Endnotes
Sam Asher

Sam Asher is associate professor of economics and public policy at Imperial College London and co-founder of Development Data Lab, a research and policy organization that works to mobilize 21st century data sources and make them available to researchers, policymakers, civil society, and the private sector in India and beyond. His research seeks to understand the drivers of growth and mobility in lower-income countries, using high resolution data from unconventional sources, such as satellites and government administrative data, to test for the role of place in shaping people’s economic opportunities. Asher was previously an assistant professor of international economics at the Johns Hopkins School of Advanced International Studies and an economist at the World Bank Development Research Group.

Aditi Bhowmick

Aditi Bhowmick is the India director at Development Data Lab, where she is leading outreach and partnerships with stakeholders across India and building a gender-focused research agenda for South Asia. Bhowmick holds a master’s degree in public administration from the School of Public and International Affairs at Princeton University and has spent several years conducting field research across urban and rural India with the Abdul Latif Jameel Poverty Action Lab.

Paul Novosad

Paul Novosad is associate professor of economics at Dartmouth College and co-founder of Development Data Lab. Some of his recent projects have focused on using machine learning and new statistical methods to study bias in the judicial system; the impacts of India’s large-scale rural roads program; the impacts of mineral sector development; and changes in upward mobility in India across time, space, and social groups. Novosad started his career as a software developer, followed by many years of living in sub-Saharan Africa and Southeast Asia, building clinic software for HIV/AIDS management in Lesotho and working on agricultural development policy in Vietnam.

Sam Asher, Aditi Bhowmick, and Paul Novosad of Development Data Lab write that the current golden age of data analytics promises to improve the effectiveness of every government, every program, every investment to broaden opportunity and super-charge social mobility.

While floating hundreds of miles above the earth’s surface, India’s first astronaut Rakesh Sharma was asked what India looked like from space. In response, he quoted Muhammad Iqbal’s poem The Anthem of India: "… better than the whole world ..."

Today, the view from space isn’t merely inspiration for poets. It is providing policymakers, social investors, researchers, and civil society leaders with a powerful tool to better the whole world.

Satellite imagery, unprecedented access to government data, and advances in machine learning are fueling a revolution in governance and philanthropy — helping us identify the roots of poverty and the seeds of upward mobility. These new tools promise to shed light on one of the greatest mysteries in economic development: Why, in an era of unprecedented global prosperity, are so many people still stuck at the bottom?

Global social mobility experts use colorful language to describe this phenomenon of low social mobility: “sticky floors” and “sticky ceilings.” These phrases mean that a person’s ability to climb the economic ladder is too often determined not so much through their own hard work and skill but by their parentage and place of birth. In other words, those born poor usually stay poor — with their potential unrealized. 

Sticky floors and sticky ceilings are common in many of the most unequal societies in the world, from Brazil to South Africa, and they are increasingly common in the U.S. and Europe as well. When the ladder out of poverty is missing a few rungs, it is a tragedy for the individual, a sorrow for parents, and a catastrophe for society.

Poor social mobility is also terrible for the economy because it leaves so many nascent inventors, entrepreneurs, and change makers from humble backgrounds without an opportunity to develop their talents. When social mobility is absent, Thomas Edison remains a poor newsboy and the world never benefits from his 1,000+ patents; and the great Indian mathematician Srinivasa Ramanujan spends his life working as a clerk in a sari shop like his father, and entire areas of mathematics go undiscovered.

Our work gives us reason to worry about India’s potential Ramanujans. Social mobility in India is among the lowest in the world, and by some measures, it has not improved in the last 40 years. India’s economy is growing, but children who are born into low-income families have extraordinarily thin chances of making it to the middle class. 

Low social mobility anywhere is a problem everywhere. By failing to nurture the brilliance in the slums of Lagos, Nigeria, the villages in Bolivia, or the farms of Louisiana, U.S., we are going without the transformative discoveries of minds like Edison and Ramanujan. 

Global social mobility experts use colorful language to describe this phenomenon of low social mobility: “sticky floors” and “sticky ceilings.” These phrases mean that a person’s ability to climb the economic ladder is too often determined not so much through their own hard work and skill but by their parentage and place of birth. In other words, those born poor usually stay poor — with their potential unrealized. 


Flying Blind

Until recently, policymakers in India and elsewhere have struggled to identify the root causes of low social mobility and to pass effective policies that broaden access to opportunity. 

A big part of the problem is that decision-makers have been simply flying blind. According to the World Bank, half of low-income countries have not conducted a census in the last 10 years. Even in India, normally a leader when it comes to data collection, the last validated national poverty survey was conducted in 2012. Essential government decisions — like where to place a new health center — are being made with only the coarsest information on where services are most needed.

An information revolution — both in data creation and in our ability to learn from it — is poised to change this.  

As government activities are increasingly electronic, they leave a valuable digital footprint with information about every interaction with every citizen. Researchers have begun to harness this data and combine it with other records such as satellite images, mobile phone records, and business data, to understand local needs, and more finely guide policies and programs to increase impact. For example, researchers can use roof material and cropping patterns detected from satellite imagery to identify communities or even individual families who need assistance. With careful analysis and cutting-edge tools, these new data sources can provide real-time information that can transform how we understand our world and revolutionize policymaking. 

Early work in this space is already delivering results in some of the most challenging environments: Togo, Colombia, and the Democratic Republic of Congo used satellite imaging, mobile phone data, and machine learning to identify the poorest of the poor and connect them with emergency financial assistance during the COVID-19 pandemic. Ghana is using machine learning and satellite data to target subsidies for clean water access; Kenya is using similar tools to map inequities in access to health care facilities. 

Machine learning and satellite data are used to target subsidies for clean water access, Ghana, 2017.

We are only scratching the surface of the possible. Sophisticated use of the data already at our fingertips promises to make every government and philanthropic program more effective. 


Location, Location, Location

When we use these cutting-edge data sources to study social mobility in high resolution, one of the things we learn is that neighborhoods matter a great deal, in both high-income and low-income countries. Children growing up in similar households, perhaps separated only by a mile or a railroad track, experience very different life trajectories based largely on the characteristics of the places that they grow up. 

This confirms what many of us recognize intuitively: The barriers people face are often highly local. Children generally attend the school in their community — not one in the next village or across town. As a result, a poorly performing school casts a large shadow over a generation of children in a particular area. Likewise, sick people and pregnant women generally seek care at the closest health center and may forgo health care altogether if there is no facility nearby, and workers can only access jobs within a small commuting radius. High resolution data lets us identify precisely the neighborhoods that are being left behind by poor access to essential public services like schools, clinics, and basic sanitation. Often, these are slums or shantytowns crowded with low-income families and minorities.

The data also confirms another intuition: Cities are engines of opportunity. But some cities are better at delivering on migrants’ dreams than others. In too many places, a tangle of governance and market failures results in cities where millions of people live in slums or are unable to move to a city at all. For example, in Delhi — a city of nearly 20 million — construction regulations bar buildings over four stories tall. These restrictions result in dire conditions visible to the naked eye — migrants living in segregated neighborhoods with ramshackle housing and no clean water or sanitation — and also visible in our data. This provides a road map for where change is most needed.

The Delhi Mobility Map presents the geographical distribution of upward mobility across neighborhoods in Delhi. The measure of upward mobility describes the expected national education rank of a son born to a father in the bottom half of the parent education distribution. Source: Asher, Novosad, and Rafian (2022).

These poor outcomes are avoidable. By zooming in and studying the neighborhoods and cities that succeed — like Ramanujan’s hometown of Erode, which is still an island of high social mobility 140 years after his birth — we can generate evidence on the government policies that accelerate social mobility and implement those policies in the places being left behind. 


Data Is the New Oil

We are in a golden age of data analytics. Governments and businesses are generating more data than ever before, but few policymakers have the tools and resources needed to leverage these vast untapped riches. 

Organizations around the world, including the Massachusetts Institute of Technology’s Abdul Latif Jameel Poverty Action Lab, Opportunity Insights at Harvard University, University of California Berkeley’s Center for Effective Global Action, the eGov Foundation in India, and our own organization, Development Data Lab, among others, are working to fortify governments’ abilities to learn from big new data sources. This is a space ripe for engagement and support from both social investors and government leaders.

There is no longer any reason to fly blind. By making better use of the information that is readily available, we can ensure that all the future Ramanujans, regardless of where they are born, can achieve their potential, with all the world better for their talents.

Endnotes
Sam Asher

Sam Asher is associate professor of economics and public policy at Imperial College London and co-founder of Development Data Lab, a research and policy organization that works to mobilize 21st century data sources and make them available to researchers, policymakers, civil society, and the private sector in India and beyond. His research seeks to understand the drivers of growth and mobility in lower-income countries, using high resolution data from unconventional sources, such as satellites and government administrative data, to test for the role of place in shaping people’s economic opportunities. Asher was previously an assistant professor of international economics at the Johns Hopkins School of Advanced International Studies and an economist at the World Bank Development Research Group.

Aditi Bhowmick

Aditi Bhowmick is the India director at Development Data Lab, where she is leading outreach and partnerships with stakeholders across India and building a gender-focused research agenda for South Asia. Bhowmick holds a master’s degree in public administration from the School of Public and International Affairs at Princeton University and has spent several years conducting field research across urban and rural India with the Abdul Latif Jameel Poverty Action Lab.

Paul Novosad

Paul Novosad is associate professor of economics at Dartmouth College and co-founder of Development Data Lab. Some of his recent projects have focused on using machine learning and new statistical methods to study bias in the judicial system; the impacts of India’s large-scale rural roads program; the impacts of mineral sector development; and changes in upward mobility in India across time, space, and social groups. Novosad started his career as a software developer, followed by many years of living in sub-Saharan Africa and Southeast Asia, building clinic software for HIV/AIDS management in Lesotho and working on agricultural development policy in Vietnam.

Stay in Touch

Subscribe to receive our future updates.

* Indicates required information.

Thank you!

We will keep you updated.

Something went wrong while submitting the form.
Please double-check your details and try again.