Link to research statement

Poverty from Space: Using High Resolution Satellite Imagery for Estimating Economic Well-being and Geographic Targeting

Measuring poverty is important for targeting aid and formation of policy in developing economies. Poverty measurement is impeded by the high administrative and labor costs of large-scale surveys. This paper investigates the ability of high spatial resolution satellite images to accurately predict poverty and economic well-being. We extract both object and texture features from satellite images of Sri Lanka. These data are then used to train models of local area poverty and economic well-being. The important features include the number and density of buildings, shadow area (a proxy for building height), number of cars, density and width of roads, type of farmland, and roof material. These variables are used to estimate poverty rates and average log consumption for 1,287 Gram Niladhari (GN) Divisions. Predictions from a baseline binomial logit model, using only these satellite features as explanatory variables, explain sixty percent of poverty and sixty-five percent of average log consumption at the GN Division level. We control for overfitting by using Lasso regularization. Our policy simulations find that these poverty estimates perform as well as official estimates for geographic targeting. In contrast to a popular, low-cost alternative measure of poverty, night time lights, our measures are two to eight times as efficient for geographic targeting, as night time lights is a poor measure of poverty in sparsely populated areas. We conclude that the use of satellite data has the potential to revolutionize poverty measurement, reducing survey costs and making geographically targeted programs more effective.

[latest draft] [slides]

[Press: BrookingsBloomberg, Atlantic City Lab, Fast Company, Borgen Magazine]

Unintended Consequences of the African Growth and Opportunity Act: The Role of Trade Diversion and Structural Change (with Klaus-Peter Hellwig)


This paper investigates the effects of preferential trade programs such as the U.S. African Growth and Opportunity Act (AGOA) on the direction of African countries’ exports. While these programs intend to promote African exports, textbook models of trade suggest that such asymmetric tariff reductions could divert African exports from other destinations to the tariff reducing economy. We examine the import patterns of 177 countries and estimate the diversion effect using a triple-difference estimation strategy, which exploits time variation in the product and country coverage of AGOA. We find no evidence of systematic trade diversion within Africa, whereas diversion from other industrialized destinations to the US was significant, in particular for apparel products. At the same time, we show that, more than diverting trade, AGOA had positive spillovers on the product composition of trade, which suggests that the product coverage of preferential trade agreements can influence structural change in Africa.

[latest draft available by request] [slides]

Building a better model: Variable selection to predict poverty in Pakistan and Sri Lanka (with Marium Afzal and David Newhouse)


This paper uses out-of-sample validation techniques to evaluate alternative prediction models of household poverty. Using household data from Pakistan and Sri Lanka, we compare the model accuracy using manual selection, stepwise regression, and Lasso-based procedures, and also examine how much incorporating publically available satellite data improves model accuracy. The main findings are that: 1) Lasso outperforms both discretionary and stepwise models in Pakistan, where the set of potential predictors is large. 2) Lasso and stepwise models give comparable results in Sri Lanka, with its fewer predictors. 3) Model accuracy model depends considerably on the poverty threshold. 4) Including satellite data improves poverty predictions in Sri Lanka, where predictors are scarce, but not in Pakistan. 5) Including satellite data increases the benefit of using Lasso in Sri Lanka. We conclude that among the three model selection methods considered, Lasso-based models are preferred for generating poverty predictions, especially with a rich pool of candidate variables, and incorporating publicly available satellite data can considerably improve the accuracy of regional poverty predictions when the pool of candidate variables from household surveys is smaller.

[latest draft] [slides]

“Robust Determinants of Trade Flows,” (with Marianne Baxter), May 2015.


What are the policies and country-level conditions which best explain bilateral trade flows between countries? As databases expand, an increasing number of possible explanatory variables are proposed that influence bilateral trade without a clear indication of which variables are robustly important across contexts, time periods, and which are not sensitive to inclusion of other control variables. To shed light on this problem, we apply three model selection methods – Lasso reguarlized regression, Bayesian Model Averaging, and Extreme Bound Analysis — to candidate variables in a gravity models of trade. Using a panel of 198 countries covering the years 1970 to 2000, we find model selection methods suggest many fewer variables are robust that those suggested by the null hypothesis rejection methodology from ordinary least squares.

[latest draft]

Historical Health Conditions in Major U.S. Cities,” (with Carlos Villarreal, Brian Bettenhausen, Eric Hanss). Historical Methods: A Journal Of Quantitative And Interdisciplinary History Vol. 47 , Iss. 2,2014.

The Historical Urban Ecological data set is a new resource detailing health and environmental conditions within seven major U.S. cities during the study period from 1830 to 1930. Researchers collected and digitized ward-level data from annual reports of municipal departments that detail the epidemiological, economic, and demographic conditions within each city. They then drafted new geographic information system data to link the tabular records to ward geographies. These data provide a new foundation to revisit questions surrounding the urban mortality transition and the growth of U.S. cities.

[latest draft]

“Sweet Diversity: Overseas Trade and Gains from Variety after 1492,” (with Hans-Joachim Voth), 2011.


When did overseas trade start to matter for living standards? Traditional real-wage indices suggest that living standards in Europe stagnated before 1800. In this paper, we argue that welfare rose substantially, but surreptitiously, because of an influx of new goods as a result of overseas trade. Colonial luxuries such as tea, coffee, and sugar transformed European diets after the discovery of America and the rounding of the Cape of Good Hope. These goods became household items in many countries by the end of the 18th century. We use three different methods to calculate welfare gains based on price data and the rate of adoption of these new colonial goods. Our results suggest that by 1800, the average Englishman would have been willing to forego 10% or more of his income in order to maintain access to sugar and tea alone. These findings are robust to a wide range of alternative assumptions, data series, and valuation methods.

[latest draft]

Works in Progress

“Firm Entry and Export Performance when Firm Credit is Constrained; Evidence from Ethiopia and the African Growth and Opportunity Act”