Stacked affluence: how vertical neighbourhoods shape the segregation of affluence in Brazilian cities

For an upper-middle class household in Latin America, residing in an apartment tower  with good access to jobs, services and amenities seems like a natural choice. In cities with high levels of congestion, low decentralization of jobs, highly concentrated amenities, and few transport alternatives, vertical neighborhoods appear as the demand for access of the more affluent grows.  In fact, many well-to-do neighborhoods in Latin American cities are also vertical neighborhoods.

In recent work on income segregation in Brazilian cities, published in the OECD report Divided Cities: Understanding Intra-Urban Inequalities, I explore whether verticalization bears any relationship with the high observed levels of segregation of the affluent. Preliminary visual evidence for some cities, including Rio de Janeiro, clearly points to a correlation between the percentage of affluent people and the percentage of people residing in apartment towers across neighborhoods.

To test the relationship between concentration in vertical neighborhoods and income segregation, I construct an index of exposure to dwellers in other type of buildings and a measure of segregation of affluence based on an ordinal entropy index. I find that vertical neighborhoods are indeed part of the explanation behind the segregation of the affluent in Brazilian cities, even after controlling for city size, income inequality and other city-level variables. Whether land use regulations have encouraged the surge of vertical neighborhoods concentrating high income households, and how can they be used to ensure affordable housing in central areas in Latin American cities are two interesting questions for future research.

Working with Origin-Destination data in R made easier by stplanr

My co-author Robin Lovelace has recently made available  a new version of his R package stplanr, which helps processing and visualizing Origin-Destination transport data in R. It integrates the useful dist_google function to get network distances and time using the Google Maps API, which we used in our accessibility to schools in São Paulo paper. Highly recommended!

Reproduce Joy Division cover with Population Density Lines in R

An excellent post by Rafael Pereira about tools for making interactive nice-looking density lines maps using R and ggplot2. You can also reproduce this famous cover if you are at it!

 

 

 

New Working Paper on Income Segregation and Urban Spatial Structure

A working paper co-authored with Miquel Àngel García López entitled “Income Segregation and Urban Spatial Structure: Evidence from Brazil” is now available as part of the CAF Working Paper Series. In this work, we estimate the effect of urban spatial structure on income segregation in using data for 121 Brazilian cities between 2000 and 2010. We show how the effect of local density varies between monocentric and polycentric cities, and between income groups.

This paper is part of a line of research trying to link the distribution of employment within cities with the distribution of the population by income groups, in order to understand the possible causes of residential segregation by level of income in urban areas.

 

A great idea: R Markdown for Undergrads

Great post about introducing students to the fundamental concept of reproducibility in science while teaching them R

Landscape Ecology 2.0

A recently published paper by Baumer et al (2014) caught my eye today (HT to Bruce Caron).  I wanted to share it here because I thought it was cool and also had a few comments to make about some of the issues the authors raised.

First, a bit about the paper.  Partly in response to all the media attention to the crisis in reproducibility in science (e.g. Nature) Baumer and colleagues made some changes to introductory statistics classes at Duke, Smith, and Amherst.  The primary change was to require the use of R Markdown for all homework.  RStudio was the editor they used and it appears any cutting and pasting of code, figures, etc. was not allowed.  They conducted a survey of the students early in the class and after the class.  The end result was that students preferred using R Markdown over the typical mode of cut and…

View original post 259 more words

New R package to characterize the spatial structure of point patterns

Erik Marcon, Stephane Traissac, Florence Puech and Gabriel Lang have developed the R package dbmss that provides a toolbox to characterize point patterns. The package greatly simplifies the task of obtaining distance-based agglomeration and coagglomeration indexes, such as the Duranton and Overman’s Kd function and the Marcon and Puech’s M functions.