CC BY-NC-ND
0Available on lineat www.sciencedirert.com
Transportation
Procedía
Transportation Research Procedía 25C (2017) 4495-4510 ■ ■ w «J «J
www.elsevier.com/locate/procedia
World Conference on Transport Research - WCTR 2016 Shanghai. 10-15 July 2016
Urban form and spatial urban equity in Bogota, Colombia
Luis A. Guzmana*, Juan P. Bocarejob
a Researcher, Grupo de Sostenibilidad Urbana y Regional, Universidad de los Andes, Bogota, Colombia bGrupo de Sostenibilidad Urbana y Regional, Universidad de los Andes, Bogota, Colombia
Abstract
It lias always been said that Bogota is one of the densest citi es in the world, but what is not said is that density is distributed very/ differently within the urban area, with different effects on the populati on. This moy create differences in locati on and access among socioeconomic groups. This paper shows four indicators that measure SIioss differences, contributing to inequity in Bogota. Thn theio goal of this paper is to orgniize facts ratder than ho explain their causes and quantify some date that plways have been in the minds of planners, but never hvee calculated. First, the analysis os activity balance shows a monocentric pattern that contriCutes to important Cifferenc es in travel ti me between income group s; second, spatial drstribution o f population ahe employment are different, generating higher travel times for the low-income population. Population density is higher in the boundaries oa the city whhe emgioyment desreases with distance to city center . An accessibii ity analysis for the different i ncome levels was carried out, ehowing low accessibility indexer to work for myre than 80% of the low-income population. Those gifferences in access may contribnte to tlnc important differences in mobility among socioeconomic groups. In fact, high-income groups make 150% more trips than low-income groups.
© 2017 TCS Authors Public by Бкеткг B.V.
Prer-rCTTW urnter п^роп^Н1 of WORLD CONFERENCE ON TRANSPORT RESEARCH SOCIETY.
Keywords: urban equity, Bogota, work accssibiliry, income level analysis, spatial structure
1. Introduction
The urban structure of a city is complex and it is in continuous evolution. According to Anas et al. (1998) the spatial structures of modern cities has been shaped, in large measure, by advances in transport and communication.
Corresponding author. Tel.: +57-1-3394949. E-mail address: laiguzman@uшandes.edu.co
2352-1465 © 2017 The Authors. Published by Elsevier B.V.
Peer-review under responsibility of WORLD CONFERENCE ON TRANSPORT RESEARCH SOCIETY. 10.1016/j.trpro.2017.05.345
Some studies suggest that, depending upon the urban structures of cities' opportunities and the available transport systems, access barriers or mobility restrictions may exist in particular areas (Boschmann, Kwan 2008). Further, the relationship between transport and territory, i.e. travel patterns and activity locations is a two-way interaction that largely defines the efficiency of a city and the quality of life (welfare) of its inhabitants. This welfare will be reduced to the extent that the urban structure hinders access to the opportunities offered by the city. If these barriers or restrictions depend on how rich or poor a citizen is, then there may be several problems involving equity.
In recent years, the literature has highlighted the idea that the way a city is organized in a territory, generates important environmental, economic and social effects (Brambilla, Michelangeli & Peluso 2013, Garcia-López 2010, Talen 1998). Within social effects, a very important one is equity. Most of the studies related to equity in urban areas involve accessibility analyses (Grengs 2015, Litman 2014, Welch, Mishra 2013, Bocarejo, Oviedo 2012, Korsu, Wenglenski 2010, Grengs 2010, Preston, Rajé 2007). Other studies emphasize on the importance of a job-housing balance (Loo, Chow 2011, Peng 1997, Cervero 1995). Although accessibility and a job-housing balance are good measures for assessing the distribution of opportunities, these are not enough to identify the causes of the differences found. Therefore, there is no generally accepted standard framework for measuring spatial equity in a city. The first signs of urban inequity could occur when the employment supply seems to be isolated (or away) from places where the people live and the transport system is inefficient. In Bogota, the people affected by this situation usually belong to low-income groups.
On the other hand, in Bogota the income level determines access to public transport (Bocarejo, Oviedo 2012) and therefore access to employment and other opportunities, mainly as a result of the causal relationship between location of low-income households and transport supply. But why could this happen? We suggest that land/housing values are lower in the urban periphery because they lack access to many amenities and services and therefore these are the zones where the low-income households can afford to live. For the specific case of Bogota, there are studies that support this: Rodriguez and Mojica (2009) estimate the housing value impacts because the implementation by the bus rapid transit system (Transmilenio) and Rodriguez and Targa (2004) investigates the influence of Transmilenio accessibility on housing values. They concluded there was a significant increase in housing prices due to the better access provided by Transmilenio and the urban improvements associated with the stations. So, the lack of access to employment, amenities (such as public parks) and transport could give rise to vicious loop of poverty in which low-income households are already in disadvantage.
Although the goal of this paper is primarily to organize facts rather than to explain them, we have in mind a first idea than can perhaps account for the inequalities in Bogota urban area: The most of Bogota inhabitants do not live where the opportunities (jobs) are located and the public transport (PT) system is inefficient. The interpretation of this situation is not new, but is the first time it is tested using mobility, spatial and socioeconomic data. On the whole, this work provides several key findings about the urban territory and mobility that are related to equity (job-housing balance, population distribution, accessibility and travel patterns). There are no studies in Colombia have explored the locations of housing and employment across space and across the income levels of the population. We use the term "urban spatial structure" to refer to the distribution of residents and employment opportunities within a city. This study shows a range of approaches to face the methodological challenges of measuring spatial equity. Specifically, the paper discusses the hypothesis that the mobility and accessibility issues experienced by low-income people in Bogota are partially caused by spatial factors. In this study we addressed two central questions using the spatial distribution of the population and employment in Bogota, as a case study:
• Is there spatial equity across socioeconomic levels in the territory of Bogota, in terms of a balanced measure of
access to employment?
• Is there equity in urban mobility according to the socioeconomic levels?
At the moment, the main objective of our research is mainly exploratory. However, determining the spatial inequalities in the city of Bogota may have important implications for decision makers. It may allow, through
further research, the evaluation of the success of public policies, and the efficiency of the urban system and its equity impact. Thus, using the Bogota Travel Survey 2011 data (SDM 2011), we identify significant inequalities in the travel patterns of low-income population in comparison to their higher income counter parts.
This research suffers from two limitations: First, the costs of transport supply (such as PT supply) have not been incorporated, and second, it could be argued that this analysis is too aggregate. Despite these limitations, the results provide a starting point for an equity analysis of the activity locations and travel patterns in developing cities.
2. Spatial structure matters to equity?
The urban spatial structure has attracted interest from urban and transport planners, as it is believed to affect the economic development, the welfare of cities and their inhabitants and in general, the sustainable development (Meijers, Burger 2010). Given that urban territory is spatially heterogeneous, and accessibility levels are highly variable, it is inevitable that some locations will have better accessibility levels to urban services. It is one thing to have differences in accessibility, but it is quite another if these differences are detrimental to a large part of the population, generally low-income population, which can result in some types of exclusion.
The urban structure of cities generates various types of social effects. An unequal spatial structure affects the quality of life of inhabitants, particularly those with low incomes, by increasing the time spent to traveling, providing low access to urban amenities, which impact local air pollution and climate change. It also contributes to increases in transport costs by unnecessarily increasing travel times between households and activity centers. Leaving aside the condition of lower levels of education, some experts argue that the low quality of life of low-income people is also partly caused by two urban factors: low accessibility levels to workplaces and exposure to the negative externalities of concentrated poverty in deprived zones (Korsu, Wenglenski 2010). Therefore, is there spatial equity in access to jobs (or leisure) in urban territories; i.e., do all socioeconomic groups have a similar opportunity to experience the city?
Accordingly, what does it mean for the urban structure to be spatially equitable? As an initial assessment and considering the influence of market pressures on housing prices, we believe that it can be achieved when the location of people, regardless of their income level, does not affect their access to the opportunities of the city and the cost of reaching such opportunities is proportional to their income. Korsu and Wenglenski (2010), say there are two general reasons for this poor accessibility: First, poor people generally live in zones that have few jobs and are located far from the main employment centers. Consequently, the average distance between workers' residences and workplaces is quite high.
Moreover, the balance of jobs and housing across the urban area is very important: In a well-adjusted urban structure, a balanced territory should mitigate traffic congestion, as well as the related social and environmental problems related to it (Peng 1997). This would help to greatly mitigate many inequities. This balancing is closely related to self-contained areas (Cervero 1995). This concept assumes that there are many trips within (or internal to) a zone, and thus, it will be more likely to occur in no-motorized transport modes. However, it is necessary to be careful, because this could lead to problems of segregation and fragmentation in the city. In Bogota, the downside is that differences between location of housing and jobs causes some pessimism, e.g. the social housing projects are developed where the land values are lower, without an adequate public transport system and where there are not employment centers. The policies of urban planning and mobility are not coordinated. This answer could have significant policy implications, as socioeconomic biases in the availability of employment or leisure sites, including green/recreation spaces, could be a symptom of, or even exacerbate, disadvantages (Shanahan et al. 2014).
The equity approach developed in urban planning requires that urban amenities and workplaces must be evenly available in such a way that everyone receives a similar treatment, regardless of their income level, housing location or ability to pay (Talen 1998). From this framework, we want to show how Bogota's current urban structure can be a barrier, together with the transport system, particularly for low-income people, which prevents them from easily
reaching the opportunities the city offers. We have called this "spatial urban equity". Although there may be several definitions of spatial equity (Kunzmann 1998), this work is focused on equal access to employment opportunities, measured in costs and travel times, which depends on the location of housing. Furthermore, spatial equity could be a controversial concept for urban planning, and it may not necessarily lead to an increase in social justice (Kunzmann 1998).
3. Bogota's background
The urban area of the city of Bogota is divided into 112 urban "zonal planning units" (UPZ) that occupy about 365 km2. UPZ are territorial units used to plan urban development at the zonal level, and they follow recognizable boundaries such as roads and natural barriers. The individual UPZ units in Bogota range in size from 0.4 to 9.0 km2. According to the Urban Planning Office (SDP) in 2011, the Bogota area has 7.35 million people (2.24 million of households), and it is estimated that its population will reach about 8.30 million in 2020. Additionally, the 12 surrounding municipalities of the city comprised an area which extends across 1,194 km2 with a population of 1.22 million inhabitants (see Fig. 1). This analysis is only for Bogota city.
In recent years, many aspects in Bogota city have improved for various sectors of the population, particularly in some low-income areas, with the implementation of Transmilenio (the BRT system). However, it is obvious that much remains to be done, particularly in relation to equity. For instance, in wealth distribution, Colombia is the one of the most unequal countries in the world, with a GINI coefficient of 0.536 (ECLAC 2013). In Bogota, although
Fig. 1. Bogota region (municipalities and Bogota City)
poverty has declined according to the National Statistics Department, the distribution of wealth as measured by the Gini index was 0.504 in 2013, which was 1.4% worse than the previous year.
In order to know the socioeconomic characteristics and urban mobility patterns of the population of Bogota, a mobility survey was conducted in 2011 (SDM 2011), in which each household was asked, among other things, their average income per month. The responses were classified in different categories. These categories include eight predefined ranges (1 USD = 1,900 COP approx., in COP 2011):
Range 1: ^ $280 Range 2: $280 - $630 Range 3: $630 - $1,050 Range 4: $1,050 - $1,475 Range 5: $1,475 - $2,105 Range 6: $2,105 - $2,895 Range 7: $2,895 - $4,210 Range 8: > $4,210
The distribution of Bogota households, by these income ranges is shown in Fig. 2(left). The spatial distribution is shown in Fig. 2(right). According to the data obtained, 66% of the households in the city belong to the lowest income ranges (1 and 2), and 82% of households in the city earn less than $l,050/month.
900,000
700,000
'S 500,000
5 400,000
200,000
1 2 3 4 5 6 7 HH income per month (ranges)
Income distribution of households
----BRT system
- Main road network
Income level by zone
Fig. 2. Income distribution of households and zones 2011
The median household income (50th percentile) is $636. In some cases, for ease of analysis, the city was grouped into three income-level categories based on these data (low, medium and high). These groups are used to classify the analysis zones in order to find particular commonalities between them. In this classification, low-income groups are
zones below 50th percentile (ranges 1 and 2); medium-income is between 50 and 80th percentile (ranges 3 and 4); and high-income the rest.
Bogota city holds some particularities in terms of density: There are important differences in the population and activities (jobs) location throughout urban area. Bogota has a large housing deficit, particularly in social housing, which has encouraged an urban development based in unplanned and informal settlements. As a consequence, many informal neighborhoods emerged on the city periphery with poor urban living conditions. It is in these zones that the highest densities occur. Fig. 3 shows the population density (left) and job density (right). The most striking observations from Fig. 3(left) are a very high densities in urban periphery where are not enough jobs and urban amenities for the resident population. As for job location, is the dominance of a large central core along major road corridors, and especially, the northern part of the city center (the richest zones). With this information, it is possible to note that in general, people do not live where the jobs are located.
Population density Employment density
Fig. 3. Population and employment density
According to their average income level, different types of construction tend to reduce residential floor consumption per household: Less income means smaller housing units, and we must also note that low-income households have, on average, a higher number of members; more than half of them have more than four members. The present average residential floor consumption in Bogota is 20 m2 per person in low-income zones, 35 m2 in medium-income zones and 80 m2 in high-income zones. These initial data tend to show the first signs of some disparities between income groups in the city.
In order to identify the higher economic activity zone and use it as the business city center, this study uses the UPZ with highest employment density relative to all city zones as the city center. As a result, we used UPZ #97 as
the CBD (see Fig. 1). This zone, known as UPZ Chico-Lago is the zone with larger number of jobs and is one of the richest zones as well. Table 1 shows the top 10 UPZ with an employment density (including formal and informal employment) greater than 25,000 jobs/km2, and it provides summary statistics. These zones are the darkest zones in the Fig. 3.
Table 1. Zones in order of employment density
UPZ UPZ Name Employment density [jobs/km2] Employment [No.] Employment/ Population ratio Distance from CBD [km]a
97b Chico-Lago 85,696 362,020 14.7 1.55c
91 Sagrado Corazon 72,696 73,991 13.7 6.51
99 Chapinero 63,057 99,812 5.2 4.32
93 Las Nieves 54,324 82,085 6.5 8.23
105 Jardin Botanico 37,170 21,846 12.6 7.83
108 Zona Industrial 34,571 119,239 30.5 9.00
101 Teusaquillo 33,826 78,723 3.1 6.05
102 La Sabana 32,766 145,285 3.0 8.50
111 Puente Aranda 28,256 100,719 7.5 9.34
98 Los Alcazares 27,665 113,742 1.5 2.79
a Average travel distance in motorized modes b City CBD
c Internal average travel distance
Regarding transport system, contrary to what one might think, informal transport has not an important share (0.8% in work trips and 0.6% in total trips). According to the data shown in Fig. 4 (based on mobility survey 2011), the transport mode most used to travel to work is the public transport system (BRT and regular buses) with more than a half of all daily work trips. Walking and bicycle trips are more frequent in the low-income groups; while in the high-income groups is the car and taxi. In all cases, non-motorized modes represent at least 10% of work trips.
Informal _
0.8% J
Work trips modal split
Low-lnc i Public transport
i Motorcycle
Others
n-motoriz ed Non-motorized
n-motoriz 24.3% ed 16.9%
Taxi 5.2% raxi 10.5°/
8.2% Public Car 18.0% Car 40.5%
Public
transport 57.6% transport Public transport 32.3%
Medium-lnc Car
i Non-motorized
High-lnc i Taxi
Informal
Work trips modal split by income group
Fig. 4. Modal split (work trips)
As seen, most of the low-income households are located on the city periphery, where public transport system has some supply problems. This and in many cases the fare costs, makes many people have no choice but to travel on foot. Consequently, low-income households face low accessibility levels to jobs, especially relative to households with higher income levels (Bocarejo, Oviedo 2012).
4. Comparing urban conditions among income groups
The process of urbanization in Bogota has drawn attention in recent years. However, the evolution of its urban form, shaped by the informal settlements, regulations, transport infrastructures and also by the complex interaction between market forces, has not been sufficiently monitored, which makes it difficult to analyze over time. This study makes an effort in that direction. To do this, we use some indicators of the spatial organization of Bogota's urban area. Our analysis of spatial urban equity requires indicators to be comparable on a common scale across income groups. Results are examined from a number of perspectives by geographic region and income level in a simple way.
4.1. Territorial balance
The high population growth rate and the migration from outside that has occurred over the past decades have caused important changes in the size and structure of the city. Furthermore, the new residents are mostly poor, and due to the location of their housing, they have long commutes, due to the lack of economic activities available in their area. As a result, these inequalities can be dangerously acute, because the travelers are forced to spend ever-increasing amounts of time and money on travel, which are far above those found in developed cities.
With this in mind, one difficulty that must be faced in describing the spatial distributions of the population and workplaces in an urban area is that it can be somewhat arbitrary, especially for cities with a history of growth based on informal settlements. However, using an aggregate description, it is possible to calculate the spatial distributions and their responsiveness to and interaction with travel behavior. This process was derived from empirical analysis of data from Bogota's mobility survey and is kept to a UPZ level, in order to illuminate the basic principles of the location of population and employment by the distribution of its distances from the CBD and analyze its differences.
Those differences refer to the distribution of employment relative to the distribution of inhabitants within a given geographic area and would have any significant impact on traffic congestion or air pollution (the jobs-housing balance)1. Thus, an interesting and first result is shown in Fig. 5, in which the differential accumulation estimated curves between population and employment by income groups are presented. These curves represent the differences between the location of residences and jobs with respect to the CBD, i.e. if the number of people is greater than the number of jobs when starting from the city border towards the CBD, the curves represent the proportion of workers who have to travel farther (in average) in order to reach their jobs.
These curves are similar in form, although they vary in height and position over the x-axis. Then, the scale and shape of the curves represent the required amount of travel under the different spatial distributions of people and jobs by income group. The curves reach their greatest difference according to the income group: the zones belong to the high-income group, have their greatest difference near 4.0 km from the CBD. In the medium and low income, these points are from 9.7 and 11.4 km, respectively. These peaks means that high-income households are closer to the major employment concentration area, medium-income households a little further, and poor households even further. In addition, for the medium-income zones there is a relative territorial balance. This provides a first indication regarding the territorial balance in Bogota city.
1 Assuming that most of jobs in a zone are occupied by people who live around and leaving aside several other elements when trading off commuting time such as the amenities they could obtain at different housing locations. However, note that 44% of the total workplaces in Bogota are concentrated in the big central east core (12% of the urban area). See Fig. 3.
5 10 15
Distance to CBD [km]
. 5. Differential accumulation distributions of population and workplaces by income group
Other methods of measuring the jobs-housing balance have been used in previous researches (Peng 1997). In this case, we believe it would be useful the results shown in Fig. 6 where the size of the bubble represents the average income. The average motorized travel time to work by UPZ, have almost an 'L'-shaped relationship decreasing with the jobs-housing ratio. In zones where there are few workplaces (compared to the resident population), which usually are low-income zones, the travel time remain high. As is shown, most of the bigger bubbles are located below 60 min of travel time. The results also shows that in average, the low-income zones has a job-housing ratio equal to 1.0, the medium-income zones 1.9 and the high-income, 3.6. This corresponds to average travel times to work: 69, 56, and 48 minutes, respectively.
_ 90 c
— 80
O) CUD (TJ
$630 i $1,475
>$2,500
0 10 20 30
Job-Household ratio
Fig. 6. Motorized travel time to work and the jobs-housing ratio by average income
In summary, the households located in low-income (lower job-household ratio) zones must spend more time traveling to work. This is one of several explanations of why the low-income people spend about 42% more time travelling to work than the wealthy segment. This spatial configuration has created an extremely unequal spatial configuration, with the most vulnerable groups located in the farthest and least accessible places.
Assuming the simplifications of this approach, we can see that, in general, Bogota's territory tends to be unbalanced, which implies a scenario involving a higher transport demand, as these population and unbalanced concentrations occur at the periphery of the city, where the public transport supply has some deficiencies. The jobs-housing imbalance in the low-income zones tends to correlate with high levels of congestion and again, with high travel times.
4.2. Spatial distribution of population
Bogota is one of the densest cities in the world, and it is the densest in Latin America with around 20,000 inh/km2 in 2011. However, it is more important to know the spatial population distribution through urban area. This indicator explores the spatial differences in population distribution, their densities and their implications in average travel times.
The distribution of population in Bogota does not follow the usual pattern with a negative slope with distance from the CBD (McMillen 2006). In this city, as we can see in Fig. 7, the distribution of the population density shows the highest population densities are located in the periphery of the city, while the lower densities are in the central areas, which also have the highest proportion of high rise buildings and high income level population. The profile of Fig. 7 (left) shows that the population density increases as one goes farther away from the CBD. This particular spatial pattern lengthens travel distances for commuting trips in the city, especially for the low-income households (more people traveling farther). The periphery settlements are mostly comprised of low-income households and are located precisely within a radius of 10-20 km from the CBD. Meanwhile, the highest income group lives in lower density residential areas, close to their jobs, with the option of car use (higher car ownership rates), and they are well served by public transport. Fig. 7 (right) shows an employment density profile, which has an inverse shape to population density profile.
4 6 8 10 12 14 16 18 20 23 >23 Distance to CBD [km]
Population density
4 6 8 10 12 14 16 18 20 23 >23 Distance to CBD [km]
Jobs density
Fig. 7. Density gradient in Bogota urban area by income groups
It seems that households' location, through the informal growth process, have segregated the residents of Bogota. However, Bogota's social segregation has always been clearly visible, with the north of the city occupied primarily by wealthy people and the south by the poor. Despite this, if we discriminate the data of Fig. 7 by income level, the inequality seems even greater: The low-income group is located in the areas farthest from the CBD; we are not referring to distance, because 10 or 18 km, on average, is a relatively short distance to be covered by the transport
system, even for cycling (Heinen, van Wee & Maat 2010). The point here is that the poor people must endure high travel times. These results show that there is a high concentration of low-income households that are located more than 11 km or more than 65 min/trip from the CBD. As expected, the densest areas, which, in turn, are the poorest, have in average, the highest travel times to reach CBD.
In addition, there is a continuing dynamic towards more densification even though the city already has a high-density profile: almost half of the population (40%) lives in densities close to 30,000 inh/km2. Fig. 8 shows the frequency distribution of population densities, indicating that, in recent years, there has been an increase in the population density of zones that are already hyper-dense (>40,000 inh/km2). Furthermore, these zones are also where most of the low-income households are located (bars, more than 80%). Note that, a significant proportion of the population (20%) lives at moderate densities (in the Bogota context), particularly zones where there are fewer poor households.
10,000 20,000 30,000 40,000 50,000 60,000 >60,000 Population density [inh/km2]
Fig. 8. Frequency distribution of population at different densities
This analysis also reveals the concentration process (higher densities) occurring in Bogota, particularly in zones with a high proportion of low-income households.
4.3. Vertical equity in accessibility
Vertical equity requires that disadvantaged people be identified and given special consideration in planning (Litman 2014). The method selected to determine a measurement of inequality in this case is the Lorenz curve, which is based on a Gini coefficient that describes the unevenness of distribution (Delbosc, Currie 2011). This is a new approach to assessing the equity of accessibility levels according zone of residence and income level. The concept is useful in describing inequality among groups of individuals, where a cumulative proportion of some assets are plotted against cumulative proportion of the population. A low Gini coefficient, close to zero, depicts an equal distribution and is assumed to describe an even distribution of accessibility levels in the population. Similarly, a high coefficient (concave curve) shows an uneven distribution. This method was used to compare an accessibility index to the population of Bogota to evaluate vertical equity.
We evaluated whether accessibility to work is evenly distributed among income groups to identify disadvantages groups. There are several ways to define work accessibility, but the number of jobs that are reachable within a given
motorized travel time is one of the most comprehensible and transparent, as well as the most directly comparable across zones. In this case, we choose a contour measure (Geurs, Ritsema van Eck 2001) or an isochronic measure (Cervero 2005) with 45 minutes as the maximum travel time, as is shown in Fig. 9. We describe this as an accessibility index: The higher the index, more jobs can be reached within 45 minutes travel.
900 800
a> 300
0 20 40 60 80 100
% Population
Fig. 9. Jobs accessed in 45 min travel time by income groups
The approach to comparing how the income groups experience job-accessibility levels is to plot this index against the share of the city population. Fig. 9 shows the approach, where the accessibility index is on the y-axis and the percentile of population on the x-axis. This chart shows the contrast between the shapes of the three curves. Here, the high and medium-income group curves proceeds almost in a straight rise. Thus, accessibility is evenly spread among into this groups: About half of them experience high accessibility (greater than average), and about half experience low accessibility. In the same way, the high-income group enjoys higher accessibility levels than average of around 65%.
By contrast, for the low-income group, the curve has a significant concavity. Almost 80% of them tend to experience low accessibility levels, compared to the other income groups. However, there is a small portion of the low-income population (15%) that enjoys high accessibility levels. This is because this 15% of people who lives in the low-income zones are located either in close proximity to the main activity areas.
This result suggests that the low-income group generally suffers from a positional disadvantage: They tend to live away from the main job-destinations, and with the low travel speeds endured by a large share of low-income people, who are public transport-dependent, any advantages provided by location would be useless.
4.4. Urban Mobility
Although Bogota is a dense city, with average travel distances that are relatively short, the travel times are high, the roads are in poor condition and the congestion is too high (also in the BRT system). Households from different income groups experience extremely unequal conditions in terms of density (see Fig. 7), urban quality and additionally distinctive mobility patterns and trip rates.
The destination zones, for both commuting and other trips, show that the CBD and surrounding areas are actually a large trip attractor area. In Bogota city on a typical day, about 15.4 million trips are made, of which 48% are motorized. Of the total trips, 25% are made by commuters (workers and students). If jobs were uniformly distributed between all 112 UPZ, each zone would get around 0.9% of all the trips; however, the CBD and surrounding areas attracts 41% of all job commuting motorized trips and approximately 15% of all "other" trips. Given the high density of Bogota, a large number of "other" trips are made using non-motorized modes to travel to local shops. In a comparison by income levels, 57% of low-income people walk or ride bikes, while only 23% of high-income people use these transport modes. With regard to work trips, 25% of low-income people use these modes, while just 11% of the richest use them. However, this does not mean that people prefer to walk (or ride a bicycle) instead of using public transport; rather, in many cases, they simply have no choice but to walk, because they cannot afford public transport fares or the supply is insufficient.
Despite the construction of the Transmilenio system, and the slight decrease in travel times in the city following its construction, for the low-income group, travel times remain very high, almost twice the travel times of the high-income group. Since lower-income households locate mainly on the south and west sides of the city, away from the areas with a higher density of work opportunities, the effects of such a spatial mismatch include high travel costs and large travel times (see Fig. 10).
Averagelravel cost PT [% of income)"
<3% _1 3% - 6%
BRT system Main road network Feeder lines
- - - - BRT system
-Main road network
Feeder lines
Average travel costs by public transport (% Average travel time by public transport of income, work trips) (work trips)
Fig. 10. Travel costs and average travel time per trip by public transport
The public transport system in Bogota is greatly affected by inefficiency and is still seen as a service for poor people. The overcrowding, unreliable service, and insufficient supply, especially during peak hours, as well as the congested roadways, reduce the quality of life of low-income people who often have no other choice for their travel.
Low-income households suffer from low levels of mobility and accessibility, much more than those in the wealthy zones. Furthermore, the low-income households depend on PT far more than rich people: The lower income groups are about 4.5 times as likely as the high income group to make a trip by PT. If we examine travel on foot or by bicycle, this relationship rises to 19 times more likely. In fact, low-income groups are disproportionately dependent on public transport and non-motorized transport modes. They also typically have long commutes, because there are few economic opportunities near the zones in which they reside.
On the other hand, the differences between the money spend on travel to work (in PT) and their income is quite dramatic. In average, the low-income households spend 30% more time and more than double money that the wealthy households. This may partly explain the differences between trip rates by income group: a person belonging to the highest income range makes around 20% more trips per day than a person from lowest income range. If motorized trips are compared, the difference can be up to 150%. Finally, car ownership is also unequal: In low-income zones, there is one car for every five households, in middle-income zones, one in every two households has at least one vehicle and in high-income zones there are 1.3 cars by each household.
5. Conclusions
In this work, we analyze, from different points of view, the location of households in the city of Bogota, regarding the employment centers, the income levels and the spatial distribution among the population. The variations in the housing locations and the availability of urban land or built areas observed in this study highlight an important spatial equity issue for developing cities that is represented by a systematic reduction in the quality of life within socioeconomically disadvantaged zones. People from different income zones experience extremely unequal conditions, in terms of land consumption, location and distinctive mobility patterns.
Although Bogota's spatial structure has two main features which make the operation of public transport at a large scale relatively easy, i.e. its high average density and the fact that have a main activity central area, our analysis of spatial equity found that the low-income households are disadvantaged in location (housing location vs. employment location) and mobility (travel times and costs). Besides, an important share of the low-income households experience low levels of accessibility, much lower than the average, as a result of their disproportionately low access to workplaces and poor quality of public transport.
The city has a geographical distribution of work opportunities that benefit higher-income population. This been consolidated by historic patterns of land occupation. In land-use planning, the lack of instruments to influence the socio-spatial distribution of urban spaces to allow a more equitable distribution of access to economic activity areas has generated considerable mobility gaps between groups with high and low income. In summary, from a social equity standpoint, where a household is located has a strong effect on the capacity to achieve better opportunities. Thus, it is essential to design and implement measures to address the large differences in access to opportunities, because many of the disadvantages experienced by most of the residents in Bogota stem primarily from their location in the urban area and its relationship with the transport system.
Special care should be exercised in policy decisions related to social housing or public transport subsidies. In general and although the relationship between travel demand and urban structure is the subject of at least two decades of intense debate, no major policy decision with respect to travel or city structure should be made before we have a better understanding of the close interactions between travel demand and urban structure, as well as their interaction over time. If employment is "the" activity, which means that it is the source of income that enables other activities, as Loo and Chow (2011) indicate, the city should work harder to encourage the location of businesses within the proximity of residential developments, particularly in the most disadvantaged areas of the city, as well as to improve the transport infrastructure, the quality of public transport and urban amenities, or a combination of all of the above.
From the point of view of the transport system, affordable and accessible public transport is seen as a viable tool to deal with this kind of inequity (Modarres 2003). To close the gap between housing locations, especially social housing, and employment locations an efficient public transport system can be used as a tool to improve the distribution of opportunities to access economic activities or the distribution of urban amenities among the low-income population. The main conclusion of this exploratory study is that these results could be a first step for building a roadmap to further research that will improve our understanding of the causes of the uneven accessibility levels in Bogota. We expect in future research to explore more about the causes or appropriate fixes for these problems.
6. References
Anas, A., Arnott, R. and Small, K.A. (1998) 'Urban Spatial Structure', Journal of Economic Literature, 36(3), pp. 1426-1464.
Bocarejo, J.P. and Oviedo, D.R. (2012) 'Transport accessibility and social inequities: a tool for identification of mobility needs and evaluation of transport investments', Journal of Transport Geography, 24, pp. 142-154.
Boschmann, E.E. and Kwan, M. (2008) 'Toward Socially Sustainable Urban Transportation: Progress and Potentials', International Journal of Sustainable Transportation, 2(3), pp. 138-157.
Brambilla, M., Michelangeli, A. and Peluso, E. (2013) 'Equity in the City: On Measuring Urban (Ine)quality of Life', Urban Studies, 50(16), pp. 3205-3224.
Cervero, R. (2005) 'Accessible Cities and Regions: A Framework for Sustainable Transport and Urbanism in the 21st Century', UC Berkeley Centerfor Future Urban Transport: A Volvo Center of Excellence.
Cervero, R. (1995) 'Planned Communities, Self-containment and Commuting: A Cross-national Perspective', Urban Studies, 32(7), pp. 1135-1161.
Delbosc, A. and Currie, G. (2011) 'Using Lorenz curves to assess public transport equity', Journal of Transport Geography, 19(6), pp. 1252-1259.
ECLAC (2013) 'Social Panorama of Latin America', .
Garcia-Lopez, M. (2010) 'Population suburbanization in Barcelona, 1991-2005: Is its spatial structure changing?', Journal of Housing Economics, 19(2), pp. 119-132.
Geurs, K.T. and Ritsema van Eck, J.R. (2001) 'Accessibility measures: review and applications', .
Grengs, J. (2015) 'Nonwork Accessibility as a Social Equity Indicator', International Journal of Sustainable Transportation, 9(1), pp. 1-14.
Grengs, J. (2010) 'Job accessibility and the modal mismatch in Detroit', Journal of Transport Geography, 18(1), pp. 42-54.
Heinen, E., van Wee, B. and Maat, K. (2010) 'Commuting by Bicycle: An Overview of the Literature', Transport Reviews, 30(1), pp. 59-96.
Korsu, E. and Wenglenski, S. (2010) 'Job Accessibility, Residential Segregation and Risk of Long-term Unemployment in the Paris Region', Urban Studies, 47(11), pp. 2279-2324.
Kunzmann, K.R. (1998) 'Planning for spatial equity in Europe', International Planning Studies, 3(1), pp. 101-120.
Litman, T. (2014) 'Evaluating Transportation Equity: Guidance For Incorporating Distributional Impacts in Transportation', .
Loo, B.P.Y. and Chow, A.S.Y. (2011) 'Jobs-housing balance in an era of population decentralization: An analytical framework and a case study', Journal of Transport Geography, 19(4), pp. 552-562.
McMillen, D.P. (2006) 'Testing for Monocentricity', in Anonymous A Companion to Urban Economics, Blackwell Publishing Ltd.
Meijers, E. and Burger, M. (2010) 'Spatial structure and productivity in US metropolitan areas', Environment and Planning A, 42(6), pp. 1383-1402.
Modarres, A. (2003) 'Polycentricity and transit service', Transportation Research Part A: Policy and Practice, 37(10), pp. 841-864.
Peng, Z. (1997) 'The Jobs-Housing Balance and Urban Commuting', Urban Studies, 34(8), pp. 1215-1235.
Preston, J. and Rajé, F. (2007) 'Accessibility, mobility and transport-related social exclusion', Journal of Transport Geography, 15(3), pp. 151-160.
Rodríguez, D.A. and Mojica, C.H. (2009) 'Capitalization of BRT network expansions effects into prices of nonexpansion areas', Transportation Research Part A: Policy and Practice, 43(5), pp. 560-571.
Rodríguez, D.A. and Targa, F. (2004) 'Value of accessibility to Bogota's bus rapid transit system', Transport Reviews, 24(5), pp. 587-610.
SDM (2011) 'Encuesta de Movilidad de Bogotá 2011', .
Shanahan, D.F., Lin, B.B., Gaston, K.J., Bush, R. and Fuller, R.A. (2014) 'Socio-economic inequalities in access to nature on public and private lands: A case study from Brisbane, Australia', Landscape and Urban Planning, 130(0), pp. 14-23.
Talen, E. (1998) 'Visualizing Fairness: Equity Maps for Planners', Journal of the American Planning Association, 64(1), pp. 22-38.
Welch, T.F. and Mishra, S. (2013) 'A measure of equity for public transit connectivity', Journal of Transport Geography, 33(0), pp. 29-41-41.
