Interactive Feature

Mapping China's Global Investments and Inequality

Geocoded Chinese Development Projects and the Diffusion of Economic Activity

Working Paper

Connective Financing: Chinese Infrastructure Projects and the Diffusion of Economic Activity in Developing Countries



Richard Bluhm, Axel Dreher, Andreas Fuchs, Bradley Parks, Austin Strange, Michael Tierney

Geocoded China Data

AidData's Geocoded Global Chinese Official Finance, Version 1.1.1



This dataset geolocates Chinese Government-financed projects that were implemented between 2000-2014. It captures 3,485 projects worth $273.6 billion in total official financing. The dataset includes both Chinese aid and non-concessional official financing.

China's Investments in Connective Infrastructure

Argentina’s Hidden Railroads

Can you spot the cities in the image on the left? It's difficult to visually discern patterns of economic activity in this daytime satellite image of central Argentina from 2016. Man-made features like urban areas are difficult to pick out from swathes of forest and grassland.

Argentina after dark is a different picture altogether. Bright dots of towns and cities are strung like glowing beads along the invisible thread of Argentina's national railway system. The night landscape appears overlayed by a grid of evenly spaced towns and cities that established themselves as commerce flowed along and out from the railway.  

Rather than concentrating economic output in a small number of cities, connective infrastructure projects like Argentina’s national railway serve a vital role to disperse economic activity to outlying areas, connecting rural households and firms to distant markets for goods, jobs and services. China, a major supplier of infrastructure finance and connective infrastructure, has invested billions in Argentina's railway system since 2010.

NASA Earth Observatory images by Joshua Stevens, using Suomi NPP VIIRS data from Miguel Román, NASA GSFC, and MODIS data from the Land Processes Distributed Active Archive Center (LP DAAC) and Google Earth Engine. Map by Borah Kim for AidData.

The Geographic Dispersion of Nighttime Light Intensity

West Africa’s Shifting Nighttime Coast

How do patterns of nighttime light change over time on West Africa’s coast? Toggling the slider reveals differences between 2012 (left) and 2016 (right) that may be subtle from space but significant on the ground. Some cities have become brighter while others have grown dim; increased economic activity along roads manifests as glowing lines, like those that connect Ghana's coast to the interior; and clusters of fishing vessels swarm near different ports.

By analyzing high-resolution satellite images like these of nighttime light intensity, researchers are able to estimate economic output at the local level. This measure is especially useful in remote, poor, and conflict-prone areas that lack reliable census, survey, and administrative data.  

But since economic progress is often uneven, it’s not just the amount of nighttime light within a geographic region that matters. The geographical dispersion of nighttime light also tells us something about how widely or narrowly the benefits of economic growth and development are being shared. 

A growing body of research demonstrates this kind of spatial inequality can have far-reaching consequences. It can increase political polarization, slow economic development, provoke social unrest, and elevate the risk of violent conflict and terrorism.

Photos by NASA’s Earth Observatory. Map by Soren Patterson for AidData.

Measuring Economic Inequality with Nighttime Lights

Spatial Inequality in West Africa

How equally or unequally is economic activity distributed within countries? Using annual data on the geographical dispersion of nighttime light, the Connective Financing report calculates a Gini coefficient to consistently measure economic inequality from year to year in more than 32,000 localities around the world.  

Each locality receives a Gini score that varies from 0 to 1. A Gini of 1 represents perfect economic inequality: a province or district in which a single 1 km x 1 km square of land contains all the economic activity in that administrative area. A Gini of 0 represents the opposite: a province of perfect economic equality where every single 1 km x 1 km grid cell in the province has the same level of economic activity. Compare the 2012 Gini scores for ADM1 districts in West Africa (left) to the satellite image of those distric's nighttime lights in 2012 (right). Districts in yellow and green contain more equal dispersions of economic activity, while districts in orange and red are less equal in how economic activity is distributed.

By observing how economic inequality changes over time, and by combining these measures with new data on the locations of Chinese Government-financed projects worldwide, the Connective Financing report evaluates how Chinese investments influence the geographic distribution of economic activity within provinces and districts in low- and middle-income countries.

Photo by NASA’s Earth Observatory. Map by Soren Patterson for AidData.

About the Report

Using the most comprehensive dataset of Chinese development project locations ever assembled, a new AidData Working Paper shows that Chinese infrastructure investments narrow economic inequalities within poor countries. Published by a team of economists and political scientists from William & Mary, Leibniz University Hannover, Heidelberg University, Helmut Schmidt University Hamburg, and Harvard University, Connective Financing leverages a new geolocated dataset of Chinese Government-financed projects worldwide to evaluate how these investments alter the geographic distribution of economic activity within provinces and districts in low- and middle-income countries. 

About the Dataset

After releasing the first-ever global dataset of Chinese development projects in October of 2017, AidData embarked on a far-reaching effort to assign precise geographic coordinates to thousands of projects. The result of thousands of hours of geocoding by dozens of research assistants is a first-of-its-kind geocoded global dataset that pinpoints 3,485 Chinese development projects worth USD $273.6 billion implemented in 6,190 locations in 138 countries over a fifteen-year period (2000-2014). This dataset only includes projects that were both recommended for research in the original dataset and projects whose status was completed or implemented.