We apply deep learning to daytime satellite imagery to predict changes in income and population at high spatial resolution in US data. For grid cells with lateral dimensions of 1.2 km and 2.4 km (where the average US county has dimension of 51.9 km), our model predictions achieve R2 values of 0.85 to 0.91 in levels, which far exceed the accuracy of existing models, and 0.32 to 0.46 in decadal changes, which have no counterpart in the literature and are 3–4 times larger than for commonly used nighttime lights. Our network has wide application for analyzing localized shocks.
Each year in the US, hundreds of billions of dollars are spent on transportation infrastructure and billions of hours are lost in traffic. We develop a quantitative general equilibrium spatial framework featuring endogenous transportation costs and traffic congestion and apply it to evaluate the welfare impact of transportation infrastructure improvements. Our approach yields analytical expressions for transportation costs between any two locations, the traffic along each link of the transportation network, and the equilibrium distribution of economic activity across the economy, each as a function of the underlying quality of infrastructure and the strength of traffic congestion. We characterize the properties of such an equilibrium and show how the framework can be combined with traffic data to evaluate the impact of improving any segment of the infrastructure network. Applying our framework to both the US highway network and the Seattle road network, we find highly variable returns to investment across different links in the respective transportation networks, highlighting the importance of well-targeted infrastructure investment.
The use of air quality monitoring networks to inform urban policies is critical especially where urban populations are exposed to unprecedented levels of air pollution. High costs, however, limit city governments’ ability to deploy reference grade air quality monitors at scale; for instance, only 33 reference grade monitors are available for the entire territory of Delhi, India, spanning 1500 sq km with 15 million residents. In this paper, we describe a high-precision spatio-temporal prediction model that can be used to derive fine-grained pollution maps. We utilize two years of data from a low-cost monitoring network of 28 custom-designed low-cost portable air quality sensors covering a dense region of Delhi. The model uses a combination of message-passing recurrent neural networks combined with conventional spatio-temporal geostatistics models to achieve high predictive accuracy in the face of high data variability and intermittent data availability from low-cost sensors (due to sensor faults, network, and power issues). Using data from reference grade monitors for validation, our spatio-temporal pollution model can make predictions within 1-hour time-windows at 9.4, 10.5, and 9.6% Mean Absolute Percentage Error (MAPE) over our low-cost monitors, reference grade monitors, and the combined monitoring network respectively. These accurate fine-grained pollution sensing maps provide a way forward to build citizen-driven low-cost monitoring systems that detect hazardous urban air quality at fine-grained granularities.
We propose a methodology for defining urban markets based on builtup landcover classified from daytime satellite imagery. Compared to markets defined using minimum thresholds for nighttime light intensity, daytime imagery identify an order of magnitude more markets, capture more of India’s urban population, are more realistically jagged in shape, and reveal more variation in the spatial distribution of economic activity. We conclude that daytime satellite data are a promising source for the study of urban forms.
Occasional widely publicized controversies have led to the perception that growth statistics from developing countries are not to be trusted. Based on the comparison of several data sources and analysis of novel IMF audit data, we find no support for the view that growth is on average measured less accurately or manipulated more in developing than in developed countries. While developing countries face many challenges in measuring growth, so do higher-income countries, especially those with complex and sometimes rapidly changing economic structures. However, we find consistently higher dispersion of growth estimates from developing countries, lending support to the view that classical measurement error is more problematic in poorer countries and that a few outliers may have had a disproportionate effect on (mis)measurement perceptions. We identify several measurement challenges that are specific to poorer countries, namely limited statistical capacity, the use of outdated data and methods, the large share of the agricultural sector, the informal economy, and limited price data. We show that growth measurement based on the System of National Accounts (SNA) can be improved if supplemented with information from other data sources (for example, satellite-based data on vegetation yields) that address some of the limitations of SNA.