• David López García

Access to public transport and inequality of urban transaction costs in Mexico City

David López García

Ph.D. Candidate in Public and Urban Policy

The New School

Panel presentation at the American Association of Geographers (AAG) 2019 Annual Meeting, April 3rd, 2019, Washington, D.C.

Download the slides in this link


This research looks at inequalities of access to public transit in Mexico City’s metropolitan area. However, the focus of the study is not just transit infrastructure. Rather, this is an inquiry into the triangle formed by the production of housing stock, the spatial distribution of economic activity, and the role of the state in bridging people to jobs through public transit infrastructure. Above all, I’m interested in better understanding the socio-economic consequences of the role of the state in this triangle, and what these consequences mean for urban practice.


To investigate this relationship, I am conducting a qualitative study to look at the demand side of public transport in the eastern zone of Mexico City Metropolitan Area. As of now, I have conducted 34 in-depth interviews with transportation specialist and people residing in my area of study, as well as doing direct observation. My findings so far allow me to hold that in addition to the availability of public transport infrastructure, daily transport decisions made by Mexico City residents are also influenced by the circumstances that explain why people live where they do in the first place, and with the kind of jobs and the income that people have. These findings have implications for urban policy because they call for a shift in the way we plan transit infrastructures. It seems that decisions on how to transport oneself across the city are not taken solely in the sphere of the public transit options. Instead, these decisions are taken in a relational way that takes into account other spheres of the city which are usually overlooked when planning transit infrastructures.


My research consists of a qualitative study on the access to public transport in the eastern part of Mexico City metropolitan area –the zone comprised by the municipalities of Iztapalapa in what was formerly known as the Federal District, and Nezahualcóyotl, Chimalhuacán, La Paz, and Ixtapaluca, in the metropolitan neighboring State of Mexico. This area offers a fruitful research setting. It is a zone which has seen an exponential production of housing stock, in the 50s – and especially over the 80s and 90s, with a relatively small share of public infrastructure development and private productive investment. Inhabitants of Mexico City refer to this zone as being beyond “la frontera” –the border– that divides the have and the have nots of the metropolitan region. It is also referred to as a “ciudad dormitorio” –a dormitory city– where people live only to sleep. With 4.33 million people spread over 579 square kilometers, the area under study represents 7.3% of Mexico City metropolitan region’s territory but is home to 21% of its population.

I started this research by trying to assess the influence of the Subway Line A over the population residing in the area of study. The slide shows the location of the subway Line A. It started operations in 1997 and represented the only mayor infrastructure investment in the zone until 2013. However, very early in my fieldwork I realized that the influence of Line A in the zone was very small, almost insignificant. I drew up a 500-meter buffer zone on each station and went to the territory to observe how Line A had influenced people’s lives and its surrounding territory. To my surprise, I found that people near the stations only rarely considered using Line A for their daily commutes. Instead of using this line, people preferred to use other public transit systems. Something was going on. I soon realized that to really understand how people rely on public transit in this zone of the city I had to zoom out and re-adjust the scale of my analysis to adopt a regional view.


But before going any further, let me share with you the context in which I situate my study. Over the last four decades, Mexico City has gone through an intense process of urban sprawl coupled with high concentration of economic activity. The available empirical studies consistently find that urban sprawl in Mexico City continues to occur at a fast rate. The central area of the city is increasingly losing population, which is migrating to the new urbanized areas in the periphery of the metropolitan region. The rapid pace of Mexico City’s sprawl can be observed in this slide. You can see that the pace of urban sprawl between 1950 and 1980 was relatively contained in comparison with the pace of urban sprawl between 1980 and 2010.


Additionally, there is consensus that until the 70s, urban based economic activities in Mexico City followed a monocentric urban form. Since then, economic activity has dispersed towards continuous agglomerations that, together, form a large corridor of economic activity. In their influential study, professors Manuel Suárez and Javier Delgado from UNAM found that over 70 percent of the formal and informal jobs of Mexico City are located within what they call the central agglomeration. That is, a very large area contiguous to the Central Business District that takes the shape of a large corridor. The map in the slide shows the central agglomeration as identified by Suárez and Delgado. The photograph on the top is the burgeoning district of Polanco, one of the top global city hubs of Mexico City located at the heart of the central agglomeration. The photograph on the bottom shows a view from the subway station in the municipality of La Paz, the last station of the subway Line A located in my area of study and at the east of the metropolitan region. You can easily see the disparate landscapes of economic activity.


Urban sprawl and concentration of economic activity have put enormous pressure on the state’s capacity to deliver public transit systems to its population. Scholars agree that Mexico City suffers severe landscapes of spatial injustice in the access to public transit infrastructure. The slide shows an accessibility index to transit infrastructure at the census tract level developed by Gerardo Contreras and Héctor Hidalgo (2012). The map shows how the central area of Mexico City metropolitan area has increased access to transit infrastructure, while the accessibility index plummets as census tracts are more distant from the city center.


The available evidence also suggests that people residing in the peripheries face higher commuting costs both in money and time. In a comparison of transportation costs between residents of the Federal District and those of the state of Mexico, professor Priscila Connolly from the Metropolitan Autonomous University found a stark difference between the two political demarcations. The cost for a bus ride in the Federal District ranges between 0.23 and 0.46 USD, and 23% of the residents spend more than 20% of their monthly income on transportation. However, the average cost of a bus ride in the state of Mexico ranges between 0.54 and 3.84 USD and only 37% of the population spend more than 20% of their monthly income on transportation. The increasing costs can also be seen when looking at commuting time. Professor Connolly found that 7% of the Federal District residents spend more than 3 hours per day on their commutes, while this figure rises to 20% in the case of the state of Mexico residents.

The explanation usually put forward for these landscapes relates to the metropolitan governance problems. Being a tri-state area comprised of 76 municipalities, governing the provision of public transit has been historically one of the biggest problems for the city. The Federal District and the metropolitan neighboring State of Mexico have completely different transit systems that hardly interact with one another. There appears to be a natural frontier that public transit systems from each political entity cannot cross. These governance problems increase transport cost for people residing in the State of Mexico who have to commute to work in the central agglomeration. This is so because, to get to their jobs, people living in the peripheries have to use the State of Mexico bus system first, and then the Mexico City system. They have to pay twice, as the systems are not interconnected.


In this context, I went into the field to Mexico City to test the following hypothesis: inequality of access to public transit infrastructure sparks an unequal distribution of what as of now I’m calling “urban transaction costs”. In the field of economics, the term “transaction costs” is defined as the cost of making any economic trade when participating in a market. By the term urban transaction costs, I’m referring to the costs associated with living in the city to be able to participate in urban-based economic activities. This includes costs like transportation, rent, housing services, education, food, commuting time, or any other cost that people pay in order to access their economic activity. I centered my work in the analysis of urban transaction costs associated with transportation. Before going into the field, I hypothesized the underlying mechanisms through which inequality of access to public transport increases urban transaction costs associated with transportation: when people rely on public transport to access their economic activities, those living in areas with less access to public transport spend more of their income to use them. As housing prices in these kinds of neighborhoods are lower, they also tend to attract people with lower incomes and often lower educational levels. The interaction between high commuting costs and low-paid jobs increases urban transactions costs for people living in these neighborhoods. This is due to people having to spend more of their income to get to their economic activities: when they have low-paid jobs, commuting costs demand a higher percentage of their budget structure.


To test these hypotheses, I set up a qualitative study in the eastern part of Mexico City Metropolitan Area. With funding from the Mexican Council for Science and Technology and from The New School Student Research Fund, I spent the Fall of 2018 doing a research stay at the Mexican National Autonomous University (UNAM). The stay gave me the opportunity to conduct direct observation in the areas of study and 34 in-depth interviews among mobility experts, public officials, scholars, and the residents of the areas of study. I will continue to do more fieldwork over the summer. I’m planning to conduct more in-depth interviews and at least four focus groups with people living in the eastern part of the metropolitan area. Nevertheless, the empirical work conducted thus far allows me to sketch some preliminary findings. I will share three of them.


The first finding is that my area of study urbanized through three different types of housing production, and that the way through which people acquired their housing units in the area seems to have an influence on their transportation needs. I call these types of housing production a) the market, b) the governmental, and c) the social models. The market model took place during the 1950s and the 1980s. People who migrated to Mexico City from other parts of the country were eager to buy a plot of land to build a house. As land prices in the inner city were skyrocketing, the eastern part of Mexico City Metropolitan Area offered an affordable alternative for urban dwellers lacking access to either private or public housing loans. The governmental model has taken place since the 1980s and is comprised of the state efforts to implement a national housing policy that provides housing for workers. This model is best represented by the massive housing developments built under the wing of INFONAVIT, the Mexican national housing fund. Finally, the social model relates to social movements occupying land and reclaiming it to fulfill the housing needs of the poorest population of the city. This model is best depicted by social production of habitat movement identified by Enrique Ortiz and his colleagues.

These models have different implications for people’s mobility needs. On the one hand, it seems that the people that build their houses from the governmental model are more bound to their formal jobs located in the big agglomeration. This means that despite having moved kilometers away from the inner city, they need to retain their formal jobs to keep their access to their housing credits and social safety net, which means that they are inescapably bound to a 2 to 3-hour commute. On the other hand, people who got their housing units through the market and social models are less bound to a specific area of the city for access to their economic activities. These people seem to be more prone to work within the eastern part of the metropolitan area, therefore facing less demanding commutes.


A second finding suggests an interesting paradox. People living in the area of study are now third or even fourth generation since the area developed. Younger generations are increasingly improving their educational levels to achieve a college degree and even a graduate school education. As a byproduct of the concentration of economic activity in the central agglomeration of the city, the area under study continues to be deprived of high value-added economic activity. This is due to several factors. The zoning regulation by and large impedes the arrival of the large-scale industry or service sectors. The area lacks the economies of scale and economies of urbanization offered by the central agglomeration, so the zone is not competitive enough at the intra-urban scale. As a consequence, the jobs available in the zone are usually informal and low-paid. The paradox then lies in the following phenomenon: people in the eastern part of the city are increasingly more educated, however, if they aspire to get a job that is formal and that pays what people think they deserve in return for their education, they must look for that job in the central agglomeration. This, in turn, is a mechanism that inescapably increases their urban transactions costs associated with transportation.


Lastly, I found out that the decisions on how to commute to work –and thus the consequences to people’s urban transactions costs associated with transport– are less determined by the availability of public transit systems, and more determined by the kind of job that people have and by their income. During my interviews, I asked people to sketch the different alternatives that they have to commute to work. In almost every case people sketched three or four different routes. There were usually three of four different bus alternatives that people could use to connect themselves to the Mexico City subway system. There was also the alternative of buying a car and commenting to work through a private vehicle. While some of the alternatives required more time, others required to spend more money; while some alternatives require to take several transit systems, other require less effort. I then asked them to indicate which one of the alternatives is the one that they choose to use and what are the reasons for their decision. To my surprise, a clear pattern among my interviewees started to emerge. When faced with different commuting alternatives that imply a different mix of time, money, and effort, people tend to choose the one that is the fastest -even if that means spending more money and requiring more effort. This trend usually holds, unless people have a very good job that pays well enough. Let me elaborate.

When losing their job for being late is a possibility –think for instance of jobs such as a security guard, a front desk clerk, a secretary, or a worker in a factory line– people tend to choose the fastest alternative even if it is costs more money. In the words of one of my interviewees: “people in this part of the city are literally living against the clock”. These people tend to choose an expensive private express-bus service that will connect them to the overcrowded subway system. This is more costly than other alternatives, but at least they have more control over the time that they are spending on their commute and the time that they will arrive at their jobs.

Contrary to this situation, if your job pays well enough, and there is no risk of losing it for being late –think for instance of a manager in a bourgeoning industry, a lawyer in a law firm, a public official in a managerial position, or a self-employed person in the creative economy– people prefer to use their private cars, the commuting alternative which is most comfortable, even if that means spending more time and money. In Mexico City, everyone knows that getting into a car and into the traffic means spending at least a couple of hours of your day in the traffic jam. However, people who control the time at which they start work and that have well-paid jobs have no trouble investing such time and money into their commutes.


Now, what are the implications of these preliminary findings for urban practice? I will put forward one of them and leave the rest for further discussion.

In the current urban practice, public transit policy tends to be exclusively centered in transit infrastructure. However, my preliminary findings suggest that we should expand the analysis to include other variables of the urban economy. As scholars and practitioners, we tend to conduct our analysis under the assumption that people will make their commuting decisions based on the availability of transit infrastructure alternatives. As an example of this, the field of sustainable mobility is usually based on assuming that providing an improved user experience -through sustainable mass transit systems- will be enough to influence people to leave their cars in their garage and use transit systems instead. This assumption might be partially correct, but this might not be the whole story. It seems that in addition to public transit availability, people also ground their commuting decisions on the type of job that they have and their income. When people have a vulnerable job, they are willing to spend a larger share of their incomes on transport in order to minimize the chance of losing their source of income. When people have a secure and well-paid job, the share of the income that they spend in private cars to be more comfortable is considerably lower than the share of income that poorer people spend.

In addition to this, it seems that commuting decisions are not only explained by where people have to go within the city, but they are also bound by the circumstances that explain why people live where they do in the first place. In the words of professor Tom Slater from the University of Edinburgh: where people live affect their life chances, that is true, but it misses the key structural question of why people live where they do in cities in the first place.

According to my preliminary findings, it seems that public transit policy should expand its scope to include the analysis of other interacting spheres of urban policy. At least, transit infrastructure analysis should take into account the production of housing stock and why people live where they do, and the kind of labor markets and the quality of jobs provided the urban economy.

Thanks for your attention, I’m looking forward to your feedback.

16 views0 comments

Recent Posts

See All