Externalities of Megaprojects
An Investigation of the Impacts of Mass-Transit Expansion
on Neighborhoods in Amsterdam and Washington D.C.
Taylor Beswick
Master Thesis Urban and Regional Planning
August 2014
Student Taylor Beswick (10635661)
Supervisor
dhr. dr. J.A. (Julio) Soria Lara
Second Reader
dhr. dr. J. (Jochem) de Vries
This master thesis has been written as part of the Master Urban and Regional Planning Program in the Graduate School of Social Sciences at the University of Amsterdam. August 2014
[Abstract: The ever growing need for cities to become as efficient as possible in order
to compete in intercity, regional and global economies has required the urban planner and decision maker to utilize all economic growth tools at their disposal. Expansion of public transit infrastructure by means of megaprojects is one way to attract business growth with an efficient transportation network. While many of these megaprojects are touted for their successes, little thought goes into the effects on populations of the city – especially those most vulnerable to radical market shifts. My study looks into the net effects that transportation megaprojects have had in two national-flagship cities: Washington D.C. and Amsterdam. Using quantitative research strategies with a mixed quasi-experimental design, I analyze socio-economic data to bring further light on the implications of future projects to the most vulnerable populations of cities.]
TABLE OF CONTENTS 1. Introduction ... 2 1.1. The Problem ... 2 1.2. Scientific Relevance ... 3 1.3. Societal Relevance ... 4 2. Theoretical Background ... 4 2.1. Megaprojects ... 4 2.1.1. Aims ... 7 2.1.2. Impacts ... 8 2.1.3. Shortfalls ... 10 2.2. Accessibility ... 12
2.3. Gentrification and excluded Populations ... 13
2.4. Controversies ... 15
3. The Cases – Washington D.C. & Amsterdam ... 15
3.1. Background ... 16 3.1.1. Washington D.C. ... 16 3.1.2 Amsterdam ... 17 3.2. Comparison ... 18 4. Research Question ... 19 4.1. Main Question ... 19 4.2. Sub-Questions ... 19 4.2.1 Sub-Question 1 ... 19 4.2.2. Sub-Question 2 ... 19 4.2.3. Sub-Question 3 ... 19 4.2.4. Sub-Question 4 ... 20 4.3. Hypotheses ... 20 5. Methods ... 20 5.1. Empirical Approach ... 20 5.1.1. Research Design ... 21 5.1.2. Units of Analysis ... 21 5.2. Concepts... 22 5.3. Variables ... 27 5.4. Method of Analysis ... 33 5.4.1. Missing Values ... 39
5.5. reliability and Validity ... 40
6. Data ... 40
6.1. Collection ... 40
7. Results ... 42
7.1. Analysis Question 1 – Accessibility Difference ... 42
7.1.1. Washington D.C. ... 42
7.1.2 Amsterdam ... 45
7.2 Analysis Question 2 – Change in Income ... 48
7.2.1 Washington D.C. ... 48
7.2.2 Amsterdam ... 53
7.3 Analysis Question 3 – Correlation ... 57
7.3.1 Washington D.C. ... 57
7.3.2 Amsterdam ... 59
7.4 Analysis Question 4 – Controlling for Movements ... 60
7.4.1 Washington D.C. ... 60 7.4.2 Amsterdam ... 67 7.5. Summary of Results ... 73 8. Conclusion ... 73 8.1. Discussion ... 73 8.2. Final Statements ... 75 References ... 76
1. INTRODUCTION
1.1. THE PROBLEM
In order for large urban areas of today to sustain their positions as competitive economic centers, city planners and city officials use all tools at their disposal that can create the appropriate conditions that attract business and economic development. One condition that is required for an economically vibrant city is ample transportation
accessibility for all workers living in the city. A common characteristic of a healthy urban economy is an efficient transportation network. City populations in North America and Western Europe have been burgeoning since the 1990’s, and the most efficient way to transport high-density populations is via public mass transportation (Rodrigue 2013). Public transportation megaprojects are often aimed to spur economic development in cities and metropolitan regions. Industrial growth, commercial expansion and increased middle-class expansion are often the touted results for the expensive infrastructure improvements. These fields of the economy are thought to have distributive effects that benefit the lives of socially excluded communities within the urban area. Yet this is just a perceived
understanding of the true interrelationships between poverty and urban transportation and it is actually poorly understood (Hook 1998). Even though equity and fairness is commonly advertised in transportation projects as a main concern, it has in fact received little research (Litman 2002), as have the other distributional effects of urban
transportation megaprojects. The realizations of these truths have led to new criticisms of major projects real impact on poverty that has been acknowledged by the World Bank (Gannon, Liu 1997). Publicly-funded transportation megaprojects bring many founded benefits to neighborhoods across applied urban areas including decrease in car use,
decreased commute time, increased labor market accessibility and economic incentives led by increased real estate demand (Debrezion, Pels et al. 2011, Welch 2013, Dubé, Rosiers et al. 2011, Korytárová, Hromádka 2014). Yet there are reasons why significant publicly-funded megaprojects are uncommon occurrences both in North America and Europe (Spiekermann, Wegener 2006). Both elected and government budget commissions shy away from the unattractive features of publicly funded transportation megaprojects. It is difficult to agree to these plans due to the amount of space required in urban centers, the political fallout from unpopular construction delays (Bruzelius, Flyvbjerg et al. 2002,
Flyvbjerg, Bruzelius et al. 2003), the unknown future socio-economic consequences (Golub, Marcantonio et al. 2013), very long project completion periods, and the risks involved with securing funding for very high price tags. These project budgets often do not account for the likely scenario of cost overruns (Taylor, Garrett 1999, Flyvbjerg 2013). Mixing the little understood externalities of the benefits from recently completed transportation
megaprojects with known issues, such as high financial risks and costs, simultaneously creates a real problem for vulnerable stakeholders living in the city. Thus, I identify the main problem as being the unknown effects from the many variables changed by
megaprojects, and how those changed variables interact specifically with the vulnerable socio-economic neighborhoods of a city.
1.2. SCIENTIFIC RELEVANCE
This thesis is concerned with examining transportation megaprojects distributional effects on vulnerable socio-economic neighborhood populations. Examining two former metro-subway megaprojects occurring in 1990s in the cities of Washington D.C. and Amsterdam offer the chance to use geographic and statistical data to observe benefits or effects on the low-income neighborhoods and minority neighborhoods in each city.
The cases used in the study hold scientific relevance because they allow
examinations of very large megaprojects in two cities in North America and Europe that are similar in purpose, time period, size and city-specific project location. Acknowledging differences in real estate market structure, country economies and governments between the Netherlands and the United States (Priemus, Dieleman 2002, Schwartz 2013); findings from this study can be used to target future scientific research. Inquires can be used for separate studies to find how each case’s different mechanisms and environments interact and effect the resulting effects, either good or bad, on the vulnerable socio-economic groups of cities and regions undertaking transportation megaprojects.
Table 1.1 – Similarities of the cases of Washington D.C. and Amsterdam
Washington D.C. Amsterdam
Population (2014) 646,449 811,185
Size 177 km2 219 km2
City Role National capital (United States) National capital (Netherlands)
Origination of metro network 1976 1977
Megaproject studied (finish date) Multiple subway line expansion (1999) ‘Ringlijn’ metro expansion (1997)
Future megaprojects (finish date) Silver Line (projected 2015) ‘Nord-Zuid Lijn’ (projected 2017)
Source: urbanrail.net 2014; District of Columbia 2014; Gemeente Amsterdam 2014 The current and future transportation plans for both Washington D.C. and
Amsterdam show scientific opportunity for more research into this thesis topic. Concurrently in Washington D.C. and Amsterdam, there are new metro-expansion megaprojects occurring. The projected completion date for each is between 2015 and 2017. Improved understanding of the effects of past megaprojects in Washington D.C. and Amsterdam on vulnerable populations in each city can be used to lead studies that are more specific on interactions and processes with the current megaprojects. Additionally the increased knowledge of the externalities of past megaprojects in Washington D.C. and Amsterdam could prove to be very valuable information for policy makers and planners in which they can apply to improve the planning policy of the megaprojects in each city going forward.
1.3. SOCIETAL RELEVANCE
The implications of transportation megaprojects on the poor populations of
metropolitan areas are enormous. Costs for megaprojects are very high and if the projects are not significantly improving the lives of all in the city, they are likely not worth their cost. Washington D.C. is currently expanding their metro line at a current estimated cost of $5.6 billion with an estimated completion date in 2015. This is one of the largest
transportation infrastructure projects currently under construction in the United States (McVeigh 2013). Another case of a city currently investing in expansion of their metro line is in Amsterdam. Currently there is construction occurring on the North-South Line that will expand the metro line for this city. The estimated cost was initially €1.2 billion but has ballooned to the current estimate of €3.2 billion (roughly $4.25 billion), with a completion date of 2017(Essbai 2013).
The aims of my study are to better understand the relationship of transportation megaprojects and their resulting effects on socio-economic vulnerable neighborhood populations. To do this, I would like to analyze previous megaprojects and their effects in a European and American context. The specific case studies I will be looking at are the cities previously mentioned: Amsterdam, Netherlands and Washington D.C., United States. I chose these particular cases not only because they both have megaprojects currently under construction, but also because they both completed major metro line infrastructure
improvements during the 1990s. In the case of Amsterdam, the Metro line 50, otherwise known as the ‘Ringlijn’, was completed in 1997, so I will analyze the city from 1996 through 2006. In the case of Washington D.C., the Metro Green Line was completed in 1991, so I will analyze the city from 1990 through 2000. Looking at the spatial accessibility level
improvements to neighborhoods before and after the metro expansion in each city and relating it to change in social equity, I hope to find out how past megaprojects have affected the poorer populations. This knowledge could give planners, policy makers and city
officials helpful insight that could be implemented on current megaprojects in Washington D.C. and Amsterdam.
2. THEORETICAL BACKGROUND
2.1. MEGAPROJECTS
Megaprojects are a different breed of construction projects because of their size, complexity, time, risk, environmental impact, effects on communities, and controversy (Capka 2004). All of these issues create a unique atmosphere where many stakeholders are combined in order to complete the mega task-at-hand. Capka (2004) writes that of all the unique characteristics of transportation megaprojects, size and complexity stand out as the biggest hurdles when looked at through a historical perspective. These two
characteristics are negatively synergistic if a transportation megaproject does not have proper, sustained management for the duration of the whole project. The US Federal Highway Administration defines transportation megaprojects as, “major infrastructure projects that cost more than US$1 billion, or projects of a significant cost that attract a high level of public attention or political interest because of substantial direct and indirect impacts on the community, environment, and budgets.” Bent Flyvbjerg’s research has
looked into the “megaprojects paradox” of why they seem to have such daunting odds when improperly managed (Flyvbjerg 2013).
In an article entitled “MEGAPROJECTS: Grand Schemes Need Oversight, Ample Funding” (Sykes 1998), international advisor on megaprojects Allen Sykes describes and defines megaprojects and their differences from other large, yet less complex, projects. He outlines nine characteristics that distinguish megaprojects:
Size and the likelihood of multiple owners
Public opposition to the likely social, economic, political, and environmental impacts
Time – a decade or more to plan, design, finance, and build Located in remote and/or inhospitable places
Potential to destabilize markets because of the demand on labor and supples Unique risk, especially when project spans economic cycles
Financing difficulties
Insufficient experience, especially in managing complex undertakings Career risks, because most of the undertakings do not advance past the
planning stage and, therefore, pose an unpopular career course for senior managers.
These characteristics for megaprojects coincide with many researchers definitions and outline the problems that can be found when planning or implementing such a project.
A research paper authored by G. Oliomogbe and N. Smith from the University of Leeds, takes a survey of literature concerning transportation megaprojects. They show the trends in megaproject literature to be a movement from ideas of ‘value management’ to ideas of ‘understanding how stakeholders values different things’ (O Oliomogbe, J Smith 2013). In their paper, a collection of some of the characteristics of megaprojects is
presented, as defined by the literature of various planning researchers. Table 1.2 shows the collection of characteristics by author.
Table 1.2 - Characteristics of Megaprojects: Author Overview
Megaproject Characteristics Authors
High investment expenditures of US$1 billion or more Long lifetime of 50 years and more
Considerable uncertainty with respect to the demand forecasts and cost estimates
Considerable share of indirect benefits which cannot be captured by the operator (benefits not occurring to the users of the project rather than to third parties)
(Bruzelius, Flyvbjerg et al. 2002)
The project delivers a substantial piece of physical
infrastructure/capital asset with a life expectancy measured in decades
The client is often a government or public sector organisation The main contractor or consortium of contractors are usually
privately owned and financed
The contractors often retains ownership stake in the
infrastructure/asset after the construction phase is completed-typically a minority shareholder in a special purpose vehicle (SPV) and is paid by the client for the service that flows from the asset’s operation or use over a number of years.
(Sanderson 2012)
Frick calls the six C’s:
Colossal in size and scope
Captivating because of their size, engineering achievements or aesthetic design
Costly- often undercosted Controversial
Complex
Have control issues
(Sturup 2009)
Multiple organisations seeking success with different objectives Changing priorities by project objectives
The project being subject to the impact of a wider socio-political environment
(Ruuska, Artto et al. 2009)
Value of over £150million
Complex management structure
Politics playing an important role in how senior management appointments and activities are defined.
(Stoddart-Stones 1988)
Source: Oliomogbe and Smith, 2012 Examining this table, Oliomogbe and Smith found that the definitions of megaprojects, based on the year the articles were written, evolved from a more simplistic definition to a broader, more complex definition. This can be seen from the Stoddart-Stones’ definition in their article in 1988 of having simply over £150million and complex management structure to Sanderson’s definition from his paper in 2012 which lists more descriptive
characteristics and includes the stakeholder as a important player in the megaproject process. Oliomogbe and Smith took the definitions of the many articles and developed a modern and comprehensive megaproject characteristic overview:
Time: More than a decade (often more than one political dispensation) Cost: Greater than £100 million
Colossal use of resources (money, human, equipment, etc.) Owner: Government/Public sector
Large Size
Risk and uncertainty
Technological innovation / insufficient experience Social, political, economic and environmental impacts Multiple owners
Complexity
Poor performance (cost, quality, performance, etc.) Control issues/changing priorities
Indirect benefits to non users of the project Located in inhospitable places
Career risk
The greatest differences between the definitions of a transportation megaproject are the quantifiable costs involved. Different studies have found this number threshold to be anywhere from £100 million to €1.5 billion (Stoddart-Stones 1988, O Oliomogbe, J Smith 2013).
2.1.1. AIMS
More now than ever in world history, cities, states, provinces and nations in the transportation community are required to plan for solving very tough challenges. The challenges are exasperated by the fact that populations in urban areas have exploded and more and more people can afford to use personal transportation, such as a car, for their daily travels. The challenges faced by transportation planners are multifaceted and their aims needed for the megaprojects are numerous. They include a demand for safer systems, greater transportation capacity, complex multimodal solutions, environmentally enhancing context-sensitive designs, and an urgent requirement to renew or replace an aging and over-stressed transportation infrastructure (Capka 2004).
Multiple stakeholders have both direct and indirect roles with planning, financing, building and implementing a megaproject. The aims and purpose of a project can be very different from project to project, depending on the level of power of certain stakeholders involved in the megaproject. The balance or imbalance of power divided by the
stakeholders results in different values being seized upon for the project (O Oliomogbe, J Smith 2013). An example given by the authors, is the London 2012 Olympics. When a transportation megaproject is built for the fans, athletes, media and tourists, time is valued highest over other concerns such as environmental impacts, long-term planning purpose, and impacts on the local population. Because the London 2012 Olympic officials are the stakeholders with a high amount of power, normal planning processes for a metro-expansion or other transportation megaproject are not fully used.
The stakeholders that are involved in a megaproject can be categorized as multiple firms, public organizations, authorities, political decision-making bodies and several owners (Ruuska, Artto et al. 2009). These stakeholders are all described to be directly involved with the megaproject and its processes, while Russka and Artto et al. explain all other parties who are indirectly involved with the megaproject as external stakeholders. Oliomogbe and Smith, as well as Flyvbjerg, raise a problem with this viewpoint because it fails to recognize the possibility of unknown impacts to external stakeholders. This causes a situation where the megaproject is planned and constructed with the internal
stakeholders’ aim and values most considered, but when the megaproject is implemented, the former external stakeholders are most adversely affected because their involvement. Therefore, their aims and values were not considered because of the predisposed
irrelevance of their connection to the megaproject. Oliomogbe and Smith identify with Flyvbjerg’s six megaproject stakeholders which are citizens/public, non-governmental organizations (NGOs), various levels of government, industrial interests/trades, scientific and technical expertise, and media (Flyvbjerg, Skamris Holm et al. 2003). These six
stakeholder categories give better recognition to the citizens and public who could become involved with the megaproject after it has started.
2.1.2. IMPACTS
The aims of certain stakeholders in a transportation project process, put value in the end result of increased land value, jobs, decreased congestion, and an improved
environmental-friendly image. Bornstein (2010), looks at positive outcomes from megaprojects in the cities of Montreal, Vancouver and Los Angeles. She highlights the overcoming of the fears from towards megaprojects of the local community of
gentrification, displacement, and change or loss of city character. In these three case studies, Bornstein reviews how innovative practices are used to prioritize the needs of low-income neighborhoods and the overall quality of residential areas. Despite these
prioritizations, the number one objective of the megaprojects in these cases was to redefine the city’s image and build an allure of a modern and high economy urban center. The long-term success of these megaprojects in the three cities concerning prioritization of low-income neighborhoods was inconclusive in the article. This fact seems to be a common denominator of many studies that highlight the beneficial efforts aimed at the vulnerable populations of cities undertaking transportation megaprojects. There are few studies that look at the overall value of a megaproject compared to that of multiple smaller scaled projects (O Oliomogbe, J Smith 2013, Flyvbjerg 2013, Hilton, Stoney 2007).
An example of a study performed to examine the economic impacts of a new subway-metro expansion megaproject was based on the case in Madrid, Spain (Calvo, de Oña et al. 2013). The Calvo, de Oña et al. study examines the impacts of the new expansion on the whole network looks at changes in residential development and population. The authors found that there were noticeable increases in the inner-urban core of Madrid of population and residential development. These findings were expected as they agree with literature which proposes that increased subway-metro line capacity and accessibility leads to an increase in residential demand near the station locations (Bowes, Ihlanfeldt 2001). A finding in the study, unexpected by the authors, was the pronounced increases in suburban areas of Madrid and in satellite towns that had sprung up simultaneously with
the building of the subway-metro expansion. Also found in this study were the benefits of subway-metro expansion to areas not directly served by the transit network. Because of these areas, the authors concluded that there are realized benefits from megaprojects in areas directly affected by expansion and areas indirectly affected. Integration of land use planning and multimodal transportation planning, with the subway-metro expansion, demonstrated the increased yield of a megaproject at this scale.
Another impact study looks at an increase in supply of a rapid-bus transit service in Quebec City, Canada. (Dubé, Rosiers et al. 2011). It examines through a quasi-experimental design, the economic impacts increased supply of bus service along the corridor.
Specifically the study analyzes the differences in housing prices from before the supply increase to after the supply increase. The particular corridor being analyzed is only a small part of the overall transportation network. Findings from the difference-in-difference analysis of the change in housing prices by the authors found there was an increase of 2.9% to 6.9% in housing prices where new bus-rapid transit was introduced. These findings are then translated so they can be applied to the region as a whole, where they find that
introduction of new stations raises the city government $6 million in increased tax revenue and plus-value for local property owners of $35 million over a 12 year period. The
implications of these findings show reasons why cities and regions should plan and implement transportation expansion megaprojects in order to inject growth into the local property value market (Dubé, Rosiers et al. 2011).
The previously stated studies look at the direct and noticeable economic benefits found from several cases of transportation megaprojects. These studies do not take into consideration complex social, environmental and economic costs in there evaluations of the net profit gains from these projects. A paper authored by researchers at the University of Birmingham (Korytárová, Hromádka 2014) looks into other impacts of megaprojects which can affect the overall financial calculation of its generated impact. The paper does not directly analyze a case study but instead looks into the use of a previously developed transportation megaproject impact model. This model called the HDM-4 model, takes in to account the impact of time consumption, changes in operation costs for vehicles (of all modes), social costs, environmental costs, safety costs related to accident incident rates and economic impact on related areas. The authors show how the model has extended uses to model impacts of a transportation megaproject such as barrier effects caused by infrastructure lines (i.e. highways, above ground rail, station footprints), effect on urban sprawl and residential density patterns. The author’s exploratory examination of the increased uses of this model to expand the knowledge of current literature in impacts of transportation megaprojects brings to note many possibilities which could improve current practices in megaproject management and planning. This paper is one of the few studies which looks into tools used to assess the impacts, specifically that of transportation megaprojects in the United States. These impacts though made for the United States
context, can easily be changed for use to examine impacts associated with the planning and implementation of transportation megaprojects located in South American or Europe. These locations have different dynamics in which the model could be tuned for and also have values which have higher or lower influences than that of a megaproject located in the United States.
2.1.3. SHORTFALLS
Flyvbjerg looks into the troubling statistics of cost overruns in transportation
megaprojects that have occurred in the last 25 years (Flyvbjerg, Skamris Holm et al. 2004). In his paper, he describes a war is on from the “Great Independence from Space”, and the key weapon in this war is infrastructure development. Flyvberg notes how hundreds, if not thousands, of billions of dollars – public and private – are currently tied-up in the provision of new infrastructure around the world. Despite all of these megaprojects, there is little knowledge on the costs, benefits and risks involved with past cases. His paper goes to examine 258 transportation megaprojects and the efficiency of their budget finances. He concludes that a majority of the cases have cost overruns. He in fact finds that cost overruns are “the rule – not the exception” in transportation megaprojects and identifies this problem as a global phenomenon. Because most transportation megaprojects are financed with public funding (Skamris, Flyvbjerg 1997, Bruzelius, Flyvbjerg et al. 2002, Daito, Chen et al. 2013) most of the risk and uncertainty which goes with large budgets and cost overruns lies with the public. Flyvbjerg argues that the decision-making process can be improved at the planning stage of transportation megaprojects, which could enable the whole process to have transparency, sound judgment, and isolation from private business bias.
A specific study explores the reasons for cost overruns in transportation-related megaprojects in the Dutch context (Cantarelli, Molin et al. 2012). The study looks a
performance metrics of megaprojects in the Netherlands and finds that cost overruns and cost underruns occur at the same frequency based on their methodology. Though both cost overruns and underruns occur with similar frequency the savings realized from underruns do not compensate for the average inflated costs of transportation projects which have gone over budget. The data results find that there is an average 16.5% cost overrun on megaprojects which have gone over their budget. Further examination by the study reveals that the cost overruns in the Dutch projects studied occur most often in the pre-construction phase of the megaproject rather than the actual pre-construction phase. The authors note that cost overrun estimates in this study could be deflated based on the fact that the data was based off of early project estimations which do not account for the lock-in phenomenon. The lock-in phenomenon is created when decision making for a
transportation megaproject becomes path dependent and the policy makers, planners and politicians are already committed to the megaproject, well before the actual decision phase of the project has begun (Sager 2011). The lock-in phenomenon is labeled as a major problem with planning of megaprojects Flyvbjerg and Cantarelli. In their articles they explain the same principle, effectively saying that the problems lies with the pressure and bias which can overcome a decision maker to commit to a certain megaproject earlier than they would feel comfortable. The earlier a decision maker unofficially commits to a
megaproject, the more uncertain and unreliable the data is about the megaproject, which increases the likelihood of cost overruns in the long-run for the project (Flyvbjerg,
Cantarelli 2014, Flyvbjerg, Skamris Holm et al. 2004). An example given by Flyvbjerg of a megaproject that turned into a ‘megadisaster’, because of early-committal by decision makers and a high level of disinformation, is the undersea rail tunnel linking the United Kingdom and France. This project had cost overruns of 80% of the forecasted costs, or over £2 billion in 1985 prices. Another example is the Central Artery/Tunnel project in Boston,
Massachusetts, United States which suffered a cost overrun of 275%, or over $11 billion at the time of the build in 1982 (Flyvbjerg, Cantarelli 2014).
Another study co-authored by Flyvbjerg (Skamris, Flyvbjerg 1997) looks into the problem of misinformation in the planning stages of a transportation megaproject, specifically that of traffic forecasts. The authors give a reason for their research, because they found little research has been carried out on the before-and-after studies of traffic forecasts and their relation to costs in the context of a transportation megaproject. The few studies which have been done on the subject of traffic forecasts and costs in the planning stages of megaprojects showed that often there tended to be under-estimations of costs and over-estimations of traffic forecasts. Given these previous findings, the authors investigate seven Danish tunnel projects along with some analyzed examples from other countries. The results conclude that the tendency for over-estimation on traffic forecasts and under-estimation on costs are present in a majority of the megaprojects on a global scale. Data results show that 50-100% cost under-estimations (overruns) are common and overruns of over 100% are not uncommon. Coupled with these statistics, traffic forecasts, in which the budgets and validity of the megaprojects are based on, are in error 20-60% when compared to realized development and traffic counts. The authors conclude that it is important that reliable and unbiased information is given to decision makers during the planning stage of a project before they make their decision. It is also important that the decision makers themselves do not become too attached to a megaproject, in which they can no longer objectively make logical and wise decisions for their constituents.
There are problems and risks associated with not committing early to a project. A case study of the a light rail megaproject in Ottawa, Canada investigated how decision makers, stakeholders and citizens were at fault for not getting the light rail transit line finished (Hilton, Stoney 2007). Instead of the problems outlined by the previous authors such as Flyvbjerg and Cantarelli, the issue in Ottawa was that a general consensus could not be reach by the City Council, resulting in seven years of planning deliberations and the eventual cancellation of the megaproject as whole. The financial costs which were amounted in that time equaled over $73 million in survey and planning costs, of which many were reiterative and redundant, as well as “not less than the sum of $175.3 million” in lawsuits pending (Hilton, Stoney 2007). An example of the negative risks and
uncertainties associated with megaprojects are shown with the Ottawa light rail transit megaproject case. Megaprojects can cost the taxpaying citizens high amounts of money, even if the plans of the megaproject are not approved. The author’s examination on why the megaproject failed revealed it was the result of lack of trust and transparency between the City Council members and their dependency on unreliable, biased information sources for the megaproject details. Had all of the planning committee members and City Council members been given full, transparent cost estimates, traffic forecasts, and private contract negotiation details – the likelihood of the Ottawa light rail transit failing would have
drastically been reduced. Not only did the project fail, but the taxpayers carried the burden from the incurred costs of the political stagnation.
2.2. ACCESSIBILITY
Examples of poor transportation access correlating to decreasing socio-economic indicators has been shown to occur in both the United States (Lucas 2012, Mare, Bruch 2003) and also in Europe (Musterd 2005). Further investigating this topic within the context of Europe and North America, I have found many studies which have detailed how and what transportation accessibility is in rural, socially excluded areas. However, there is found to be a gap in how transportation is interrelated with continued urban social
exclusion (Hook 1998). The authors found serious societal and economic issues which arise when transportation access or transportation mobility is low. Reduced transportation accessibility and transportation mobility, especially for those dependent on public transit mode, has been shown in many studies to lead to disparities in equity within a
transportation network (Bowes, Ihlanfeldt 2001, Church, Frost et al. 2000, Freeman 2005, Kushto, Schofer 2008).
A study of the housing market and its reaction to a implementation of a new subway line was examined by Vladimir Bajic in Toronto, Canada(Bajic 1983). This is an older study which first examined the relationship between accessibility to a subway-transit line and its direct economic effects on the surrounding area. This study is unique because it examines a subway network, where there was not one before, in a time when major urban centers began to decline in growth. Bajic’s paper uses a hedonic pricing model to estimate housing prices before and after the implementation of the subway-metro network in Toronto. He notes that the theoretical justification for doing this is to test the importance of accessibility attributes and the importance of externalities of the subway-metro
implementation on the neighborhood characteristics. His data analysis into the subway’s effect on the housing stock most accessible to the new stations, found that the cost savings from commuting were capitalized into the housing values.
Another study takes into account accessibility values based solely on the mode of walking to the transit station. The transit mode which was expanded in this study is the bus rapid transit in Bogotá, Colombia (Munoz-Raskin 2010). The author’s examine the relation between walking accessibility and residential property values that are located within walking distance to the bus rapid transit stations. The analysis found that housing stock in the immediate area of the stations of the bus rapid transit network were placed at a premium. The author’s note that previously low income populated areas of the housing stock located near the bus stations, were being sought after by mostly the middle-income economic strata. These findings coincide with the general theory of gentrification, albeit at a mild tone based on the case study’s context and location.
For the next part of the literature review I will review studies which have examined specifically the role of segregation and/or socially excluded population and transportation. In the problem I would like to analyze is the socio-economic impact (benefits and costs) that are distributed throughout a population in a city or urban area (Litman 2002). The subject of socio-economic disparities has been studied thoroughly in the United States context since the 1960s. In Europe, some studies have looked at the development of segregated communities in highly accessible and mobile cities relative to that of United States accessibility and mobility standards (Spiekermann, Wegener 2006). Through this particular study, transportation pricing policies and transportation infrastructure
investments were run through scenarios to see which would have a greater overall impact on increasing various socio-economic indicator variables. Infrastructure investments showed to have minimal correlation with increases in economic indicator variables in these European studies. Other authors looked at urban areas with high poverty in the United States, which had either the first or extra addition of a mode of public transportation
(Glaeser L., Kahn E. et al. 2008). The Glaeser, et al. paper concluded that additions of modes of transportation can be seen to aide in the disbursement of certain economic disparity.
2.3. GENTRIFICATION AND EXCLUDED POPULATIONS
The population group that this paper deals with are groups present in neighborhoods of urban spatial inequality that accompany related socio-economic disadvantages and vulnerabilities. Lack of car ownership and inadequate public
transportation has been linked to worsening social, economic and racial exclusion in all parts of the world (Bullard 2003, Gannon, Liu 1997, Church, Frost et al. 2000).
Church, Frost et al. highlight the spatial concentrations of poorer populations in cities and focus on why they are higher frequencies in American urban centers compared to suburban areas. More than 19% of people in American cities are poor, when compared to just 7.5% percent of people living in suburban areas (Church, Frost et al. 2000). This fact leads them to look into the elastic demand of the wealthy wanting to live where land is cheap but has a trend of improving services, better known as the process of gentrification. According to the authors, this traditional explanation of gentrification and the elasticity demand model, which causes it, does not fully explain the urbanization of poverty. They find that urbanized poverty exists because of better access to public transportation in certain areas. This goes against many author’s findings of data from housing demand and accessibility that shows a relation to accessibility-poor areas to be more likely
concentrations for the urban poor (Sanchez 2008). The debate on whether public transportation is the reason why urban poor are concentrated in some areas when compared with others differs on a case-to-case basis. A regional study of European cities found that some cities have centralized areas of poverty where there is average or above average public transportation accessibility (Kingsley, Pettit 2003, Spiekermann, Wegener 2006). Other cities looked at in this study showed just the opposite in that the poorer populations where concentrated in the periphery, where accessibility to public transportation is more sporadic, with increased distances between stations and less frequent stops made at each station. These studies show that the concentration of poorer populations is complex and quite controversial as to what are the causal factors that influence the spatial concentration, or vice versa.
Social exclusion, despite the controversy of is causal foundations, is still very relevant to policy planners as well as transportation planners. Social consequences from unequal mobility affect the urban structure and make transportation systems inefficient (Lucas 2012). The effects are documented on how social exclusion has affected past transportation planning projects, but the amount of verifiable and applicable knowledge and theory which can be used for contemporary issues, as well as in future plans, is far behind where it needs to be (Kingsley, Pettit 2003). Kingsley and Pettit look into the concentration of poverty in the American context. Another policy overview, looks at
possible policy responses to social exclusion instances using an adaptation of Amartya Sen’s theory of entitlement (Preston, Rajé 2007). The paper analyzes former and current United Kingdom transportation and poverty policy to provide enhancement for
accessibility planning. The authors find that though the national government planning policy recently has promoted it is prioritization of helping the socially excluded, it critiques the policy by finding it to be to concentrated on the spatial attributes of the issue. They recommend that a schema be used that helps produce a more balanced socially,
economically and spatially differentiated conceptualization of socially excluded immobile populations. The paper highlights one of the few examples of an advanced effort by a national government to implement policy to try and solve the problem of social exclusion.
Social exclusion in the United States is both an economic and sociological issue (Mare, Bruch 2003). It has long been a topic of discussion and academic studies which use cross-sectional data to investigate patterns of segregation in American cities. The
descriptive nature of the segregation patterns areas, frequency, and history are rich with years of historical data to use. The data collected on segregation in large American cities has been heavily themed in sociological data values with intermittent historical economic data studies and evidence. Despite the history of segregation in America, little work has been done successfully in explaining the mechanism that drives neighborhoods to change on a socio-economic level (Lucas 2012, Mare, Bruch 2003). Mare and Bruch look at the city of Los Angeles and try to apply former theories to explain neighborhood change in the city’s past. They conclude that the models used do not dissect the issue enough to separate the causal patterns of racial bias and economic factors. This paper underlines the
complexity in trying to find causal relations and mechanisms for the issue of social exclusion and segregation in urban cities.
The causes of segregation have been studied with the context comparison of both European cities and American cities. A comparative study overviews the frequencies and occurrences of segregation both in the contemporary context and historical context in Europe and America (Musterd 2005). The study finds that the nature of segregation, categorized by racial diversity, is very complex in Europe because each country has very different histories, which have caused concentrations of different minorities from different originations. Adding to the historical complexities of the European continent’s segregation history is the differing practices of ‘the welfare state’ style government in each relative to a the American example. Despite the differing histories of segregation makeup, the problems of segregation in American are similar to the problems of segregation in European
countries. Groups, who are socially disadvantaged and excluded, create fewer
opportunities for the community as whole because they are less integrated with the rest society. Musterd shows that segregations problems are similar in Europe when compared to America but occur on at more moderate level. The author notes this should only be taken as a general guideline because the differences are not absolute and a case-by-case analysis would be required to compare segregation in two cities from each continent.
2.4. CONTROVERSIES
The field of transportation economics and accessibility-driven social development deals with highly complex real-world problems for which variables are hard to contain. The main controversies I have found so far with the topic of urban poverty and the effect public transportation has on it are the different conclusions between transportation
infrastructure direct investments and transportation policy-based adjustments. Since many socio-economic indicators, including income disparity, have grown or become more
concentrated in urban centers, especially in the United States, there have been some authors who argue that the current efforts of infrastructure investments are not working. I have found so far that there exists inconsistent data to support these arguments. In some cases there have been studies which show the regressive aspects of some public
transportation projects. In the United States, for instance, many of the larger, old
Metropolitan areas are serviced by public transportation systems which are accessible by over 60% of the Metropolitan population but only 20% of the jobs are accessible (within 90 minutes) by public transportation. This, in effect, creates an income inequality because individuals who are able to afford private transportation, like a personal car, have much greater work opportunities and therefore are less socially excluded.
There are many questions concerning the relation of certain socio-economic indicating variables and different types transportation. It has been debated how the mode of
transportation impacts the dispersion of economic inequalities, as public transportation has historically been used to service poorer populations. (Sandoval, Cervero et al. 2002). What I want look into more closely is whether recent transportation infrastructure
investments on already existing modal networks, have affected the socio-economic makeup of areas in spatially isolated urban neighborhoods. This has been looked at in Israel for the year of 1995 between core and peripheral cities (Leck, Bekhor et al. 2008). I would like to look at a United States city (Washington D.C.) and European city (Amsterdam, NL) to compare the impacts of a recent transportation infrastructure improvement (expansion of metro lines in the 1990s) on previously socially excluded neighborhoods within each city. I expect that the findings will be very different from the United States case to the Dutch case because of differences in economic distribution, metro line efficiency and job market locations.
3. THE CASES – WASHINGTON D.C. & AMSTERDAM
For the results of this study’s findings to be of any use for future inquiry, it is important to maintain significant external validity and internal reliability. These were important considerations in choosing the case studies of Washington D.C. and Amsterdam. For the selection of the case studies, criteria and characteristics were set which needed to be met. They had to have similar population sample sizes, population growth rates, age of transportation system implementation and city type. Both Amsterdam and Washington D.C. are scheduled to have new transportation megaproject improvements to their metro lines in the next few years. In the case of Amsterdam it is the North-South Line and in the case of Washington D.C. it is the Silver Line. The context for the cities of Amsterdam and
Washington D.C. are different in respect to each being located in different countries, yet both currently have socially excluded urban populations. Identifying whether megaprojects
were regressive or progressive in respect to the urban poor can help better understand the future impacts of the current construction projects.
3.1. BACKGROUND
The cases of Washington D.C. and Amsterdam hold the common denominator in that they were both chosen for their past subway-metro expansion, transportation
megaprojects. Other similarities exist, some shown in Table 1.1 in the introduction section of this paper. The cases are treated separately in this study because there exist different contexts, histories and populations for each case.
3.1.1. WASHINGTON D.C.
Washington D.C. is the national capitol of the United States and is a center for many governmental agencies, and thus much of the population living in it and in the outlying areas works for the federal government. The Washington D.C. city area has recently seen resurgence in its population, with over 600,000 individuals living in the city. Since the 1970s, the city has fallen victim to massive urban flight, witnessing a peak population of 802,178 in 1950, which by the year 2000 fell to 562,475 (District of Columbia Office of Planning 2014).
The demographic makeup of the city includes high numbers of American minorities, which currently account for over 60% of the city population. Despite the white population in the city being an actual “minority” within the contexts of the city, they possess a very disproportionate amount of the wealth in the city.
The specific demographics and socio-economic facts given of Washington D.C. hold true for many other American cities such as Chicago, Atlanta and Pittsburg. A major
difference between Washington D.C. and other larger American cities is the structure of the government. Washington D.C. is located in a federal district which is overseen directly by presidentially-appointed commissioners as well as elected local officials. This fact has helped Washington D.C. at times for funding of large projects such as the megaproject studied in this paper.
A transportation megaproject, which consisted of a multiple-line subway expansion, was completed in the late 1990s. The current route of this line is called the Metro Green Line and can be viewed in Figure 1.1. The project fits the definition of a transportation megaproject (Sanderson 2012, O Oliomogbe, J Smith 2013) because it required billions of US dollars in 2014 currency amounts. This project was fueled by popular interest of local politicians and the federal government to infuse an economic stimulus to attract business and people back to the city. Fighting the decreasing population trends was the primary aim of public stakeholders, while private stakeholders saw an opportunity to invest in property enhancement and potential future consumer growth in property markets.
Currently in Washington D.C. there is another megaproject planned for subway-metro expansion. This project has faced many problems associated with megaprojects (Flyvbjerg, Skamris Holm et al. 2004) including cost overruns and construction delays. The Silver Line Metro expansion is the second largest transportation megaproject currently undergoing construction in the United States at the time of this paper (McVeigh 2013).
Figure 1.1 - Washington D.C. Metro Map
Source: urbanrail.net, 2014 3.1.2 AMSTERDAM
Amsterdam is the national capital of the Netherlands and a center for national tourism, academics, and international companies. The city has a current population of 811,185. The city has seen large increases in total population over the past 20-30 years.
The demographic makeup of this city is a majority non-immigrant, or native to a Western European country. In 2005 the city consisted of 15.26% non-native inhabitants or minorities. These populations have a lower amount of capital and our spatially segregated in clusters throughout the city.
The government structure in place in Amsterdam is like that of all other Dutch cities, with the caveat that the national government views the city of Amsterdam as their
international example of national pride (Tieleman 2013).
A transportation megaproject, which consisted of multiple-line subway expansion, was completed in the late 1990s. The current route of this line is called the Metro Ring Lijn or Line 50. The route of this line can be viewed in Figure 1.2. The project fits the definition of a transportation megaproject (Sanderson 2012, O Oliomogbe, J Smith 2013) because it required close to the equivalent of 1 billion US dollars in 2014 currency amounts.
. Figure 1.2 - Amsterdam Metro Map
Source: urbanrail.net, 2014 Today the city is undertaking another transportation megaproject which can be viewed in Figure 1.2 as Metro Line 52. The project, titled the ‘Noord-Zuid Lijn’ has faced many of the problems large megaprojects run into, and has caused so much delay and extra costs that the subway line when opened will not recover its costs over a 100 year timespan (Korff, Mair et al. 2011). The project’s estimated completion date at the start of
construction in 2002 was to be 2012, but now has been pushed back to at least 2017.
3.2. COMPARISON
Comparisons of Washington D.C. and Amsterdam can be made which make them very similar cases. Amsterdam has a population density of around 4,900 people per square kilometer, and Washington D.C. has a density of 4,100 people per square kilometer. Both cities are located on each respective country’s seaboard and both cover a similar land area, Amsterdam 165 km2 and Washington D.C. 159 km2. The population of the Federal District
of Washington is 646,649 (a 2013 estimate) and for Amsterdam the population was 809,859 in December 2013 (US Census Bureau 2014, CBS Statline (in Dutch) 2013). It
should be noted that both Washington D.C. and Amsterdam have had similar population growth trends in the last 10-15 years. In the case of Washington D.C., it was one of the few large American cities to see a large population increase during the economic recession. Because of this, the housing market in the Washington D.C. area is creating some of the most expensive land values in the country at this time. Amsterdam also has high housing demand and an expensive private market. The populations in each city are becoming more and more ethnically segregated which can lead to equity issues for accessibility. Going further, both cities are major centers of government, both being capital cities. Both cities are also relatively old when compared to other cities in each respective country.
The differences of each case’s location, population trends, modal use and other factors will be in context when examining each case for cause and relations between the variables.
4. RESEARCH QUESTION
4.1. MAIN QUESTION
Based on my prior problem statement, I hereby state my main research question for my study: What are the spatio-economic effects of recent public transportation megaprojects
on socially excluded populations in the cities of Washington D.C. and Amsterdam?
4.2. SUB-QUESTIONS
In order to systematically answer the research question, I have formulated the following questions to give step-process for my research. Structuring the
sub-questions in this way will allow me to investigate the main question in depth through the data analysis results. I will further explain how these sub-questions will structure my research in my research design (Chapter 5) and will define the underlined concepts in my conceptual framework (Chapter 5).
4.2.1 SUB-QUESTION 1
1. Is there a measured increase in accessibility to populations in Amsterdam and/or
Washington D.C. after completion of each megaproject?
4.2.2. SUB-QUESTION 2
2. Did socially excluded populations in Amsterdam and/or Washington D.C.
experienced measurable increases in equity compared from before and after each megaproject?
4.2.3. SUB-QUESTION 3
3. Were changes in economic levels and changes in accessibility levels in the
4.2.4. SUB-QUESTION 4
4. After completion of the megaprojects in Amsterdam and/or Washington D.C., did the
previously socially excluded populations remain stationary or were they displaced by the process of urban gentrification?
4.3. HYPOTHESES
I have expectations for the results of my research and data analysis. Taking into account the previously discussed literature and theories, I expect that the results of my quasi-experimental research will show that there is a relationship between the
neighborhood accessibility level increases and the neighborhood equity level increases. I also expect to find that previously socially excluded populations will not have witnessed true increases in equity but rather displacement due to gentrification. Combining these two hypotheses, I expect that overall socially excluded populations are negatively affected by public transportation expansion megaprojects due to the process of gentrification which results in increased land values.
5. METHODS
The previous chapters so far have explained the aim the paper, it relevance to the scientific community, the existing literature concerning concepts core to the topics discussed in this paper, existing studies and an contextual examination of the cases of Washington D.C. and Amsterdam. The intent of this section is to describe the scientific research method used in trying to answer the main research question. This chapter consists of the research approach, explanation of concepts and variables, and concludes with the method of analysis.
5.1. EMPIRICAL APPROACH
The research design can be defined as the process and plan or framework of data collection and analysis for a study (Bryman 2012). The main method of my research is deductive by nature based on setup of my research question and hypotheses (Bryman 2012). I have based my hypotheses off of existing theories reviewed in literature on the subject of public transportation, accessibility, and equity. The general orientation to the conduct of my research, which can be defined as the research strategy (Bryman 2012), is of a quantitative research study. More specifically, my research design is a quasi-experimental design with two case studies embedded into it. Therefore my research design is a mixed methods design. Characteristics of quantitative research studies which are in my research design include that I am obtaining my variables from datasets which are compiled from housing census and survey data. These datasets are numerical and are in a normal
tabulated format. I am also obtaining quantitative data from spatial measurements using GIS tools.
5.1.1. RESEARCH DESIGN
The intent of this study is to examine possible known and unknown post-effects of two large subway-metro expansion projects on neighborhood populations in the cities each of projects are located. As covered in the previous sections, megaprojects span long
periods and take place over large swaths of cities. Therefore, the variables that interact with the planning, constructions and implantation of these projects are numerable. The variables differ on the context of the megaproject. By examining two projects of similar scope, type and time but having the megaprojects taking place in two different continents could lead to highlights of different outcomes. Furthermore, the choice of implementing the aspect of case study research into the overall research design of this paper allows for extended validity in a geographical sense for the results of the research. With the selection of two cases of recently completed transportation megaprojects, analyzing each of these cases pre-megaproject and post-megaproject leads the research to follow that of quasi-experimental design. A quasi-quasi-experimental design research is used to test hypotheses about the effects of manipulable treatments but lack the process of random assignment that occurs with true experiments (Shadish, Luellen 2005). The design of this research is to conduct a pre-test of the neighborhoods in Washington D.C. and Amsterdam and then a post-test is each city for different variables. The pre-test and post-test in this research design refer to the before and after analysis of the changed factor, the transportation megaproject.
Taking the main research question, four sub-questions are stated to perform a systematic analysis of different variables that will then help to answer the main research question. The breakdown of the main research question into four research sub-questions make the systematic approach more attainable given the time allotted for research for this particular study. Another useful aspect of this process analyzing the research question in multiple specific questions is for the easier management of the geographic analysis and statistical data analysis being used for the method of analysis. The methods of analysis for each sub-question will are explained in the Methods of Analysis section later in this
chapter.
5.1.2. UNITS OF ANALYSIS
The intent of this paper is to investigate effects of megaprojects on neighborhoods in Washington D.C. and Amsterdam. The true, underlying purpose of this type of
investigation is to find casualties that can help improve the planning process in future megaprojects which are aimed at helping the people of these cities. The fact that this study involves quantitative secondary analysis requires neighborhood statistical zones be the unit of analysis and not the individual. The statistical neighborhood zone (SNZ) is the name given to the unit of analysis in this study in order to operationalize the neighborhood
statistical and geographic data and make it easier to compare findings derived from either case study. For the case of Washington D.C., United States Census Bureau organizes
geographic, demographic and economic data into geographic zones at the smallest scale available for this study called ‘census tracts’. The tracts differ in size and shape and do not follow any particular rule for their boundaries. For the case of Amsterdam, the
Netherlands Central Bureau for Statistics (Dutch: Centraal Bureau voor de Statistiek) organizes land data, demographic data and economic data into spatial zones at the smallest level available for this study called ‘buurts’ (English: neighborhood). Like census tracts, buurts also differ in size and shape.
The units of analysis, referred to as statistical neighborhood zones forth after, are of comparable size for the purposes of this study, despite being from two different countries’ governmental statistical agencies. The case of Washington D.C. uses 187 statistical
neighborhood zones for analysis while the case of Amsterdam uses 95. Performing secondary data analysis on historical data defined using statistical neighborhood zones allows the comparison and study of variables over time or in a before-after scenario, as is the intent for this study.
When comparing the same statistical neighborhood zone from two different periods of time there are cases because of the small scale of the spatial zone when a statistical boundary does change or is merged with another statistical neighborhood zone. These instances are due to a number of factors. Construction projects which take up a whole zone such as those of roads or rail stations can cause the zone to either be spit or merged. Unique to Amsterdam is when land is added to the city through the addition of islands. In these cases the ArcMap parcel join/split function was used when possible and the data was calculated for the area.
5.2. CONCEPTS
In this section of my paper, the concepts of my research are stated in the main research question and sub-questions. Additionally the internal logic of how the concepts mentioned in the research question as well as other concepts relevant to the overall conceptual framework for research design are shown first in Figure 1.3 & in terms of concepts stated and their interrelations.
To analyze the stated research questions I will first define the concepts within each question in order to operationalize them. The following concepts will be defined:
spatio-economic; public transportation megaproject; socially excluded populations; accessibility level; equity level; process of urban gentrification
The operationalization of spatio-economic for the purposes of this study refers to any observed spatial/location changes combined with changes in equity associated with the results from the public infrastructure-improvement project. Sub-question 4 seeks to answer the “spatio” aspect of megaprojects externality effects on neighborhoods while the
“economic” aspect is analyzed in sub-questions 2 and 3. As seen in other cases, a possible
outcome for spatio changes is the movement of goods and services or the increase or decrease of demand for particular areas, such as when gentrification is evident. Economic side of spatio-economic is referred to in the main research question and singularly, in sub-questions 2 and 3. The economic variable used to analyze the neighborhoods in
Washington D.C. and Amsterdam is per capita income. Per capita income is the total income produced by a statistical neighborhood zone divided by the total population for that particular zone, for the purposes of this study.
The concept of recent public transportation megaproject is defined as a major transit project which the city has undertook to increase the capacity or service of the network. The description of “recent” defines the fact that the megaprojects being examined were
completed within the last 20 years. As covered in the theory section, megaprojects are large enterprises which have the potential to affect communities directly and indirectly over long time periods. A quantifiable definition of a public transportation megaproject as defined by the United States Federal Highway Administration is any public transportation project exceeding US$1 billion in cost and which receives a majority of governmental and public funding (Capka 2004). For the case of Washington D.C., their megaproject was all construction related to the subway expansion on the Green Line in 1991. For the case of Amsterdam, their megaproject was all construction related to the subway expansion on Metro line 50 in 1997. Though not a part of this study, current megaprojects ongoing in Washington D.C. and Amsterdam are considered future public transportation megaprojects. These projects, the Silver Line in Washington D.C. and the North-South Line in Amsterdam, have yet to be finished at the writing of this paper but offer further opportunity for
scientific studies using methods in this paper.
Social exclusion has been reviewed and defined in many instances of different existing literature and has been reviewed in the previous chapter. To operationalize the concept of
socially excluded populations, they are groups of people who have limiting factors which
repress their ability to interact with society (Vollmer 2012). In the cases of Amsterdam and Washington D.C., high percentages of socially excluded populations represent ethnic minority (Musterd 2005, Fischer 2010) and have low income levels relative to their nation’s per capita average – limiting their spending choices.
Accessibility levels referred to in my main research question and questions 1 and 3 of
the sub-questions are a function of the spatial Euclidean distance which relates to walking distance to the nearest public transportation station. The accessibility level is measured using an accessibility index derived from geographic analysis of the pre and post subway-metro locations with their relation to the neighborhoods in the cities of Washington D.C. and Amsterdam. The accessibility index methodology and process is further explained in the method of analysis section, 5.2.
Urban-gentrification in my research study is defined as the displacement of lower income
individuals by in-migrant higher income individuals due to changes in a neighborhood’s relative demand or structural makeup (Freeman 2005).
Figure 1.3 - Conceptual Framework Model
Source: Author adapted from: Banister and Thurstain-Goodwin, 2011
Internal Logic:
Investment in Public Transport (Megaproject): The process of the transportation
megaproject starts with its construction and finishes with its implementation. This section of the conceptual framework is the start of the internal logic for the process which affects populations of a city unequally. For this study, the investments in public transportation megaproject are the subway-metro expansion project occurring in the 1990s in
Washington D.C. and the subway-metro expansion project occurring in the late 1990s in Amsterdam.
Transport Demand: Transport demand is affected by the investment in public
transportation megaprojects indirectly but is driven by development in areas where accessibility changes has in made neighborhood conditions right for development.
Accessibility Change: The implementation of a transportation megaproject creates an
increase supply in accessibility through the form of more subway-metro stations. This process as shown in the framework, indirectly affects transport demand through land and property markets. For the purposes of this study, the accessibility changes being analyzed
