Cong Peng

/Economics & Big Data/


My research features in using real-time data to solve pressing problems in economics, by connecting advances in data science with econometrics. I have been working with data generated from the Internet of Things (IoT) platforms, user-generated content, open source GIS data, and high-resolution satellite data. I am interested in applying state-of-art causal inference methods to draw inference from Big Data.

Curriculum Vitae

Does E-Commerce Reduce Traffic Congestion? Evidence from Alibaba Single Day Shopping Event
Job Market Paper

Traditional retail involves traffic both from warehouses to stores and from consumers to stores. E-commerce cuts intermediate traffic by delivering goods directly from the warehouses to the consumers. Although plenty of evidence has shown that vans that are servicing e-commerce are a growing contributor to traffic and congestion, consumers are also making less shopping trips using vehicles. This poses the question of whether e-commerce reduces traffic congestion. The paper exploits the exogenous shock of an influential online shopping retail discount event in China (similar to Cyber Monday), to investigate how the rapid growth of e-commerce affects urban traffic congestion.

Colonial Legacies: Shaping African cities
with J. Vernon Henderson and Neeraj G. Baruah

Differential institutions imposed under colonial rule continue to affect the spatial structure of African cities and day-to-day life. Based on a sample of 318 cities across 28 countries with satellite data on built cover over time, Anglophone origin cities compared to Francophone ones are more sprawling and have less regular spatial layouts overall. They especially have more leapfrog development at the extensive margin.

Does paving roads improve quality of life? Examining the Zambian context using AI
with Wenfan Chen

Given the scale of road investment and Zambia's mounting fiscal deficit, this study assesses the socioeconomic impact of improvements in road quality between the years 2009 and 2013. Using Michelin maps and Open Street Map as a basis, the study constructs a digitalised road network of Zambia. With a number of available documented road upgrade projects and high resolution satellite images, we trained a Convolutional Neural Network model to predict paved roads at the national level. The model allows tracking the changes of road conditions over time, and therefore travel time and market access of settlements. Using two batches of household data from the Demographic and Health Surveys in the same period, the study finds evidence for enhanced socioeconomic outcomes for households arising from improvements in market access driven by better road quality.

Valuing the Environmental Benefits of Canals Using House Prices
with Steve Gibbons and Cheng Keat Tang
commissioned by Canal & River Trust of the UK

The canal and waterway network in Britain provides a potentially valuable recreational and environmental amenity. We value access to this amenity using a property value approach. We improve on standard methods by controlling for micro-geographic fixed effects for small neighbourhoods and by applying a difference-in-differences method to analyze the effect of the restoration of canals in the late 2000s.



79 JFK St
Cambridge, MA 02138
United States.