional Policies As indicated previously, Regional Growth Strategies should address transportation related issues such as the avoidance of urban sprawl, encouragement of walking and cycling, promotion of land use patterns that minimize the need for automobile use, economic development, efficient movement of goods and people, provision of affordable housing, and energy conservation. The LRS has several basic high level goals, all of which are directed at containing urban sprawl: • Protect the green zone. There is a strong desire to preserve the Agricultural Land Reserve (ALR) and to protect existing natural areas and park land. • Build complete communities. Provide mixed land-use to provide work, living, shopping, and recreation possibilities within a community. Ibid p. 15. RapidBus, for example was given a "0% probability" of generating adequate development, although this technology is the basis of the transit system in Ottawa, Ontario and was used to create the highly compact form of Curitiba, Brasil. Curitiba has the same population as Vancouver and has been often been cited as one of the best examples in the world of compact, transit oriented development. See Jonas Rabinovitch and Josef Leitman "Urban Planning in Curitiba," Scientific American, March 1996: pp. 46-53. 104 • Achieve a compact metropolitan region. Develop a network of regional town centres instead of a single metropolitan core. • Increase transportation choice. Reduce the dependence on the private automobile by increasing opportunities to walk, cycle, or take transit. Transportation and land use are seen as key factors for the achievement of these policies. The land use and transportation goals and objectives of the GVRD and M o T H for the Lower Mainland, which influenced the development of Transport 2021 and the LRS, were based on the principles of Creating Our Future, including:187 • Livability. Minimize the impact of transport on the quality of life; • Economic Development. Ensure that transportation systems support and promote desirable regional social and economic development; • Land Development Interaction. Ensure that transportation systems are compatible with, and promote, regional development plans; • Social Equity. Ensure the equitable distribution of transportation services and costs. The LRS specifically calls for significant changes in the projected population and the numbers of jobs in several key subareas. Transportation infrastructure and zoning are seen as the major policy levers. For example, Surrey and the Northeast Sector are expected to increase their population growth rate and dramatically increase their job rate. The South Region and the Langleys are expected to decrease their population growth rate and increase their job rate. While the urgency of these two factors is underlined, there is unfortunately no indication as to how land is to be released for affordable housing closer to employment or how employers are to be enticed to create jobs into the outer suburban areas. GVRD/MoTH Transport 2021 Technical Report 1: Goals, Objectives and Criteria for Developing a Long-range Transportation Plan for Greater Vancouver (May 1993). 105 LRS studies were careful to state that the relationships between transportation infrastructure and land use shaping effects are not fully understood.188 Nevertheless, the necessity of coordinating land use and transportation emerges throughout GVRD planning documents. The reports recommend that transportation investments should be prioritized in favour of those projects most likely to result in desired land use patterns, in particular those that will create a compact urban form, although there is little elaboration of which methods should be applied to determine this.189 Transport 2021 was a jointly funded project between the GVRD and M O T H . The project's mandate was to recommend a comprehensive series of transportation plans for the GVRD in support of the LRS. All of the GVRD's member municipalities had direct input into the development of transportation policies. The results of the project were the Interim Improvements Plan (to 1995), the Medium-range Plan (to 2006), and the Long-range Plan (to 2021). All the plans seek a modest increase in the modal share of transit, while increasing occupancy rates of the automobile and decreasing the need to travel through strategic investments. Transport 2021 Interim Improvements Items in the short term plan were intended to fill the gap until the medium-range plan provisions came into effect. The criteria for selected improvements were: • No negative impact on long range land use goals; • Consistency with modal priority goals, i.e., transit, HOV, and then mixed use; • Enhance system safety, reliability, and continuity. It would appear that the emphasis of these interim construction projects was on the "continuity" criteria. The financial expenditures on increased road capacity in the last two years will almost There does seem to be an enormous body of knowledge, however. A simple search of the UBC library catalogue for the keywords "transportation" and "land use" results in 81 references, most directly related to the land use shaping effects of transportation. 1 8 9 GVRD/MoTH Transport 2021: A Medium-range Transportation Plan for Greater Vancouver (October 1993): "Conclusions on growth management and transportation," p. 10. 106 certainly fuel residential growth in the Fraser Valley and reduce the ability of providing alternative modes to compete, particularly transit. Selected construction projects included widening of the Mary Hill Bypass in Coquitlam and the Barnet Highway in Burnaby and Port Moody, adding counterflow lanes on the Pitt River Bridge connecting the North Fraser subarea to the regional core and the removal of the last controlled intersection on the Trans Canada highway between Hope and Horseshoe Bay in North Vancouver.190 Transport 2021 Medium-range Plan Due to the considerable uncertainties in forecasting long-range system requirements over a thirty year time period,191 particularly with available transportation models, the medium-range plan appears to be the focus of the Transport 2021 study. The base year for comparison was 1991, and the plan is oriented to the period 1995-2006. As the medium-range plan covers most components of the long-range plan, the major common elements of both plans are outlined here: • Growth Management Using Transport to Shape Growth: • Make transportation investments which encourage inner suburbs; • Coordinate land use and transportation. One should not lead the other; • . Withhold capital investments until appropriate local land use policies are in effect; • Expand the mechanism of partnership agreements between levels of government. • Transportation Demand Management. The report stresses the importance of public education before the implementation of any measure seen to be a "stick", e.g., taxes and tolls. Preferred T D M measures include: • Promote telecommuting; • Establish employer programs to reduce work trips by automobile; • Give transit priority in traffic; • Increase parking fees; GVRD/MoTH Transport 2021: Interim Highway Improvements (November 1993). 9 1 GVRD/MoTH Transport 2021: A Long-range Transportation Plan for Greater Vancouver. (September 1993): Sources of Uncertainty," p. 54. 107 • Increase fuel taxes; • Introduce toll on bridges. • Improved transit: • Basic transit improvements; • Provide transit priority at approaches to water crossings; • Provide transit priority measures on roadways; • Provide Intermediate Capacity Transit Systems (ICTS, or L R T ) ; 1 9 2 • HOV network development. The report calls for the provision of H O V lanes, not necessarily the construction of new lanes. Alternatives suggested include the conversion of existing lanes or preferential access to some highways by HO Vs. • Improved mixed use roadways. The report emphasizes that improvements are needed to accommodate the movement of people and goods, but recognizes that some unspecified mechanism is needed to deter long haul commuting along these routes. The report recognizes the financial difficulty of implementing all recommended LRT corridors simultaneously. One significant factor is the debt remaining from the Sky Train system.193 As a result, it is accepted that implementation of system components will be phased in, which was shown in the theory section of land use and transportation interaction to be of limited value. The report also recognizes that there are conflicting objectives that must be weighed. If transport efficiency was the priority, the report recommends that the Richmond-Vancouver corridor proceed first. If land use shaping was the priority, the New Westminster-Coquitlam corridor was recommended. If funding for two routes was available, it was recommended that both the Richmond - Vancouver and New Westminster-Coquitlam LRT be constructed. 1 9 2 LRT can be a segregated ROW, at-grade (on-street), or Automated (ALRT) service with a capacity up to 10,000 passengers per hour. 1 9 3 GVRD/MoTH Transport 2021: A Long-range Transportation Plan for Greater Vancouver (September 1993): p. 47. 108 Commuter rail is mentioned, but not considered to be a priority component. The plan notes that transit already has a high share of the modal split along the Vancouver - Mission corridor, that it does not promote intensified land use along the route,195 and that it is relatively costly for the limited function it provides.196 The major advantages seen were support from the public and a relatively short lead time, which has proven to be the case since the service started in November 1995, less than a year after the decision to proceed with the project was announced. Transport 2021 Long-range Plan The long-range plan appears to be a minor extension of the medium-range plan. Elements introduced in the medium-range plan are elaborated, such as the full implementation of T D M measures, H O V lanes, LRT corridors, and mixed, higher-density land uses. One T D M measure that is stressed is the need to send clearer price signals to users. A "pay-as-you-drive" insurance system is cited as a specific recommendation. The plan also elaborates on the promotion of walking and cycling. These also received consideration in an associated technical report with parallel policy recommendations.197 The report provides examples of bicycle modal shares in North American and European cities far higher than those in Vancouver, and notes that there is considerable potential for increasing the catchment area of transit by the provision of bicycle parking at transit stations.198 The policy 1 9 4 GVRD/MoTH Transport 2021: A Medium-range Transportation Plan for Greater Vancouver (October 1993): "Appendix 1: Note on Commuter Rail." 1 9 5 David Smitfi "Mission train service puts developers on fast track", Vancouver Sun, 09 December 1995. Article describes a residential building boom in Maple Ridge and Mission, and how several Vancouver businesses, including city hall, are considering shuttle buses for train users. 1 9 6 Glenn Bonn "Train passengers get $20-subsidy, figures show", Vancouver Sun, 09 November 1995. Jeff Lee "Taxi ride's on taxpayers: $5 fare covered for Mission train users", Vancouver Sun, November 1995. Both articles argue that commuter train users are heavily subsidized and pampered compared to other transit users. 1 9 7 GVRD, Strategic Planning Department Greater Vancouver Regional Bicycle Task Force Policy Recommendations (November 1993) and GVRD/MoTH Transport 2021 Technical Report 15: Regional Transportation Implications of Neighbourhood-Level Planning Initiatives (August 1993): Section 5, "Pedestrian and Bicycle Alternatives." 1 9 8 In a 1991 brief to the Toronto Transit Commission, the Toronto City Cycling committee demonstrated that 90% of the residents of the city were within a ten minute bike ride of a subway station. After years of lobbying TTC staff, the Commission passed a resolution to immediately install of bicycle parking at all stations. 109 recommendation related to cyclists and pedestrians was that new and re-developed urban activity centres should be designed to provide nearby housing and safe routes for walking and cycling. 2.5.5 Municipal Policies At the receiving end of the many higher level policies is local land use and street networks, two jurisdictions which are predominantly under municipal control. It would be difficult to concisely summarize the land use and transportation policies of more than twenty municipalities in the GVRD. As a proxy, the transportation and land use policies at two geographic and philosophical extremes will be compared, the City of Vancouver and Langley District. Table 3 - Comparison of Vancouver and Langley District Characteristic Vancouver Langley District Density Medium to High Low Housing prices Very high Low to Medium Population High Low Buildable Land Low High Employment Surplus of Jobs Deficiency of Jobs Form Urban Rural Employment base Services, Transportation Agriculture Transit Use Medium Very low Auto ownership Medium High Growth Rate Low to Medium Very high Source: Author's observations. Vancouver's high level policy document is CityPlan and Langley's policies are embodied in its Growth Management Plan, both dating from 1995. 110 Vancouver In CityPlan, the city reaffirms its intention to remain the region's major employment centre while at the same time protecting the predominantly single family nature of the city's residential neighbourhoods from unwanted development. The primary policy direction related to a jobs-housing balance is "encourage continued job growth at a rate that helps balance the number of jobs in the city with the number of workers who live here."200 Theoretically, this should mean a drastic reduction in the number of new businesses allowed, since 50% more people now come to work in Vancouver than there are workers who live in the city. CityPlan actually calls for a significant increase in the total number of jobs, which would appear to invalidate the goal of balancing jobs and housing. The emphasis is on white collar jobs and, with the exception of port activities, new industry is primarily focussed on activities which are "city-serving." It is interesting to note that the plan's authors seemed unaware that the city still had significant numbers of residents who were not in white collar or service sector jobs.2 0 1 CityPlan promotes the concept of higher-density, mixed-use neighbourhood centres, which appear to be quite linear and focussed along major arterial streets. Many of the city's residential areas currently have no commercial areas nearby, which inhibits walking or cycling for basic goods and services while promoting car use. The "carrot" to entice neighbourhoods to accept this concept is the ability for current residents, both children becoming adults and workers entering retirement, to stay in their neighbourhoods throughout the life-cycle. Paradoxically, this is to be accomplished without imposing on the SFD character of existing neighbourhoods.202 The plan also supports the city's responsibility to absorb "a portion" of the region's projected growth. Planning Department, City of Vancouver CityPlan: Directions for Vancouver (1995). 2 0 0 Ibid. p. 30. 2 0 1 Ibid. p. 31, "Many blue collar workers still live in the city and CityPlan seeks to maintain a range of employment opportunities for all workers." 2 0 2 Ibid. p. 13 "New housing in neighbourhood centres best meets the changing needs of current residents while preserving most of the city's single-family neighbourhoods." I l l Relative wording such as "more," "less," and "some" figures highly in the policies intended to promote a mix of housing types, instead of absolute wording such as "adequate:"203 • "Subsidized housing will provide homes for some low and moderate income individuals and families; • Private developers will be encouraged or required to provide some less costly market housing; • Regulating demolitions and strata conversion to preserve some rental housing; • Allowing secondary suites to provide more affordable homes in some areas. " The policy falls far short of a commitment to provide a range of affordable housing types based on the incomes of those working in the city. In a sense, the city has guaranteed that the polarization of income classes and housing types while commuter traffic, particularly in the south and east sides of the city, will continue to grow. A "luck of the draw" approach seems to apply to affordable housing, whereby a limited amount of social housing will be made available to whoever is needy enough or fortunate enough to find it. The long lineups to view rental apartments in many parts of the city at the start of each month would indicate that those being able to find suitable, affordable rental accommodation may be in the minority. Although Vancouver's land use and transportation policies are considered by many to be progressive, many shortcomings are evident when they are compared to every day decisions by council and staff. Practically no ground-oriented or cooperative housing has been built in the city since the 1980s while the number of new condominium apartments in the city outpaced all other regional housing starts combined.204 Pedestrians are increasingly forbidden from crossing at major intersections, presumably for their own safety, instead of providing better crossing facilities and enforcement. Council abandoned a bike lane to the downtown core 1996 after only one week as a result of public and media pressure, although initial traffic delays disappeared after several days and significant numbers of new cyclists were observed. The vast majority of the city's bus Ibid. p. 22. See later research findings section on housing trends. 112 stops do not have even a basic shelter to protect passengers against the elements, presumably due to budget shortages, while millions of dollars are spent each year for street widening, computerized traffic signals and left turn bays. District of Langley Many of Langley District's land use and transportation policies could be used to write a textbook on cutting edge sustainable urban design for the 1990s:205 • "A community should include a well-planned mix of land uses to provide a full range of housing types, employment, educational and recreational opportunities and the range of goods and services that people need for daily living; • A community should include a mix of housing types, including a variety of housing densities, tenures and prices to meet the needs of all members of the community; • A community should be designed around a central node, with the highest density of commercial and residential development in the centre to encourage walking and transit use; • Each community and neighbourhood should be designed to be efficient and convenient for a variety of transportation modes; • A community should provide a balance between jobs and housing; • A community should be well-linked to other communities and to larger centres in the region by transit and walking and cycling links. " Unfortunately, the most fleeting visit to the District of Langley indicates that the last thing in anyone's mind at the moment is to produce a mixed-use community with a balance of jobs and housing, a variety of housing types or an urban form based on any transportation mode other than the private automobile. Single family houses, strip commercial development and isolated public amenities continue to be the rule. Traffic on Highway One leading to the regional core from Langley's residential communities is now bumper to bumper for many kilometers on every 2 0 5 District of Langley, Growth and Planning Commission Growth Management: Conclusions and Recommendations (November 1995). 113 working day. The most recent and flagrant contradiction to the district's growth management policies came in May 1997, when the District council announced that it no longer supported the principle of phased development to promote complete communities, and will start to allow development anywhere outside of properties within the Agricultural Land Reserve. 2.5.6 Summary Current land use and transportation policies affecting the Lower Mainland have a remarkably consistent veneer. It often seems that municipal, state/province or national land use policies could be picked up randomly from a library shelf for any jurisdiction in North America and the policy statements would be practically identical. In an interesting parallel to being Politically Correct, governments in the 1990s seem to feel a strong need to appear Planning Correct, at least in policy documents. Their common policies could be fairly paraphrased in the single statement "we will work towards providing a higher-density mix of land uses in complete, compact communities with jobs, affordable housing and transportation alternatives available for all local residents." Beneath the surface, a long legacy of market economics, professional planning attitudes and established engineering practices is embedded in restrictive zoning bylaws, outdated building codes and a dogmatic adherence to long-established administrative procedures. These act to reinforce the status quo and prevent innovative land uses, alternative development standards or human-scaled transportation modes from being implemented. In closing, it should be noted there has been a number of attempts to coordinate land use and transportation planning in the GVRD, particularly near stations along the SkyTrain ALRT route. Burnaby's Metrotown is often cited as the best example in the region. Higher density zoning, the proximity to a pleasant parkland, access to Metrotown or Patterson SkyTrain stations and a mix of jobs and housing have drawn a great deal of development to this location. Unfortunately, with 10,000 free parking spaces available to visitors, the area around the Metrotown complex forms one of the region's most formidable barriers to pedestrians and cyclists with only 15% of users arriving by transit. Three of Surrey's four SkyTrain stations are surrounded by large expanses of 114 at-grade parking lots which service nearby malls and shopping centres. It would appear that planners, engineers and decision makers at the provincial, regional and municipal levels still have a steep learning curve ahead of them before being sufficiently fluent in the concepts of effective land use and transportation coordination. "The automobile enjoys huge subsidies and governments continue to make decisions around development without the full knowledge of the costs that the automobile has on society as a whole." ...Transit Planner 115 2.6 Housing Policies Although Canadian housing policy this century could never be described as stable, it could currently be described as being in a complete state of flux. The federal government is negotiating with the provinces to hand over jurisdiction for most housing programs. As an example of the results of this instability, housing cooperatives established under section 96 of the federal National Housing Act (NHA) are finding that the Canadian Mortgage and Housing Corporation (CMHC), eager to off-load responsibility for housing cooperatives, is reluctant to provide financing for necessary repairs and renovations. At the municipal level, funding to construct non-market housing in the City of Vancouver has been essentially eliminated during the last two council terms. "Bonusing" schemes, where density increases and the relaxation of requirements are exchanged for social housing and community amenities in new projects, are increasingly being relied upon as a primary policy tool. Numerous policies and programs under a myriad of agencies and departments impact housing in some way. For example, the provision of transportation infrastructure by the province, including SkyTrain, commuter rail, and freeways, could be thought of as a housing subsidy for those who commute to far-away jobs from cheaper housing. The identification of these highly interrelated policies could easily fill another thesis. As a result, the following sections will focus only on the major housing policies and policy instruments related to the provision of affordable housing by the various levels of government. Housing policy, as it affects neighbourhoods and communities in the GVRD, is a juxtaposition of federal, provincial, regional district, and municipal government policies and programs. The policy instruments at the local level have the most direct impact on housing and can vary widely between neighbouring municipalities.206 For more detailed information on recent developments in local housing policy in B.C., see Hayley S. Britton Decentralization and Local Innovation: The Role of British Columbia's Municipalities in Affordable Housing Policies (UBC School of Community and Regional Planning, Masters Thesis, 1995) or Hayley S. Britton Decentralization and Municipal Housing Policy in British Columbia (UBC Centre for Human Settlements, Policy Issues and Planning Responses Working Paper P18, January 1996), which is a shorter summary of the key points contained in the thesis. 116 2.6.1 De Facto Policies It could be argued that there is no overall housing policy in effect at the moment. To be more accurate, there is no comprehensive set of policies that attempts to provide sufficient amounts of affordable, suitable housing in forms that promote sustainability. Those policies that do exist could be classified as: token financial incentives, removal of some existing regulatory barriers, or off-loading of the cost of providing new affordable housing units progressively from the federal level onto buyers of other new units. Financial incentives include: • Federal policies allowing pension funds to be used for down payments and decreasing down payments to 5%. These may have stimulated the construction industry, but did so by depleting individual pension savings and encouraging larger monthly mortgage payments by households. • Provincial policies of removing the property transfer tax for first-time homeowners and large homeowners grants. Ironically, with fewer young households being able to own their homes, the homeowner grant is increasingly acting as a subsidy from general revenues paid by all provincial taxpayers to a dimishing proportion of long-time homeowners. • The City of Vancouver policy of requiring developers to provide 20% of space in major new projects for non-market, primarily social, housing units. Intended to provide "free land" for affordable housing, usually in the form of airspace parcels, the physical construction of the units is dependent on dwindling funding from the provincial government. This has resulted in negligible amounts of new affordable housing for the "core-needy," while the policy has not been extended to assist other lower-income groups with affordable housing. Regulatory barriers include: • Exclusionary zoning such as low FSRs, limitations on permitted uses, and height restrictions; • Costly building codes such as sprinkler systems and excessive minimum parking standards; 117 • Requirement that new developments to pay up front the full costs of both infrastructure and community services. These had been provided without cost in earlier developments and were usually paid for by municipal bonds financed through general revenues. The lack of a political agenda for a housing policy backed by adequate funding and effective programs means that the de facto policy is to allow the price of housing to be governed by local resistance to development and basic supply-and-demand principles. This lack of political will to take direct action has been reflected in official statements.207 Exclusionary zoning practices in established municipalities serving as employment centres hampers the achievement of regional housing policy through limiting essential density increases. In this environment, land economics dictates that new affordable housing will be located in cheaper, low-density outer suburban municipalities with development-friendly councils. This statement is not intended to be an indictment of the development industry. Developers, as in any other business, seek to maximize their profits within a set of established rules, which they will obviously try to influence in their favour. The rules are ultimately defined by government policy, which is theoretically guided by the need to promote the greater public interest, which is not always in the interest of a particular sector of the economy or all individuals. Outside of a considerable federal support for the comprehensive monitoring of housing indicators, particularly for new housing market,208 housing strategy in Canada is a patchwork of disjointed initiatives with little coordination. As the following table shows, de facto housing policy is being shaped indirectly by a number of policies developed by various mandates. For example, CMHC Habitat II - The Canadian National Report (1996), Introduction, states "Canada will stress the importance of the private marketplace in meeting the housing needs of most people. A key role of government is to implement measures to influence and facilitate the effective operation of the market." The Province of BC has a similar hands-off approach, stating in Province of British Columbia, Ministry of Municipal Affairs, Recreation and Housing Affordable Housing Builds Strong Communities (1993) that "The intent (of provincial housing policies) is to allow each community to respond, in its own way, reflecting local conditions and points of view." This latter statement would appear to legitimize NIMBY attitudes and exclusionary zoning practices. 2 0 8 Chairs of CMHC are often prominent bankers and developers, which may explain the agency's historical interest in collecting new housing data. While this data was previously made available to the development industry at no charge, CMHC is moving quickly to a cost-recovery, subscriber-based system for market reports. 118 Table 4 - Examples of Indirect Housing Policies Level Department Program Impact Federal Public Works Infrastructure Incentives to build new roads, sewers, water lines. National Revenue RRSP mortgage withdrawal exemption, CMHC insurance with 5% down payment. Encourages high ratio mortgages with increased exposure to interest rate fluctuations. Provincial M o T H Going Places (HOV lanes, commuter rail) Provides subsidized access to cheap land in outer suburbs for residential sites. Environment BC Environmental Assessment Act Excludes major urban developments and streets from impact analysis. Regional Development Services Flat Rate DCCs Increases attractiveness of large lots. Strategic Planning Livable Region Strategy Growth earmarked for areas with severe jobs:housing imbalances and poor access. Municipal Planning Land use zoning, permitted uses Exclusionary zoning puts single family housing out of reach of most households in central locations. Engineering Parking standards, building codes Stringent sprinkler system requirements in new buildings and mandatory parking adds thousands of dollars to housing prices. 119 Official housing policy statements at the local level, as were required in OCPs by Bill 20 (1992), are liberally sprinkled with enigmatic expressions such as "provide a choice," "ensure a balance," and "establish a mix" followed by the expression "of housing opportunities." These expressions appear to be carefully crafted to avoid the need to establish ascertainable criteria against which housing objectives could be evaluated, leading to the laissez-faire environment discussed earlier. An example of an ascertainable criteria would be: "a stock of housing shall be maintained which satisfies the needs of households whose primary income earner is employed locally, in terms of size, amenities, access, and affordability" Few planning departments in North America would dare to propose such a change to their community plan at the current time. 2.6.2 Federal Policies Canada has been a major player in United Nations housing initiatives and sponsored the first U N Conference on Human Settlements in 1976 (Habitat I). At the Habitat II conference held in Istanbul during June 1996, Canada reaffirmed that decent housing was a basic human right as opposed to simply a marketable commodity.209 While not directly a federal jurisdiction under the Canadian constitution, the federal government has acquired varying degrees of responsibility for housing during this century. The Halifax Harbour explosion during the First World War has been cited as the stimulus for federal involvement in housing programs.210 At the request of the provinces and the general public, the federal government created a program under the auspices of the War Measures Act to provide some replacement housing for that which was destroyed by the blast. Although only a modest contribution, it was the foothold for increased federal involvement in housing. CMHC Habitat II - The Canadian National Report (1996): Introduction. The two themes of the Canadian policy framework are "sustainable human settlements development in an urbanizing world and adequate shelter for all." 2 1 0 George D. Anderson Housing Policy in Canada (Vancouver: Centre for Human Settlements, 1992): "Canadian Housing Policy 1917-1946," p. 3. 120 This initial step was followed by the Dominion Housing Act (1937) and the first National Housing Act (1938) which created CMHC's predecessor, the Central Mortgage Bank. The legislation also defined programs to support federal policy of using new construction as a stimulus for the stagnant national economy. The National Housing Act (NHA) was updated in 1944 to prepare for housing returning veterans and in 1949 a landmark agreement was reached between the federal and provincial governments to share the cost of joint housing programs 75%:25%. The 1954 version of the N H A has set the tone for subsequent federal housing policy, with the CMHC becoming the crown corporation responsible for insuring mortgages and supporting national housing policies. Throughout the 1960s and early 1970s, with the construction of single family detached homes decreasing, CMHC became more involved in funding for urban renewal projects, social housing, and non-profit groups such as housing cooperatives. The creation of the Minister of State for Urban Affairs (MSUA) in 1971 clearly demonstrated the federal government's interest in municipalities. M S U A was dissolved in 1978, due in part to provincial concerns that the federal government had over-stepped its constitutional authority. In the late 1970s and 1980s, federal policy moved towards using fiscal incentives such as tax breaks for Multiple Unit Residential Buildings (MURBs) which intended to address shortages in the supply of rental units in urban areas and the Registered Home Ownership Savings Plan (RHOSP) tax deduction to encourage first-time home owners to save money for an initial down payment. Most federal financial incentives have been eliminated, with the exceptions of revenue neutral measures such as reducing the minimum down payment on insured mortgages to 5% and allowing RRSP contributions to be borrowed as a down payment on a first home. Since 1993, CMHC has withdrawn from all non-profit housing programs whose costs were previously shared with provincial governments, although it still maintains a role in mortgage insurance and provides considerable research support for the building industry and local governments. 121 2.6.3 Provincial Policies "In the view of the commission, there is no question that the greatly varying cost of land is the most significant factor affecting the creation and price of housing. The one thing that the province could do to affect the price of housing is to ensure a plentiful supply of land." .. .Report of the Provincial Commission on Housing Options (1992) The BC provincial government has delegated most jurisdiction over local issues to municipalities, entities created under the Municipal Act. Although not formally expressed, the underlying provincial policy since the early 1990s has been to off-load jurisdiction for affordable and social housing as well, albeit without commensurate funding. Provincial infrastructure grants to municipalities have recently been reduced and attempts have been made to transfer responsibility for a number of provincial highways in urban areas to municipalities. This financial support may not appear to be directly related to housing policy. Nevertheless, the growing requirement for suburban municipalities to pay a larger share of the full costs of infrastructure internalizes onto municipalities development costs that were previously externalized. This could lead to increased reluctance to approve remote subdivisions that burden municipal infrastructure resources. A provincial commission undertook a major review of housing policy in the early 1990s, citing many of the arguments for the need for change that are presented in this thesis, including the social impacts on families and low-income individuals, the environmental impacts of continued sprawl and the economic impacts of high land prices on the construction industry.211 The limited powers of the Municipal Act before the 1990s have been identified as a barrier to the provision of affordable housing by municipalities.212 The commission's categorical conclusion was that the price of land, due to regulatory restrictions at the municipal level, was the main cause of high housing prices in the province's metropolitan areas. Many of the review's recommendations that applied to affordable family housing were implemented in the form of Municipal Act amendments 2 1 1 BC Ministry of Municipal Affairs, Recreation and Housing The Report of the Provincial Commission on Housing Options - New Directions in Affordability (1992). 2 1 2 BC Housing for the Ministry of Municipal Affairs, Recreation and Housing Affordable Housing Builds Strong Communities, Developing Community Housing Policies: A Guide for Local Government (1994). 122 which granted new land use powers and required OCP changes in the form of Bill 20 (1992), Bill 57(1993), and Bill 31 (1994). Under these provisions, the Minister of Municipal Affairs and Housing can impel a municipality to determine projected population and housing needs in its OCP and to state how these needs will be accomodated. The essential definition of "needs" was not provided, i.e., which needs are to be accommodated: those of the region, those of existing residents, or those of households who would like to become residents? To the author's knowledge, no municipality has ever been required by the Minister of Municipal Affairs and Housing or the Inspector of Municipalities to justify any of its housing projections or its policies to accommodate anticipated housing needs. Zoning bylaws created by a municipality, such as those which change density or permitted uses, must conform to the general policies of any existing OCP. This ensures that provincial housing policies will eventually be implemented mechanism, at least in theory, in the same way that bylaws conforming to OCP Regional Context Statements are intended to implement the province's other growth management policies. Bill 20 (1992) requires that OCPs include "housing policies of the local government respecting affordable housing, and special needs housing." Bill 57 (1993) authorizes a municipality to enter into negotiations with developers to grant density bonuses and other favourable zoning changes in exchange for the provision of affordable housing and other amenities needed by the community. Bill 31 (1994) allows municipalities greater powers to protect the stock of rental housing through protecting existing rental buildings from demolition and permitting the enactment of bylaws to establish minimum maintenance standards on these buildings. The City of Vancouver has similar jurisdiction through the city's enabling legislation, the Vancouver Charter. After several years of experience with these initiatives, the province issued a five point strategy for Affordable Housing in 1996, three of which are directly related to home ownership:213 2 1 3 BC Ministry of Municipal Affairs and Housing British Columbia's Strategy for Affordable Housing (April 1996). 123 • Build Partnerships to Broaden Participation. This appears to be simply an information gathering and distribution exercise. The focus is on telling stakeholders, other than the province, what they can do to help make housing more affordable. • Reduce Land Costs. The cornerstones of this initative are to allow public land to be made available at below market value for housing and to enable municipalities to negotiate with developers to exchange density for affordable housing and community amenities. • Increase Affordable Ownership. This is a collection of minor initiatives, such as streamlining administrative procedures, reviewing development standards and providing limited financial assistance for first-time buyers. The strategy sends the clear message to housing stakeholders that the province intends to play only the role of facilitator, leaving ultimate responsibility for the implementation of affordable housing policies in the hands of regional districts, municipalities and market forces. 2.6.4 Regional Policies The GSA enables the development of Regional Growth Strategies that will provide housing for projected future populations in a sustainable manner. It is stipulated that this housing should be "adequate, affordable, and appropriate." Besides establishing voluntary targets for population and employment, the LRS's specific housing policies call for the GVRD Board to: "Seek, through partnerships: • A better balance in jobs and labour force location throughout the region; • A diversity of housing types, tenures, and costs, in each part of the region in balance with job distribution; • The identification offurther opportunities for the location of ground-oriented housing; • Achievement of adequate population and employment densities in centres and transportation corridors to support planned transit services. " 124 While the GSA and LRS establish high level growth management goals, general, high-level statements such as these demonstrate the urgent need in B.C. for a set of clear, ascertainable growth management guidelines and fixed targets to guide the development and evaluation of OCP Regional Context Statements. An independent, arms-length review board, such as Oregon's L C D C , would help to interpret and ensure uniform application of these guidelines. 2.6.5 Municipal Policies As in the Transportation and Land Use Policy section, housing policies for the regional extremes of the City of Vancouver and the District of Langley will be compared as a proxy for the range of municipal housing policies in the GVRD. City of Vancouver Under the Neighbourhood Housing Variety section of CityPlan, the general direction is to "increase neighbourhood housing variety throughout the city, especially in neighbourhood centres." Specific directions related to housing affordability include: • "Continue to provide new housing near downtown jobs and ensure this housing is suitable for different ages and incomes; • Add more housing to single-family neighbourhoods in new forms. " Under the Addressing Housing Costs section of CityPlan, the general direction is to: "increase the supply of subsidized and lower cost housing throughout the city through the use of senior government programs, private sector incentives, and City regulations and subsidies. " 125 Specific directions related to affordability include: • "Maintain or increase the ratio of subsidized housing to market housing as the city grows; • Use incentives to encourage the private sector to provide lower cost housing, or require a percentage of new units to be more affordable; • Support actions to increase the housing supply, helping to minimize price increases due to scarcity." It should be noted briefly that a large variety of policy instruments are being used at the local level to promote affordable housing around the country. These programs take advantage of municipal jurisdiction in areas such as general land use policy, regulation (zoning and standards), joint partnerships, land banks, financing, taxation, and information distribution.214 Specific regional examples from the City of Vancouver include the release of portions of the city's land bank resources for housing cooperatives, the expansion of areas currently zoned as single-family to two-family or multi-family zones to reduce land costs, and the "20% Rule" inspired by Bill 57 (1993).215 From reviewing these iniatives, it becomes clear that the wheel has been reinvented many times, sometimes in neighbouring municipalities, which cannot be very efficient. Under Vancouver's 20% Rule, a minimum of 20 % of the units in major new projects requiring rezoning must be designated for non-profit (core-need) housing, half of which must be designated for families. Vancouver's city council reserves the right to substitute other affordable housing types and target groups for specific sites with respect to housing need, neighbourhood mix, amenities, and services. If necessary, Vancouver will consider "pay-in-lieu" for the site to be used to support non-market housing in other locations, which allows the developer to build market value units in the place of the non-market units and return a share of the additional profits to the city. The policy has not yet been applied to affordable housing for other than core-needy income 2 1 4 David J. Hulchanski, Margaret Eberle, Michael Lytton, and Kris Olds The Municipal Role in the Supply and Maintenance of Low Cost Housing: A Review of Canadian Initiatives (Vancouver: UBC Centre for Human Settlements, 1990). 2 1 5 City of Vancouver, Housing Department The State of Social Housing in Vancouver (Vancouver, 1993). 126 groups, which has led to an income polarization in recently completed projects, which reflects the growing income polarization being observed on a city-wide basis. Percent Change in Population by Income Category (1981-1991) 100.0% Lower and Upper Classes are becoming polarized, especially in Vancouver 40.0% 20.0% 60.0% 80.0% 0.0% Class is disappearing! under $10,000- $20,000- $30,000- $40,000- $50,000- $60,000- over $10,000 $19,999 $29,999 $39,999 $49,999 $59,999 $69,999 $70,000 Household Income Source: Statistics Canada Figure 10 - Income Class Polarization in Vancouver and the CMA The 1991 Central Area Plan is more direct than CityPlan concerning the importance of housing in meeting overall land use goals, the relationship between land use, transportation and housing and the appropriate policy direction for housing:216 "Objective: Increase the amount of housing and create new neighbourhoods for a range of households, to add people and activity and to reduce the need to commute from outside the central area. 2 1 6 City of Vancouver Planning Department Central Area Plan: Goals and Land Use Policy (1991): p. 18, "Housing Policy." 127 Significance: ...Housingplays a transportation role. Providing more opportunities to live close to the region's largest employment concentration means fewer demands for major transportation facilities to take people to work from homes in the suburbs." The Central Area Plan, although progressive in many ways, does indicate that the transportation role of housing is subservient to a less commendable, NTMBY-serving, objective: to relieve the development pressures that are "threatening existing stable neighbourhoods." District of Langley The District's 1995 policy on housing is as generic as one could possibly imagine:217 "Include a mix of housing types, including a variety of housing densities (single family detached lots of various sizes, townhouses, apartments), a variety of tenures (fee simple, strata title, rental) and mixed forms such as secondary rental suites in houses to provide a wide variety and price of units to meet the needs of all members of the community. " The only other significant references to housing in the growth management policy is in the context of the need to keep development costs low through flexible design standards and to base developments around community centres and elementary school catchment areas. This last point is of interest in this research in that it implicitly assumes that new development will be focused on family households with young children, as opposed to a wide range of ages. On the contrary, Vancouver's policies seems to make no special provision for children in new housing and appears to be focused on couples, families with older children, and seniors. 2 1 7 Growth and Planning Commission, District of Langley Growth Management: Conclusions and Recommendations (November 1995) 128 2.6.6 Summary Housing policies at the federal and provincial levels have become increasingly irrelevant, due to a lack of real funding and an unwillingness to get involved in local land use decisions. Affordable housing is generally considered to mean social housing. Housing "assistance" for the middle class has been interpreted as relaxing conditions on access to personal pension savings for down payments and mortgage insurance for high ratio mortgages at the federal level combined with "homeowner grants" at the provincial level. Regional and municipal policies for affordable housing, primarily in the form of social housing, depend on small and dwindling funding from higher levels of government. Generic statements of principles (i.e., provide a mix of housing types with a variety of costs and tenures, etc.) form the basis for municipal policies with few hard numbers for targets and few comprehensive programs to back up policy statements. The impacts of these housing policies, or lack thereof, on emerging family households seeking affordable housing in the GVRD will now be examined empirically. "Governments should intervene less, in the sense of giving people more flexibility to use land at greater densities." ...Housing Analyst 129 2.7 Summary of Applicable Literature, Theory and Policies The preceding sections on applicable literature, theory and policies have brought together the most applicable elements of planning practices related to the effects of land use, transportation and housing on the achievement of regional growth management. The many close interactions between these policy areas were highlighted. Approaches to growth management in Cascadia were compared to demonstrate the significant gap between Oregon's top-down approach and B.C.'s concensual, bottom-up approach. All approaches recognize a relationship between growth management, affordable housing and the effective coordination of land use and transportation. Oregon has set ascertainable growth management goals and maintains an independent, quasi-judicial body to ensure the implementation of state goals at lower levels of government. In stark contrast, B.C. defines broad, general policy areas that should be addressed and exercises no control over growth management approaches agreed to at lower levels of government. Initial experience with the GVRD's Livable Region Strategy, the first Regional Growth Strategy under B.C.'s Growth Strategies Act, indicates that the GVPvD's member municipalities have publicly endorsed a common regional approach while many privately intend to continue a business-as-usual policy, stressing local objectives first. Land Use and Transportation impacts, tools and coordination were discussed in detail, primarily to demonstrate that current planning practices cannot continue if regional growth management goals are to be achieved and that there is a well-documented history of both effective and ineffective planning tools available. The common element in all successful approaches to growth management has been highly coordinated land use and transportation planning, combined with much higher densities and land use mixes than currently exist in most parts of the GVRD. Researchers in Washington State have shown that the use of alternative transportation modes is highly dependent on land use density and mix, while researchers in Oregon State have shown that housing affordability has been effectively maintained by the specification of minimum densities. 130 The influence of land economics and location theory was then presented to illustrate how the marketplace for residential and commercial development does not usually take official growth management policy pronouncements into consideration when decisions are being made. The utility of a location in terms of accessibility to workplaces, customers and amenities, in addition to price and other economic determinants, dictate the supply and demand parameters for most private sector location decisions. Land prices, the most variable element of housing costs, are indeed heavily influenced by public policy, notably through restrictions on land use density and mix, controls over building standards, and the provision of transportation infrastructure. These public policies have been dominated by local land use decisions designed to minimize neighbourhood resistance to intensification and provincial transportation decisions designed historically to accommodate long-distance commuting. Existing policies could be classified as de facto policies, or what is seen to be happening on the ground, and official policies, or what is the stated to be the preferred course of action. Official growth management policies call for regional districts and their member municipalities to address a wide range of issues to promote social, environmental and economic sustainability. De facto policies avoid the definition of clear and ascertainable objectives and criteria to support well-intended high-level goals, which has resulted primarily in parenthood statements with few enforcement mechanisms. Official land use policies and transportation policies could be concisely summarized as "promote complete, compact communities which discourage car use and promote walking, cycling and transit." With few exceptions, the de facto land use policy is to segregate high- and low-density uses locally and regionally in order to protect existing single family detached home districts from intensification. Only outer-suburban municipalities offer affordable and attractive ground-oriented housing as a result, perpetuating long-distance commuting by younger, family households. Token funding has been made available to promote walking and cycling, while transit funding has been cut and hundreds of millions of dollars continue to be poured each year into projects that increase road capacity and assist long-distance suburban commuting. 131 Officially, suitable, affordable housing is seen to be a basic right for all citizens by all levels of government. The de facto policy is that it is left to the marketplace to seek out opportunities in municipalities which do not practice exclusionary zoning, generally found in reclaimed, marginal land in urban areas for high-density development and at the outer edges of the region for low-density development. The development industry attempts to provide the most affordable housing possible given the limited availability of zoned and serviced land, and costly building standards, such as wide Right of Ways and minimum parking standards. Federal and provincial governments have effectively abandoned their roles in housing, although reluctant municipalities have been provided with a reasonable toolkit of legislated powers to make housing more affordable. These tools include old-fashioned control over density and use, as well as the ability to negotiate density increases in exchange for community amenities and affordable housing units. The following research findings sections provide empirical answers to the research questions posed earlier. In so doing, they also provide validation for the stated de facto policies, point out where observed trends vary from those outlined in official policies, and identify barriers and opportunities exist for bringing the de facto policies and trends back in line with official policies. 132 3. Research Findings 3.1 Survey on Land Use, Housing, and Transportation Issues 3.1.1 Survey Description In order to acquire an understanding of the prevailing attitudes towards the issues of land use, affordable housing, and transportation infrastructure raised earlier in the research questions, key informants consisting of senior representatives from government and the private sector were surveyed. Attempts were also made to obtain a cross-section of NGO views, but representatives of these groups generally did not feel comfortable with their level of knowledge on these issues. Respondents were encouraged to answer with "No Opinion" or "Not Applicable" if they did not have an established opinion or the necessary background knowledge in a specific subject area. Although given the option, most respondents did not request that their identity be withheld. Out of respect for those who requested anonymity and those who responded as individuals and not as representatives of their respective organizations, all responses presented here will be unattributed. Survey Questions I. How much importance do >ou assign to the following issues for their impact on growth management in the GVRD (e g , urban sprawl and long distance commuting)9 Why9 a. Providing affordable, ground-oriented housing near employment centres b. Balancing the number of jobs and employed residents in each GVRD subarea (I e , North Shore, Northeast Sector, Burrard Peninsula, Fraser Valley South, etc ) c. Achieving land use density targets that support walking, cycling, and transit 133 2. What is your definition of "affordable" housing? Do you feel that governments should be responsible for the provision of affordable housing'' 3. What barriers are there to the construction of more affordable ground-oriented, higher-density family housing near regional employment centres, such as the Burrard Peninsula9 4. Which of the policy instruments currently available to planners and decision makers for the creation of affordable housing are most effective9 Have they been fully exploited9 5. Should governments intervene more aggressively in housing and land use policy? If so. which level(s) of government should act and which programs should be implemented9 6. In general, which policy should governments follow transportation investments should lead development, or development should lead transportation investments? Why9 7. It has been claimed that some types of transportation infrastructure can act as a hidden subsidy in certain real estate markets For example, the U S. Interstate freeway system has been identified as a major factor in providing easy access to cheaper suburban land in metropolitan areas for businesses and homeowners Do you feel that there are examples of such hidden subsidies in BC 9 If so, which are the most significant and how could these be made more equitable? 8. How aware are decision makers and the general public of the significance of and need for growth management? Of the influence that personal lifestyle choices and public realm decisions such as transportation mode preferences or housing location/mix have on growth management efforts9 Which mechanisms would be best for increasing awareness? A summary of responses for each question is included below. The complete set of responses, with any remarks that could identify individual respondents removed, is included as Appendix B. 134 3.1.2 Summary of Key Informant Responses Two-thirds of the 24 key informants asked to provide their expertise responded to the survey request. Responses were received from authorities at the municipal, regional and provincial levels of government as well as from academia, the private consulting field and the real estate and development industries. Their responses are summarized below. l a . The provision of affordable, ground-oriented housing near employment centres was considered by a large majority by respondents to be of high importance. The general feeling was that if there was a good mix of housing of various types and cost distributed throughout the region, households could move easily to the location which provided the best balance between their travel needs and other considerations. An associated theme was that the current distribution of affordable, ground-oriented housing often places severe limitations on where households with average incomes are able to locate. The important problem of how affordable housing would be distributed in practice to those working nearby, assuming that significant amounts of affordable housing could be provided near employment centres, was raised by several respondents. l b . The importance of having a jobs/housing balance in each subarea of the region received mixed reviews. While most agreed that the principle was laudable, a number of respondents felt that such a balance would be difficult, if not impossible, to achieve. Reasons cited included lifestyles, the growing need for two-income earners in households and the increasing mobility of workers. As one respondent put it, "its much easier to change your job than where you live." The need for a balance of housing types and affordability which matches the income profile of those living in the a subarea, as opposed to a simple numerical balance, was also raised. l c . The achievement of higher density targets that support walking, cycling, and transit received uniformly strong support. The viability of the transit alternative as opposed to the automobile, the achievement of complete, compact communities and the possibility of realizing significant savings 135 in the costs of infrastructure, were cited most often by respondents as the reasons why the importance of higher density in the region should be stressed. 2. One of the more interesting results of the survey was that the majority of respondents assumed that the term "affordable housing" meant social housing for the core-needy. There was general support for government intervention in this particular area of affordable housing. Of the few who interpreted the term "affordable" more globally to be inclusive of households with a wide range of incomes, only one respondent held a strong belief that affordable housing was a basic value that governments should actively be involved in. 3. In spite of the ambiguous responses to the "affordable housing" question, there were very clear and uniform responses identifying the barriers to "more affordable," ground-oriented housing. These were best described in one respondent's concise answer: "High land cost. Neighbourhood resistance. Inflexible zoning." High D C C s were also mentioned by several respondents. The NTMB Y syndrome appears to be at the heart of the issue for respondents from all backgrounds. The answer given above could reasonably be expanded to: "High land cost results from the lack of land zoned for higher density and mixed uses which results from neighbourhood resistance." Another prevailing attitude was that the established majority, presuming that it is entitled to lower density housing, feels that finding affordable housing is the problem of the non-established minority. Moreover, the established majority feels that the non-established minority should pay up front for the cost of the required infrastructure and amenities for new developments that were previously shared by the entire community. 4. With respect to the tools available to planners to make housing more affordable, the use of density bonusing, the release of public land at below-market prices to non-profit groups and basic zoning were routinely identified. Respondents were unanimous in their belief that these tools were not being fully exploited. Zoning in particular was often mentioned as the single most effective tool, given that land prices are overwhelmingly the largest barrier to affordable housing in the region. Zoning was also cited as the most difficult tool to use due to potentially strong 136 public resistance. The notion of area-wide zoning was mentioned, whereby higher density and mixed uses would be prescribed over wider areas of the region to avoid localized resistance. 5. The question of the appropriateness of government intervention provided quite.a variety of responses. On one side were the non-interventionists, who argued that over-involvement of governments in land use controls had started the problems of low density housing and segregated land uses in the first place. This camp suggested that the worst thing that could happen for affordable housing would be for the government to take too much interest in the issue. On the other side were the facilitators, who supported a government-led enabling strategy to overcome identified barriers to affordable housing, such as: • Provide information and training for cost-effective development and building standards; • Mediate the often counter-productive competition for growth between regional municipalities; • Legislate statutory tools which allow municipalities to effectively provide affordable housing; • Promote a climate in society where the average citizen understands and supports the need for more affordable housing and compact communities throughout the region. A consensus was reached that municipal governments should become more proactive in providing sufficient amounts of suitably zoned land for construction and that senior governments should again be more generous in financing housing for low-income households. 6. The choice between transportation leading development or development leading transportation was not intended to present respondents with a "chicken and egg" enigma to solve, but to probe their awareness of the interaction between these two factors. It was, as one respondent guessed, a trick question. It was nevertheless interesting to receive several responses stating that either "Transportation should lead" or "Development should lead" followed only by words to the effect of "for obvious reasons." A majority of respondents, while usually showing some preference, did indicate that the two factors should be coordinated, or proceed in "staggered parallels" as one respondent put it. In addition to responses indicating that the two factors were independent, the 137 lack of understanding by some professionals was raised by an experienced strategic planner: "I am still concerned that the land use-transit connection is very fragile and needs constant nurturing." The most interesting responses on this question were from two knowledgeable persons working at a high level within the provincial government. One, involved with MoTH, cited commuter rail and HOV lanes as excellent examples of the appropriate transportation investments needed to shape desirable growth patterns. The other, involved with the Ministry of Municipal Affairs and Housing, stated that commuter rail and HOV lanes were good examples of unsustainable and costly solutions that would reinforce the tendency towards urban sprawl by subsidizing long-distance commuting. 7. The potentially contentious question of hidden subsidies and equity resulted in the greatest diversity of responses. Most respondents felt that transportation infrastructure investments did, to a greater or lesser degree, subsidize housing. Examples cited included the West Coast Express, bridges and tunnels with no tolls, freeway expansions, low automobile operating costs and artificially low property taxes in some municipalities. One respondent from the private sector stated that developers are subsidizing most transportation infrastructure. Several thought provoking comments were provided which seem to challenge the conventional wisdom: • "We have little understanding of the true costs of sprawl, which makes it hard to determine who should pay and what amount they should pay; • The de facto acceptance of urban sprawl and automobile dependence by individuals and governments may be an unfortunate, but nevertheless valid, confirmation that society values the ownership of ground-oriented, detached housing highly and is willing to provide subsidies; • The use of general revenue to pay for transportation infrastructure should be considered to be a subsidy not only for suburban home owners, but also for home owners in urban areas closer to job concentrations, as this relieves the pressure on them to accept more density. " 138 8. The final question, concerning the awareness of decision makers and the public of the need for growth management and the influence of personal decisions on growth management, also provided a wide divergence of opinion. Responses were spread evenly between the public being "not very aware at all" to decision makers being "very aware." Several stated that public awareness was at the level of "there's a lot of growth happening, I don't like it much, but I guess we're going to have to do something about it, as long as it's not me." Others indicated that awareness of the issues and personal lifestyle implications is quite a recent phenomenon, but one which is growing rapidly. A few respondents made the disturbing observation that many decision makers are not nearly as aware of the underlying causes of the problems as they think they are, which is leading to inappropriate actions. Options for the education of decision makers and the general public included public forums, op-ed pieces in major print media, and involvement in community planning processes. Surprisingly, only one respondent proposed using the potential of the school system to increase awareness of the issues of land use, housing and transportation. Another suggested that if more councils had the courage to make progressive zoning changes, the public hearing process could be used to educate the public on why such changes were necessary to meet objectives. Two respondents made very convincing cases in favour of the need to address what could be called "densi-phobia," through public education, to counter fears of higher density. Crime, lack of privacy and the paving over of greenspace top the list of these fears. Poor development standards in some GOMD housing projects may have justified the latter two fears. These respondents stated that good examples of higher density should be highlighted and development/building standards strengthened to achieve livable, high-quality, medium-density housing.218 'There's a lot of growth happening, I don't like it much, but I guess we're going to have to do something about it, as long as it's not me." • ..Growth management official's view on the public's attitude to change 2 1 8 See, for example, GVRD, Strategic Planning Department, Housing Task Group Examples of Ground-Oriented, Medium Density Projects in Greater Vancouver (July 1996), funded by the GVRD and the Real Estate Foundation. 139 3.2 Population and Job Growth Trends The heart of the LRS is the concept of directing new jobs and population into a Growth Concentration Area. According to the LRS, the types of growth trends and targets that are needed to achieve the strategy's goals include: Table 5 - Numerical Targets for Growth Management Subarea Target Share of Excess or Target Job Share of Growth Shortfall of Growth (to Growth (%) Trend (%) Growth (%) nearest 50 %) Southern Region/Langley 8 29 21 100 South Fraser Valley 3 10 7 50 Vancouver 9 13 5 50 Richmond 5 10 5 50 North Shore 2 4 2 50 North Fraser Valley 6 7 1 150 North East Sector 20 10 -10 300 Burnaby/New Westminster 16 4 -13 100 North Surrey/North Delta 30 12 -18 450 Source: GVRD Livable Region Strategy: Proposals (August 1993): Growth Management Targets, and Transport 2021: A Medium-range Plan for Greater Vancouver (October 1993): Numerical Performance Targets for 2006 - Land Use Targets. It is important to note that, while the growth trends in may not be reflecting the targets of the LRS in the Burrard Peninsula, Surrey or the Northeast Sector, Surrey and the Northeast Sector are experiencing the region's most significant long-term growth, but only in terms of population: 140 Growth (%, 1981-1991) • 57.2 to 73.5 (3) Source: Statistics Canada, 1981-1991 census data. Figure 11 - Population Growth (1981-1991) Growth (%, 1991-1996) • 27 to 50.8 (1) Source: Statistics Canada, 1991-1996 census data. Figure 12 - Population Growth in the Study Area (1991-1996) 141 The second map confirms that the general patterns of the 1981-1991 trends are continuing unabated, with the focus of growth shifting closer to the geographic centre of the Fraser Valley during the 1991-1996 time period. The only significant regional change was the U E L , where the UBC Real Estate Corporation has been creating a new residential community. The highly localized growth in the U E L , shown as black, makes the strong growth in other municipalities appear slightly weaker than the 1981-1991 period. It should be noted that these data were accumulated before the effects of recently-completed large-scale investments in transportation infrastructure in the Northeast Sector have become evident, such as the Barnet-Hastings H O V lanes, commuter rail, Mary Hill Bypass widening, and Pitt River Bridge counterflow lanes. These are discussed further in the next section. While the rate of population growth in suburban areas is strong, the essential complementary growth rates in employment are far from meeting the targets. In many cases, the rate of growth in employment is not keeping pace with the rate of growth in the residential population. In other words, the jobs/housing imbalance in many of the outer suburbs and exurbs is worsening, perpetuating and accentuating their roles as the "bedroom communities" for employment centres in the regional core. Further, the highest growth rates are being experienced in the central and eastern Fraser Valley where the impact on the A L R is greatest. It is worth noting that Richmond has not only consistently high growth rates, but also has the region's highest ratio of jobs to employed residents. This observation is due in large part to the fact that the Vancouver International Airport, a major regional employment centre, is situated in the northwest corner of city. The airport provides over 15,000 permanent jobs, equivalent to one job for every ten persons living in the city.2 1 9 In fact, large numbers of workers commute in and out of the city, which reflects the high housing prices in the municipality. 2 1 9 Vancouver International Airport Authority 1995-2015 Draft Master Plan (July 1995): p. 33, "Employment at YVR." 142 The ratio of jobs (Employment, EMP) to employed residents (Employed Labour Force, ELF), a proxy for housing units, paints a clear picture of the imbalances of jobs and housing in the region: EMP to ELF Ratio • 1.058 to 1.372 (3) Source: GVRD EMME/2 Transportation Model Database Figure 13 - Ratios of Employment to Employed Residents (1991) The trend is also important. In the bar chart below, wherever the grey bar (change in employed residents) exceeds the black bar (change in employment) and the white bar (EMP7ELF ratio) is less than 0.8, the jobs/housing imbalance was significant and worsening in 1991. West Vancouver, Port Moody, Coquitlam, Surrey and Pitt Meadows/Maple Ridge fall into this category. West Vancouver also had the dubious honour of being the only GVRD municipality to register an absolute decrease in the number of available jobs. The relative size of a municipality should also be considered as a factor in identifying the importance of a jobs/housing imbalance. The villages of Anmore and Belcarra have similarly bad jobs/housing imbalances, but amount to 0.1 % of the region's population and do not have a major regional impact. An analysis of trends in the regional balance of jobs and housing and the required direction appears in a table below. 143 Jobs (Emp) to Labour Force (Elf) Balance Trends (1981-1991) 1.20 1.00 -0.80 -0.40 • 0.20 -0.00 • • Emp: (1991-1981)/1981 • Elf: (1991-1981)/1981 • Emp/Elf (1991) An Emp/Elf ratio between 0.8 and 1.2 is reasonable, a value of 1.0 being ideal. < s o -0.20 9- 2 Source: GVRD EMME/2 database. Figure 14 - Trends in Employment/Employed Resident Ratios Table 6 - Analysis of Trends in Jobs/Housing Balance (1981-1991) Key: Description: ft Gradual growth M Moderate growth fflrfT Strong growth => Little or no growth U Negative growth o Emp/Elf < 0.80 » 0.8 < Emp/Elf < 1.20 • Emp/Elf > 1.20 144 Municipality AEmp AElf: Emp Jobs/housing Groupings Elf Assessment Burnaby => » Good balance Coquitlam flft ftft O Need more jobs, less housing Delta ftftft ft O Favourable trend Langleys ftftft ftft o Need less housing Maple Ridge, ft ftftft o Need more jobs, Pitt Meadows less housing New Westminster => ft 1 Good balance North Vancouver City => => » Good balance North Vancouver Dist. ft =^ > o Need more jobs Port Coquitlam ftft ftft o Need more jobs, less housing Port Moody ft ft o Need more jobs, less housing Richmond ftft ft » Need more jobs in southwest Surrey ftft ftftft o Need less housing Vancouver/UEL => => • Need more housing West Vancouver U => o Need more jobs White Rock ftftft ftft o Need less housing Vancouver C M A ft ft » Need better regional balance Source: Author's estimation based on 1991 census data. 145 3.3 Transportation Trends The results presented in this section will be a collection of miscellaneous and occasionally conflicting observations, which reflects the state of transportation data collection in the region. M o T H abandoned using accidents/Vehicle Kilometers Travelled (VKT) statistics in 1994 due to the difficulty in collecting accurate distance data. BC Transit planners insist that there is excess capacity remaining on main routes in Vancouver during peak periods where simple observation shows that successive buses are full to capacity with standees every day and "pass-bys" are a common occurance.220 While the 1985 GVRD Travel Survey interviewed over 25,000 households and the 1992 GVRD Travel Survey interviewed over 15,000 households, the 1994 Travel Diary study approached only 1500 households.221 A wide range of broad conclusions on regional travel trends was derived from this small sample, in spite of the high probability of statistical error. For example, the 1994 survey claimed a -0.9% decrease in the 24 hour modal share for car drivers in the critical Northeast Sector between 1985 and 1994, but the error range for this value is given as +/- 3.6%. Initial data from M o T H has shown that vehicle traffic on the Barnet-Hastings 'Teople" Mover corridor through the Northeast Sector has increased almost 50% since the original two-lane Barnet Highway was doubled to four lanes in 1996, while transit ridership has decreased. The EMME/2 transportation model, used heavily by MoTH, the GVRD and municipalities, depends on "K" factors, which are correction factors needed to make the model's results match what is actually observed by traffic counts. Standard traffic counting instruments such as air tubes are also fallible. The Vancouver's engineering department was unable to provide accurate numbers for cars and cyclists during a 1996 bike lane experiment, due to instrument failures. A "pass-by" is when a bus driver judges that a vehicle has reached a safe capacity with standees and refuses to pick up further passengers. Bus routes leading to UBC, the region's second largest transit destination, has the highest rate of pass-bys in the system. 2 2 1 The Travel Surveys were a set of random telephone interviews in which questions about travel habits were asked, while the Travel Diary was a log book sent randomly to households, who were then asked to record their trips over a one week period and return for analysis. The two research methods involve different sets of systematic errors, indicating that their results should be compared with some caution. 146 Even in the case of extremely large, concrete infrastructure elements, such as bridges and freeways, MoTH statistical technicians, when approached for historical data from the last twenty years, admitted that they had never compiled a list of the province's major transportation projects. The list eventually had to be pieced together from a set of press releases, partial lists appearing in older publications and the recollections of long-time M o T H engineers. These comments are not intended to belittle the technical capabilities of traffic engineers, who have limited time and financial resources to make sense out of an enormously complex system. Over five million trips are made on an average day within the GVRD to hundreds of thousands of destinations on tens of thousands of lane-kilometers. They are intended only to suggest that, in spite of the claims of its practitioners, the field of transportation planning is far from being a precise science and is often based on a critical lack of both complete data and a connection to socioeconomic factors.222 The reader to therefore encouraged to focus primarily on broad trends that are being observed in travel patterns, as opposed to the details. "When it comes to accurate information on transportation trends to assist in decision making, we're really flying blind here..." ...Regional Transportation Planner In Walter Stewart The Paper Juggernaut (Toronto: McClelland and Stewart, 1979) it is demonstrated that data estimation problems are not limited to transportation engineers working on road traffic problems. The ill-fated Pickering Airport project in Ontario during the 1970s, which expropriated 35,000 acres of prime farmland, displaced hundreds of people and cost taxpayers almost half a billion dollars before cancellation, was based on year 2000 traffic projections that soared during the analysis phase from 25 million/year to almost 200 million/year. The planners overlooked the obvious implication that the estimate used would require every man, woman and child projected to be living in Metropolitan Toronto to fly about once every week. 147 3.3.1 Personal Travel Trends As the table below indicates, while the Vancouver C M A has been growing quickly, the number of commuting autos in the 1985-1992 time period was growing at a rate 50% faster than the growth in population. The growth in the number of trips made by car drivers was over 100% faster than the growth in population while the average auto occupacy was falling. At the same time, trip distances grew and trip speeds decreased. Overall, the share of trips by the dominant car mode grew another 4% in the morning peak period from 1985-1992 while the share of other modes decreased, transit by 12% and cycling by over a third. It is worth noting that the majority of the decreases in transit use and cycling were for school trips, with a corresponding increase in the number of automobile passengers for school trips. This implies that a much larger share of children are now being driven to school. The 1994 survey confirmed the trend and suggesting that it reflected concerns over safety on the streets. The share for walking remained constant. Table 7 - Changes in Travel Indicators (1985-1992) Transportation Characteristic* % Change (1985-1992) Population +21 Commuting autos +32 Trips by auto drivers +48 Trips by transit . +23 Total trips +38 Trips per person +14 Auto modal share +4 Transit modal share -12 Trip distance +12 Trip time +20 Source: GVRD 1992 Greater Vancouver Travel Survey 148 An enigma in observed growth patterns is that, if the regional population density is growing, transit ridership should theoretically be increasing at a higher rate. While regional density was increasing, the densities where many new households were choosing to live was decreasing. Many new suburban housing subdivisions are low-density "greenfield" developments, i.e., where no housing existed before. As the Vancouver Regional Transit Commission (VRTC) requires basic levels of service in these areas, the level of transit service available across the region has been effectively decreasing. Recall also that a critical density is required to improve transit viability. Further, in many denser urban areas, notably in Vancouver, transit ridership has reached the system's capacity during peak periods, which has inhibited increased usage and reduced the utility of transit as an incentive to choose a smaller, more expensive housing alternative in Vancouver over a larger, less expensive suburban choice. The 1994 GVRD Travel Diary Survey was intended to fill in some of the data gaps found in the 1992 survey. Trip diaries record each trip taken by members of responding households during a week- long period. The 1994 survey suggests that transit's morning peak period ridership losses observed in the 1992 survey have been recuperated, but notes that increases in overall ridership have been limited to the regional core in off-peak periods. A significant decrease in SOV travel is reported between 1985 and 1994 at times other than the morning peak period examined in the 1992 survey (-7.3% over a 24 hour period). In spite of a reduced walk/bike mode share in both the morning and evening peak periods, gains were observed in transit ridership (+25.4%) and the walk/bike modes (+17.6%) over a 24 hour period. The number of cars per household and the number of cars per capita decreased significantly, from 1.75/household to 1.54/household and from 0.67/capita to 0.58/capita. Half of these decreases in car ownership took place after the 1992 survey, when 1.63/household and 0.62/capita were reported.223 While these latest results are cause for optimism, they should be approached with some caution. Regionally, the growth in automobiles registered for work purposes matched the population 2 2 3 GVRD Strategic Planning Department 1992 GVRD Travel Survey, Report 3: Travel and Demographic Characteristics p. 8. 149 growth very closely, and the rates of increased automobile ownership vary dramatically across the region 2 2 4 Vancouver actually witnessed a negative growth in the total number of registered and insured vehicles in the five years between 1991 and 1996, in spite of a 12% population growth, while Coquitlam, Langley, and Surrey registered 29%, 24%, and 30% increases respectively in this time period. Of special note for this thesis is that the number of vehicles insured for driving to work increased at nearly double the rate of all insured vehicles. The relative values used primarily the 1994 survey report, i.e., increases and decreases given as percentages, overshadow major increases in traffic levels and the report provides little insight as to why such major trend changes might have occured. It is difficult to explain, for example, why the number of people walking and cycling in the off-peak afternoon period would almost double throughout the region in only nine years while the number of people walking and cycling to work was apparently decreasing. The lack of absolute numbers for changes in the number of trips per capita and the total number of trips per day prompted the author to do some forensic calculations. Based on raw data from the 1985 and 1994 studies,225 the absolute number of car trips in the region had increased from 2,925,000 motorized trips per day, i.e., car or transit trips, to 4,325,000 motorized trips per day. This represents an increase of 50% in a nine year period. When the automobile modal splits reported in the 1994 report are applied, it is seen that there has been an enormous increase of over 800,000 automobile trips per day in the region in less than ten years. In other words, while the region's population was growing at an annually compounded rate of approximately 2.50%, the total number of car trips grew at an annually compounded rate of approximately 3.75%, or 50% faster. This rate coincidentally mirrors the rate at which car ownership is exceeding the rate of population growth in the 1992 survey. Needless to say, such rates of increase would be completely unsustainable in the long term and are completely incompatible with government policies at every level to reduce car use and promote alternatives. Transport 2021 's targets for transportation, are informative: GVRD Strategic Planning Department Greater Vancouver Key Facts (July 1996): pp. 42, 65, 66. GVRD Development Services 1985 Metropolitan Origin-Destination Survey. 150 Table 8 - Transport 2021 Transportation Targets (AM Peak Hour Period) Goals/Criteria Baseline Target Target (1991) (2006) (2021) Trips: All modes 390,000 560,000 700,000 Car drivers 230,000 300,000 370,000 Car passengers 60,000 100,000 120,000 Transit riders 50,000 100,000 130,000 Pedestrian/Cyclists 50,000 60,000 80,000 Transit Modal Shares: To downtown peninsula 37% 45% 49% To regional town centres 13% 23% 30% To all destinations 13% 17% 19% Automobile Occupancies: To downtown peninsula 1.29 1.35 1.37 To regional town centres 1.24 1.30 1.32 To all destinations 1.28 1.35 1.33 Transit Service Levels: Length of rapid transit routes 23 km 83 km 99 km Population < 400m from bus 87% 90% 90% Population < 1km from rapid transit 8% 25% 30% Other Indicators: Total vehicle distance* 11.1 Gkm 13.4 Gkm 16.6 Gkm Average car speed 38kph 40 kph 37 kph Average truck speed 53 kph 53 kph 49 kph Source: Compiledfrom A Medium-range Transportation Plan for Greater Vancouver, (Transport 2021 Report, October 1993): pp. 53-54, Numerical Performance Targets for 2006 -Transportation System Targets. *Gkm is Gigakilometres, or billions of kilometers. 151 The targets suffer from many of the same limitations as the 1992 travel survey in that the focus is on morning peak hour trips, which now account for less than an eighth of all trips on weekdays. A 38% increase in automobile-based trips over 15 years in the peak hour is to be aimed for, which represents an annually compound growth rate of 1.83%. The observed annual growth rate of approximately 3.75% over a 24 hour period is a factor of two higher than the target growth rate, indicating that a major reversal of current trends will be required to achieve even these relatively large absoluted increases in automobile-based trips. It is worth noting that while the target for transit-based trips is an increase of nearly 30%, the target increase in walk/bike trips is only 20%, implying that a large share of transit's increased ridership is expected to come from people who would otherwise be walking or cycling. A similar oversight almost occurred in the City of Vancouver's Draft Transportation Plan in 1996, where a net decrease in the total walk/bike modal share was initially proposed to achieve a higher transit share. These targets concede that complete, compact communities are not expected to yield many increases in people walking or cycling in the near future. It is also worth noting that the target for average driving speeds calls for a slight improvement of speeds. Given that time savings are now the predominant factor which makes most people choose the automobile over transit, it is hard to see how drivers would be lured away from their cars into buses to meet transit trip targets. A number of recent regional planning documents have identified the growth in "suburb to suburb" trips to be the most pressing transportation problem within the region.226 Recall that the GVRD defines a suburb as any GVRD municipality outside of Vancouver. Workplaces in employment rich municipalities such as Burnaby, New Westminster, and Richmond would therefore be counted as an inter-suburban trip if the employees did not have a place of residence in either these municipalities or Vancouver. This point of view discounts the fact that almost half of all worker report both a place of residence and a place of work within the regional core and that seven out of every ten workplaces are located within the regional core, as the following table shows: 2 2 6 See GVRD Strategic Planning Department Commuting Across Municipal Boundaries in the Vancouver CMA (1993), Changes in Journey to Work Patterns in Metropolitan Vancouver 1971 to 1991 (1994) and 1994 Greater Vancouver Trip Diary Reports, Report 4: Analysis and Historical Comparison of Travel Characteristics (1995): p. 39, Subregional Trip Distribution Pattern. 152 Table 9 - Inter-regional Workplace Share Workplace > Residence V Regional Core Suburbs Exurbs Total Regional Core 47.34% 4.37% 0.34% 52.05% Suburbs 20.02% 18.36% 1.29% 39.66% Exurbs 2.24% 2.34% 3.72% 8.29% Total 69.59% 25.07% 5.34% 100.00% Source: Statistics Canada, 1991 Census Place of Work data, aggregations by author. What seems to have attracted the most attention from GVRD planners is the fact that "inter-suburban" trips have been growing at twice the rate of Suburb-Vancouver and Vancouver-Suburb trips, which is a valid concern. Two factors in the gradual reduction of the City of Vancouver's share of regional employment are road congestion and higher land prices. These factors have led many new and existing businesses to locate in North Richmond and Central Burnaby, in part to locate closer to their employees. Improved transportation access for employees was an important consideration in BC Tel's decision to move its head office from Vancouver's CBD to Burnaby.227 Closer analysis of the subarea data contained in the following table shows that, of the 18% of workplace/residence combinations shown above as "suburb to suburb," most are to and from locations within the same subarea, i.e. Langleys to Langleys or South Region to South Region. This was defined earlier as "local" travel, which is a highly desirable situation. Cross-commuting between suburbs in different subareas, such as Pitt Meadows and Langley or Delta and Coquitlam, is seen to be negligible. There were also very few workplaces in other suburbs or exurbs reported by residents of the regional core. 2 2 7 David Ley, "Downtown or the Suburbs? A Comparative Study of Two Vancouver Head Offices," The Canadian Geographer, No. 29, Spring 1985: pp. 30-43. 153 O o o o CD o CD O CO in CN in o CN I m O CN O in CO o o o o in 3 X CO >» O) c CO "a. c CU s l a cu > S s s C/5 3 H £ co °> S 5 CO o CN m o or CN CN CO o SP CM o m r--o to xi in x i x i 3 CO r-- o co co CO CO CO 351 co CN cu i ro m co in o C3> CO cu £ sz o •e cu O CO co in # co oo CN o o b o m CO CO CN CO o CO o o CO C 4) •2 o go a: co CO CN BP! 3' CN cP CM O o CM CO o in ia>| co o o "co c o D) CU 1 I o> Lc 1 -*-» CU •g co o CO XI XI CO c cu cu cu XI (0 a. 8 o E .c o en co CN o oo in CN co o oo o CO c o £ o CD at S c cu o cu a. •a (A co in a. fl| m IT co CN CN in 00 m CO m o Dl CU l_ CU £ o I o c JZ a> > o o c CO > in o co CN CO iri CD CO "8 CO CD O CO o A > 8 8 co c a- £ 2 IS cu I £1 5 cu XI CO c to |o| (0 c o D)| o CU CO to c o , "> cu or co cu CO *s to cu CO 10 CL a 1 I I s 3 ON i cS n >-* a Stressing the importance of a growing "inter-suburban" travel trend problem focusses attention on suburban travel and masks the fact that the region already has a reasonably compact, transit-supportive core that could serve as an excellent base for nature intensification. Of concern to the goal of sustainable land use planning is that an undue emphasis on this one problem may have been used to justify major transportation infrastructure investments that have clearly worsened the situation. Infrastructure examples include the Richmond Connector and Alex Fraser Bridge which connect Richmond and Vancouver to North Delta and Surrey or the numerous projects that have improved access between the Fraser North and the Northeast Sector subareas. If Burnaby, New Westminster, North Richmond, North Vancouver City and Vancouver had been defined as a Growth Concentration Area twenty five years ago, when the A L R was defined and Oregon created a UGB around Portland, or even ten years ago before these transportation investments had been made, today's sprawl could have been reduced dramatically. Such a concentration would have promoted accelerated Transit Oriented Development within the region and reduced the priority given to improving access to outer suburbs. Portland's UGB is accompanied by minimum density regulations, to protect housing affordability, a concept which is lacking in the GVRD's Growth Concentration Area definition. Enormous resources have been invested into providing transit services from inner and outer suburbs to employment centres in the regional core. While transit's modal share for work trips to the CBD dropped during the late 1980s and early 1990s for users living in the Burrard Peninsula and the inner suburbs, large increases were being experienced for users living in the outer suburbs.228 Overall transit ridership levels are still extremely low in the outer suburbs, indicating that such investments have been little more than a subsidy to long-distance commuters at the expense of the rest of the transit system.229 2 2 8 GVRD Strategic Planning Department 1992 GVRD Travel Survey, Report 6: Comparison of 1985 and 1992 Travel Characteristics (1994): p. 22. 2 2 9 According to the 1996 City of Vancouver Transportation Plan "Choices" document, Vancouver composes 6% of the VRTC service area, but provides 53% of the system's ridership with only 30% of the system's buses. The remaining 70% are allocated to low-density suburban routes. 155 3.3.2 Transportation Infrastructure Investments GISs overlays indicate that many recent infrastructure projects have coincided with the appearance of unmistakeable symptoms of urban sprawl: automobile-dependent suburban travel, low density subdivisions, and long distance commuting to workplaces. The following overlays indicate that much of the major new infrastructure constructed in the last 20 years has served to provide easier access to municipalities with severe surpluses of housing over jobs and long commuting distances. Much of the emphasis before the 1990s was directed at the North Shore and South Region. More recently, the Northeast Sector and North Fraser municipalities have been targeted for large increases in mostly residential development and traffic growth. Major Infrastructure Projects X Water Crossing SkyTrain/SeaBus — H i g h w a y / H O V <—> Counterflow Lane West Vancouver N Vancouver Dist Commuter Rail N Vancouver City U.E.L. , E M P to E L F Ratio (1991) • 1.058 to 1.372 (3) • • 965 to 1.058 (2) • 0 822 to 0.965 (2) _i 0.818 to 0.822 (2) • 0.676 to 0.818 (4) • '"10.615 to 0 676 (4) • 0.611 to 0.615 (1) • 0.46 to 0.611 (4) White Rock Source: GVRD 1992 Travel Survey, MoTH, overlays by author. Figure 15 - Transportation Infrastructure and Employment Concentrations 156 Major Infrastructure Projects X Water Crossing SkyTrain/SeaBus <-^ > Counter-flow Lane Highway/HOV Commuter Rail West Vancouver \ J Vancouver * L Commute Distance (km, 1991) • 18.9 to 21 (2) • 18.5 to 18.9 (2) • 15.9 to 18.5 (3) B 15.7 to 15.9 (3) H 14.7 to 15.7 (3) • 11.4to 14.7 (3) • 10.8 to 11.4 (3) • 8.3 to 10.8 (3) White Rock Source: GVRD 1992 Travel Survey, MoTH, overlays by author. Figure 16 - Transportation Infrastructure and Commuting Distances In the following table, the municipalities in the left column are characterized by long commuting distances, severe imbalances of employed residents to employment, and high growth rates. The transportation infrastructure in the right column was added in the last twenty years or is currently being proposed and has the potential to contribute to low-density sprawl. Table 11 - Shaping Suburban Growth with Transportation Infrastructure (1976-1996) Municipality Infrastructure Abbotsford, Chilliwack Widening Trans-Canada for H O V Lanes Proposed Port Mann Bridge Widening Coquitlam, Port Coquitlam, Port Moody Barnett-Hastings People Mover Mary Hill Bypass Widening West Coast Express Proposed Broadway/Lougheed LRT North Surrey, Langley Highway 91 (Richmond Freeway) Alex Fraser Bridge SkyTrain Extension 157 Maple Ridge, Pitt Meadows, Mission Pitt River Bridge Widening Pitt River Counterflow Lanes Mary Hill Bypass Widening West Coast Express North Delta, South Surrey, White Rock Highway 91 Alex Fraser/Queensborough Bridge Massey Tunnel Counterflow Lanes Highway 99 Widening for Bus Lanes Added to these provincially funded projects are many complementary local projects. These include the construction of new arterial and local streets, the widening of existing streets to add more lanes or turn bays and the elimination of on-street parking. The importance of their cumulative impacts should not be underestimated, in terms of financial cost, quality of life and the facilitation of automobile travel. An example is the Knight Street truck route in Vancouver. Over $12 million has been spent since 1990 solely for the purpose of acquiring the ROW needed to construct new left turn bays which are facilitating annual traffic increases of up to 5%. This sum, exclusive of actual construction costs, would have been sufficient to complete the city's bike route network and provide shelters at all transit stops in the city. Schools, businesses, and numerous homes have had their frontages and playgrounds sacrificed. In one location, an entire apartment complex was demolished to widen the street. As the picture below shows, the vast majority of vehicles using the truck route are automobiles. Truncated homes along Knight Steet in Vancouver. 158 These investments provide examples of a "cause and effect" relationship between accessibility and travel behaviour and can be attributed to allowing policy to be developed primarily from an engineering point of view. Using methods developed during the massive expansions of the freeway system during the 1950s and 1960s, most of today's senior transportation engineers were trained to predict future traffic demands and to recommend supply solutions that will most efficiently accomodate this demand.230 It has only recently been shown that this approach, far from alleviating projected congestion, encourages a latent demand for travel.231 4. New Recommended Capacity 2. Original Capacity Increased Vehicles Per Hour ->-1986 1996 2006 Year 3. New Predicted Demand 1. Originally Predicted Demand Figure 17 - How Transportation Planning Generates Traffic Appropriate land use decision appear to lag behind transportation decisions in this environment. With cheap land made accessible by such heavily subsidized transportation infrastructure, land use economics dictates that developers can offer buyers affordable housing in low-density ground-oriented housing and realize a high return on investment. Another way of stating this is that, as will be shown later, households starting out in the Lower Mainland real estate market are being 2 3 0 G.M. Lamb, "Introduction to Transportation Planning: Context of Transportation Planning," Traffic Engineering and Control, January 1970: pp. 422-425. 2 3 1 The subtitle of Richard Arnott and Kenneth Small, "The Economics of Traffic Congestion," American Scientist, Vol. 82, September-October 1994: pp. 446 - 455 is "Rush-hour driving strategies that maximize an individual driver's convenience may contribute to overall congestion." 159 presented with the following choices for the same price: a small, two bedroom condo with balcony in Vancouver or a large, detached house with full basement in Surrey. As a result of this unbalanced choice, a strong residential foothold has become established in primarily rural areas that were once inaccessible due to natural barriers. The Northeast Sector and the Fraser South subareas are the clearest examples, but the Central and Eastern Fraser Valley do not appear to be far behind. Decisive action now to focus new development within existing communities appears to be the only way to prevent condemning the remainder of the Fraser Valley to a primary role as a bedroom for Burrard Peninsula workers. "If we continue to plan land development and transportation independently, trying to build our way out of congestion - without taking into account the indirect relocation effects that a new highway induces - this policy will produce far more congestion for far more people over much larger areas, and it won't have fixed anything."232 ...Stephen Putnam, Professor of City and Regional Planning and Director, Urban Simulation Laboratory, University of Pennsylvania Quoted in Chris Lazarus "Integrated Modelling" Earthword, Vol. 4, 1992, Transportation Planning. 160 3.4 Housing Type and Location Trends A plot of yearly housing starts within the Vancouver C M A shows that there have been strong, cyclical fluctuations over the last twenty years. Vancouver C M A Housing Starts 1975 1980 1985 1990 1995 Year Source: GVRD Development Services, statistical analysis by author. Figure 18 - Housing Starts in the CMA (1975-1995) The dotted lines on the plot represent linear and exponential best fit curves for the data, or what would have happened if the housing starts had increased monotonically. The Vancouver C M A experienced a 2% annually compounded population growth rate during this period, i.e., an exponential growth curve might be expected. Nevertheless, linear growth fits the housing data better than exponential growth, according to the R 2 value, a common statistical measure which determines how well data is correlated to a given curve fit. R 2 values near one indicate perfect correlation and values closer to zero indicate lower correlation. The R 2 value for the linear fit is 0.269 and the R 2 value for the exponential fit is 0.237, indicating how erratic C M A housing starts have been. 161 3.4.1 Demographic and Socioeconomic Influences Fluctuations in the number of housing starts are less surprising when the yearly fluctuations in population growth is considered, as opposed to the total population growth. GVRD Population 1,900,000 1,800,000 -1,700,000 -TION 1,600,000 -J5 "3 1,500,000 • a. e CU 1,400,000 -1,300,000 -1,200,000 < 1,100,000 -—1—1—1—1—1—1—1—1—1—1—1—1—1—1—1—1—1—1—1— GVRD Yearly Population Change 50000 -r 45000 -40000 -35000 -30000 -1 25000 -20000 -15000 -10000 -5000 -0 - H 1 1 1 1- H 1 1 1 1-Source: BC Statistics. Figure 19 - GVRD Population Trends (1976-1996) Besides population growth, a number of other socioeconomic factors should also be considered when analyzing housing starts. The trends and potential influences of some of these factors are: Table 12 - Factors Influencing Housing Trends Socioeconomic Factor Current Trend Potential Influence Demographics Population increasing, aging Potential for exclusionary SFD zoning, equity wealth ^ high income. Household size Low, decreasing, more singles living alone Number of new units, mix of units. Unemployment rate High, stable, very high for younger age groups Economic uncertainty, decision to buy instead of renting. 162 Household income Average income stagnant, gap increasing between lower and upper incomes Ability to afford a home, need to form larger "economic" families to own. House price High, increasing, highly location-dependent Market value of housing may exceed ability to pay. Discussed in next section. Mortgage rate Low, stable Payments determine ability to pay. Source: BC Statistics, CMHC, obser\>ations by author. BC Household Demographic Trends 3.50 1971 - Population (Millions) - Households (Millions) - Household Si2E 1976 1981 Census Year 1986 1991 Age of Household Maintainors in BC 25.00% • 20.00% 0.00% 1971 1976 1981 1986 1991 Census Year -15..24 -25.J4 - 35..44 -45..54 - 55..64 -65+ Source: Statistics Canada. Figure 20 - Household Demographic Trends (1971-1991) BC Per Capita Income (1986 Dollars) $12,500 • H 1 1 1 1-Unemployment in BC Source: Statistics Canada Figure 21 - Income and Unemployment Trends (1976-1995) 163 Perhaps the most important factors that do not emerge from the data is that, while unemployment, inflation and household incomes are stable, unemployment and incomes are not distributed equally among age groups. In particular, as was noted earlier, unemployment is much higher than average for the under-35 age group and the real personal and household incomes of this group are well below average incomes and are decreasing.233 Vancouver demographer David Baxter has based his GVRD housing demand projections for the 1996-2021 time period primarily on historical trends and what younger families have traditionally preferred with less emphasis placed on these crucial factors.234 This led to a conclusion that ground-oriented units would dominate apartment units by almost a 2:1 factor in any likely growth scenario.235 As will be seen below, apartments have composed half of all starts since 1990, which better reflects the likely future costs of ground-oriented housing and the current ability of younger households to pay. Five Year Mortgage Rate Change in Consumer Price Index (1986 Dollars) Sources: BC Statistics, Statistics Canada Figure 22 - Mortgage Rate and Inflation Trends (1976-1995) 2 3 3 Statistics Canada Canadian Economic Observer: Historical Statistical Supplement 1994/95 (Ottawa: Queen's Printer, 1995): Tables 8, 9. Real incomes for full-time workers the under-45 age groups has dropped between 10% and 20% between 1990 and 1995 while they rose 5% to 10% for workers in the 45+ age groups. The gap between full-time and part-time workers has quickly widened, which is resulting in many full-time employees working longer hours while unemployment and underemployment is increasing. 2 3 4 David Baxter Homes in Metropolitan Vancouver's Future: Housing Demand by Structure Type, 1996 to 2021 (Urban Futures Institute, August 1996). 2 3 5 Ibid. p. 35. 164 3.4.2 Housing Starts and Types Given the underlying long-term statistical trends which result in market fluctuations, comparisons between municipalities and different types of housing can be problematic and subtle changes in housing preferences can be difficult to detect. To help overcome this difficulty, the data for the rest of this section has been normalized, or "smoothed," to offset short-term fluctuations in the total number of C M A housing starts and better reflect the best fit curve shown above. Determination of the Normalization Factor Normalized Housing Starts (NHS) for a given year, type and location is defined as NHS(Year, Type, Location) = HSobsLrv.<i(Year, Type, LocationVNF(Year). whete housing starts observed for a given yeai, type and location, IIS(vcn<.u is from CMHC actual housing start data,-''0 and the normalization factor for the year (NF) is NF(Year) -- I -t ((HS< (Year) - HSc\ivKPtaui(Year))/ HS, \n i^pu.,,j(Year), and the expected C M A housing starts are HS«-Mvi M*uui(Year) - 294 088*Year - 568285, where the numerical values for slope and intercepted are taken from a least-squares linear curve fitting algorithm performed on the housing data Housing start data for each municipality and the CMA by year compiled by Development Services Department of the GVRD, other groupings by housing type, municipality, subarea done by the author. 165 For example, if the observed number of housing starts in the C M A was 25% above the expected number for thai year, then the normalized number of housing starts used for comparison purposes would be divided by 1.25: NHS(Year, Type, Location) (Year, Type, Location)/! 25 The resulting numbers of normalized C M A housing starts are: Before Normalization: After Normalization: Actual CMA Housing Starts 25,000 20,000 + Year -Det -Dup - Row -Apt -Total Normalized CMA Housing Starts Year Source: GVRD Development Services. Figure 23 - Normalized Total CMA Housing Starts 166 The relative distribution of housing types within these housing starts is: CMA Housing Starts Composition in r— Cr^  »—i o m f- C\ <—1 o m Year Source: GVRD Development Services. Figure 24 - Composition of C M A Housing Starts The type composition of these housing starts contains important information. Apartments can be seen to be absorbing most of the overall increase in housing units in the last five years with a mirror reflection of decreases in the share of detached houses. The relative proportion of duplexes and townhouses has remained remarkably flat over the twenty year period, with the exception of a period after the severe recession in the early 1980s when townhouses enjoyed temporary popularity at the expense of detached houses. This observation points out that no discussion of housing starts can be made in the absence of the prevailing state of the economy. The locational distribution of housing starts for different types throughout the region is also important to identify, as is shown on the following two pages. 167 > > g o o g | | . « i i . a Hi * * • + • • < x o x o x o x o v o x o x \ ° o x 0 S - ox O O O O O O O O O O O O O O O O o O O v i © «n o* v i © v i O m m fN '— 1 t>2 CO S 8 .S" *s "S s <*> M) _a "so S O w < ss .C IT/ u 3 en 00 ] B | 0 X J 0 O / o 1 S -a ii •a s ii >, ii u u s t/5 ii >• s o W B « Ml « tZ5 CJD S S o w < U o a u « vo la 9 W> fc VO SL61 xO o x O O 00 \ ° o N o x x=> o x \ ° 0s-x=> o x ox xO o x xO 0 s xO o x O O O O O O O O O O O O O O O O O O VO <N © 00 VO c4 © 1 B ; O X J O O / O These plots show that, while Vancouver and Surrey have maintained the lion's share of new housing starts over the last two decades, Vancouver's share is the greatest and is experiencing strong growth while Surrey's share is dropping rapidly. Outside of Surrey and Vancouver, the share of housing starts in Burnaby and Richmond have gradually declined while the share in the Northeast Sector and the Langleys have gradually increased. The other variable of interest is the percentage composition of the three "barometer" housing types in each municipality. The housing start plots on the following pages show apartment, townhouse and detached house starts for municipalities in the Vancouver C M A with more than 10,000 residents and various groupings of municipalities. The influence of the state of the regional economic factors introduced earlier should again be kept in mind when interpreting the data. In the years following the recession of the early 1980s, in addition to a decrease in the absolute number of housing starts, there is a period of increased apartment and townhouse starts and a corresponding decrease in the number of detached house starts. Apartment starts in Vancouver continue to overwhelm those in other municipalities, indicating that there is a strong ongoing demand for housing in spite of the high costs involved for relatively small units. There is a growing trend towards a larger share of apartment starts in the inner suburbs, complementing rapid price increases for ground-oriented housing in these municipalities. Although difficult to quantify, there is a distinct "trickle-down" effect in evidence. As housing prices increase near employment centres, there is a shift not only in the location of new housing but also the type of new housing. Households that once chose detached houses are now deciding between townhouses and apartments. Similarly, as will be seen in the final research findings section, households that once worked in Vancouver, and lived in Richmond or Burnaby, are now moving to the Northeast Sector and South Fraser subareas. Residents of these outer suburbs are increasingly moving to exurbs as far away as Chilliwack. The number of apartment and detached house starts for Vancouver is somewhat skewed for a number of reasons. For apartments, the construction of new rental apartments, dramatically reduced across the Burrard Peninsula, has practically ended in the City of Vancouver. 170 Condominium apartments are increasingly seen as investment vehicles and are consequently playing a increasing role in the city's new rental stock.237 For detached houses in Vancouver, the vast majority of new starts represent the demolition and reconstruction of existing detached homes, and therefore do not add to the region's housing stock.238 Although stable on the regional scale, the trends in townhouse starts at the local lev