Socioeconomic Status and Mobility during the COVID-19 Pandemic: An Analysis of Eight Large Latin American Cities

Abstract

This study analyzes mobility patterns during the COVID-19 pandemic for eight large Latin American cities. Indicators of mobility by socioeconomic status (SES) are generated by combining georeferenced mobile phone information with granular census data. Before the pandemic, a strong positive association between SES and mobility is documented. With the arrival of the pandemic, in most cases, a negative association between mobility and SES emerges. This new pattern is explained by a notably stronger reduction in mobility by high-SES individuals. A comparison of mobility for SES decile 1 vs decile 10 shows that, on average, the reduction is 75% larger in the case of decile 10. According to estimated lasso models, an indicator of government restrictions provides a parsimonious description of these heterogeneous responses. These estimations point to noticeable similarities in the patterns observed across cities. We also explore how the median distance traveled changed for individuals that travel at least 1 km (the intensive margin). We find that the reduction in mobility in this indicator was larger for high-SES individuals compared to low-SES individuals in six out of eight cities analyzed. The evidence is consistent with asymmetries in the feasibility of working from home and in the ability to smooth consumption under temporary income shocks.

“Does a respiratory virus have an ecological niche, and if so, can it be mapped?” Yes and yes

“…Of course, the interpretation of these factors is not as intuitive as mobility and we certainly do not wish to attribute direct causality to them. However, there are a variety of relevant factors for COVID-19 that can be related to internet access and computer usage for example, such as age, educational status, population density [21], as urban areas have better infrastructure, and mobility itself [22, 23]. As with any epidemiological or ecological model, the interpretation of the predictors as representing direct versus indirect interactions is highly non-trivial.…”

“…) and P (C | X n j = 1) for each value of the habitat variable ranges X m i =Average annual temperature and X n j =Internal labor flow of the municipality corresponding to a "climate-only" model and a "mobility-only" model. related to internet access and computer usage for example, such as age, educational status, population density [21], as urban areas have better infrastructure, and mobility itself [22,23]. As with any epidemiological or ecological model, the interpretation of the predictors as representing direct versus indirect interactions is highly non-trivial.…”

Nonlinear effects of mobility on COVID-19 in the US: targeted lockdowns based on income and poverty

“…A strand of the literature has shown evidence for unequal mobility changes across demographic and socioeconomic groups amid COVID-19, which may be effective on the spread of COVID-19. Among others Aromí et al. (2020), have shown that higher socioeconomic status is associated with more intense reductions in mobility based on data from 8 large Latin American urban areas.…”

<p>#StayAtHome: Social&nbsp;Distancing Policies and&nbsp;Mobility in Latin America&nbsp;and the Caribbean</p>

“…However, given the information we have available, it is not possible to identify these characteristics for each user, and, therefore, we cannot estimate the heterogeneous effects of lockdowns for different groups. Still, a recent paper by Aromí et al (2021) shed light on this issue. The study explored the differential change in mobility by socioeconomic status for eight large Latin American cities during the beginning of the pandemic.…”