Today, our class focused on urban theories as it applies to South Africa. It is really fascinating to learn about these topics. I’ve lived in large cities all my life - Manila, Jakarta, New York, Madrid, Buenos Aires, Abu Dhabi - and while I can describe the differences between them, I did not consider its larger macroeconomic implications.
Glaeser’s Triumph of the Cities is our primary reading today, where he argues that cities are mankind’s greatest invention, as the headquarters of the industrialization revolution that brought much of the world’s prosperity today. I agree that cities have been an invaluable part in pulling the world out of subsistence conditions in the way that it conglomerates people, capital and ideas in dense locations that facilitate rapid production and consumption. I am relieved, however, that he does not overstate its benefits and makes sense out of why wide, recreational spaces like parks exist as well. These spaces exist not only to raise property values of nearby houses, but also as insurance against the opposite of the “eyes on the street” effect. If buildings are so high that people do not feel any responsibility for its maintenance, then high-rise buildings may not drive growth. Parks, however, may help people feel a sense of ownership to the location, because the value that parks add to their property is dependent on its safety.
I am interested in the role of parks in cities not only because of the subtle ways that it adds value, but also in how humans are inherently drawn to these places for rest. This interest stems from my enjoyment of parks such as Central Park in New York and Parque de Tres de Febrero in Buenos Aires, and partly due to how the comedy Parks and Recreation glamorizes government work. My friend who works at the Monocle, a magazine about global affairs, business, culture, and design, created a short video documentary about how urban planners are reimagining recreational spaces and integrating it into regular city life. Although the projects the film features appear exciting, I am conscious about the breadth in which these resources are shared, because it appears to cater to only the urban elite. Parks should be the most democratic places in the city, where everyone has equal access and stake in its preservation.
In the second part of class, we watched a video where Angus Deaton placed doubts on randomized controlled trials (RCTs) as the panacea in the methodological constraints that development studies frequently face. Afterwards, Prof. Buckley shared with us another short video where he compares RCTs to the popular mobile application “Angry Birds” describing them as a frantic and misguided games of trial and error. On that comparison alone, I would disagree with Deaton. RCTs, as described by Ester Duflo in Poor Economics, have a specific strategy in identifying the parameters of the experiment before evaluating a project. When addressing the problem of teacher absenteeism in India, for example, Duflo did not utilize farfetched ways to try and solve the problem. Instead, she used cameras as a tool to reinforce accountability, and it was effective in improving teacher attendance. What this ultimately produces is concrete evidence of whether a particular strategy is effective or not, and making adjustments accordingly. (On another note, Angry Birds employs the same identification strategies as RCTs. Rather than blindly shooting the birds at farfetched directions, a skilled player should make an educated guess on the angle and force by which to release the birds on the pigs. But I digress.)
Of course, this is not to say that RCT is not without its problems: its three main issues would be the size of its impact, its large time horizons, and sampling errors. RCTs make small adjustments within a specific set of variables, and therefore its impact after the study will not be significant. As with the case of teacher absenteeism, although introducing cameras increases teacher participation, the quality of the education requires further testing. In addition, some treatments require years before its findings can be identified, which means the implementation of its policies may be outdated by the time of its implementation. Habits do not change overnight, and this has to be reflected in long-term trends in data. Therefore, RCTs are limited in the scope of its effectiveness. Moreover, Deaton argues that RCTs have sampling errors because it is selected out of convenience, then randomized. This part confused me, because I don’t think that it causes much of a problem. Say, for example, that an NGO wants to study the impact of a new savings program on a village in India. Therefore, from all the households in the village, the NGO will select half through a randomized program for its treatments and the rest for its baseline. Even though the village was conveniently chosen to begin with, the village itself is precisely the target of the policy intervention. In that sense, Deaton concedes that RCTs, in these and other similar instances, are precisely the type of program that delivers the exact data that policymakers need. Other methods need to be explored, however, if we want to make quicker change.
Glaeser’s Triumph of the Cities is our primary reading today, where he argues that cities are mankind’s greatest invention, as the headquarters of the industrialization revolution that brought much of the world’s prosperity today. I agree that cities have been an invaluable part in pulling the world out of subsistence conditions in the way that it conglomerates people, capital and ideas in dense locations that facilitate rapid production and consumption. I am relieved, however, that he does not overstate its benefits and makes sense out of why wide, recreational spaces like parks exist as well. These spaces exist not only to raise property values of nearby houses, but also as insurance against the opposite of the “eyes on the street” effect. If buildings are so high that people do not feel any responsibility for its maintenance, then high-rise buildings may not drive growth. Parks, however, may help people feel a sense of ownership to the location, because the value that parks add to their property is dependent on its safety.
I am interested in the role of parks in cities not only because of the subtle ways that it adds value, but also in how humans are inherently drawn to these places for rest. This interest stems from my enjoyment of parks such as Central Park in New York and Parque de Tres de Febrero in Buenos Aires, and partly due to how the comedy Parks and Recreation glamorizes government work. My friend who works at the Monocle, a magazine about global affairs, business, culture, and design, created a short video documentary about how urban planners are reimagining recreational spaces and integrating it into regular city life. Although the projects the film features appear exciting, I am conscious about the breadth in which these resources are shared, because it appears to cater to only the urban elite. Parks should be the most democratic places in the city, where everyone has equal access and stake in its preservation.
In the second part of class, we watched a video where Angus Deaton placed doubts on randomized controlled trials (RCTs) as the panacea in the methodological constraints that development studies frequently face. Afterwards, Prof. Buckley shared with us another short video where he compares RCTs to the popular mobile application “Angry Birds” describing them as a frantic and misguided games of trial and error. On that comparison alone, I would disagree with Deaton. RCTs, as described by Ester Duflo in Poor Economics, have a specific strategy in identifying the parameters of the experiment before evaluating a project. When addressing the problem of teacher absenteeism in India, for example, Duflo did not utilize farfetched ways to try and solve the problem. Instead, she used cameras as a tool to reinforce accountability, and it was effective in improving teacher attendance. What this ultimately produces is concrete evidence of whether a particular strategy is effective or not, and making adjustments accordingly. (On another note, Angry Birds employs the same identification strategies as RCTs. Rather than blindly shooting the birds at farfetched directions, a skilled player should make an educated guess on the angle and force by which to release the birds on the pigs. But I digress.)
Of course, this is not to say that RCT is not without its problems: its three main issues would be the size of its impact, its large time horizons, and sampling errors. RCTs make small adjustments within a specific set of variables, and therefore its impact after the study will not be significant. As with the case of teacher absenteeism, although introducing cameras increases teacher participation, the quality of the education requires further testing. In addition, some treatments require years before its findings can be identified, which means the implementation of its policies may be outdated by the time of its implementation. Habits do not change overnight, and this has to be reflected in long-term trends in data. Therefore, RCTs are limited in the scope of its effectiveness. Moreover, Deaton argues that RCTs have sampling errors because it is selected out of convenience, then randomized. This part confused me, because I don’t think that it causes much of a problem. Say, for example, that an NGO wants to study the impact of a new savings program on a village in India. Therefore, from all the households in the village, the NGO will select half through a randomized program for its treatments and the rest for its baseline. Even though the village was conveniently chosen to begin with, the village itself is precisely the target of the policy intervention. In that sense, Deaton concedes that RCTs, in these and other similar instances, are precisely the type of program that delivers the exact data that policymakers need. Other methods need to be explored, however, if we want to make quicker change.
No comments:
Post a Comment