Estimating Economic Characteristics with Phone Data

Abstract

Historically, economists have relied heavily on survey-based data collection to measure social and economic well-being. Here, we investigate the extent to which the “digital footprints” of an individual can be used to infer his or her socioeconomic characteristics. Using two different datasets from Afghanistan and Rwanda, we show that phone data can be used to estimate the wealth of individuals in two very different economic environments. However, we find that such models are relatively brittle, and that a model trained in one country cannot be used to estimate characteristics in another. These results suggest several promising applications and directions for future work.

Approaches and Alternatives to the Wealth Index to Measure Socioeconomic Status Using Survey Data: A Critical Interpretive Synthesis

“…Lastly, there are several research teams attempting to proxy wealth indices using new technologies. One team has developed a machine learning algorithm that can be used to roughly approximate wealth indices using phone usage characteristics in Rwanda and Afghanistan, although the models must be developed separately for each country and cannot be applied naively across borders (Blumenstock 2018). Another team has created a convolutional neural network trained on ground imagery that is able to predict 37-55% of variation in consumption and 55-75% of variation in asset wealth if trained separately for each country, although this drops to 19-52% and 24-71% if applied to other countries (Jean et al 2016).…”

Connecting Isolated Communities

“…We also include site fixed effects, ν s , in all regressions. 10 We find that CCN adoption was correlated with household wealth. Controlling for other household characteristics and site fixed effects, we find that a one standard deviation increase in the wealth index is correlated with a 3 percentage point increase in network adoption.…”

“…Upon launching the CCNs, we took steps to enable linking the rich socioeconomic data from the baseline survey to Call Detail Records (CDR), which allow us to describe phone-based communication on the community cellular network. Together, this unique data allow us to describe information access before the introduction of cellular networks as well as unpack the baseline household characteristics that correlate with the early adoption of the cellular networks [9][10][11]13].…”

Key trends in development co-operation for national data and statistical systems

“…Support through other channels is important only for some DAC members: Australia, for instance, channelled about one-third of its support between 2014 and 2018 through the Data for Health Initiative, a donor country-based non-governmental organisation founded by Bloomberg Philanthropies and the Australian government (Bloomberg Philanthropies, n.d. [66]). According to OECD data, the United States between 2016 and 2018 channelled more than USD 130 million (in 2018 prices) through domestic private sector entities, including more than USD 110 million for private companies acting as implementers of USAID's MEASURE DHS (Demographic and Health Surveys) project (USAID, n.d. [67]).…”