Economic Measurement

We introduce a methodology to detect and measure economic activity using geospatial data and apply it to steel production, a major industrial pollution source worldwide. Combining plant output data with geospatial data, such as ambient air pollutants, nighttime lights, and temperature, we train machine learning models to predict plant locations and output. We identify about 88% (94%) of plants missing from the training sample within a 1 km (5 km) radius and achieve R-squared above 0.8 for output prediction at a 1 km grid and at the plant level,  as well as both regional and time series validations. We use the model trained on Chinese steel data to reliably detect the location and measure the total annual output of steel plants in South Korea and Iran. Our approach can be adapted to other industries and regions, and used by policymakers and researchers to track and measure industrial activity in near real time. 

Livestock are difficult to measure consistently in areas with sparse agricultural statistics. This paper examines the potential and limitations of using satellite-derived methane measurements to assess livestock activity.  Combining global 0.1° methane observations with gridded livestock data from 2003 to 2020, we find that nearby areas with more livestock have higher methane concentrations. The relationship survives controls for competing methane sources and does not appear for pollutants that livestock do not emit. In the United States, satellite methane tracks concentrated dairy cattle but not dispersed beef cattle, and California cells with large dairy facilities show higher methane concentrations. Satellite methane is therefore a useful tool for screening and independently verifying large, concentrated operations, and filling facility-level monitoring gaps in countries where official statistics are sparse or outdated.