Economists have long faced challenges in collecting comprehensive, reliable data to power their models and analyses. Traditional data sources like government statistics and ground surveys often have gaps – both spatial and temporal. Many developing countries lack extensive reporting infrastructures, while conducting consistent on-the-ground data collection across countries is arduous and expensive.
Enter satellites – our eyes in the sky. With their global reach and regular image capture, satellites are filling critical data gaps to give economists new capabilities. This article explores the ways satellite data is advancing economic modeling:
Accessing Hard-to-Obtain Data
Satellites collect diverse geospatial information difficult to gather otherwise. For example, satellite imagery enables tracking infrastructure like roads, buildings, ports, and energy facilities. This allows models to quantify infrastructure’s economic impact and role in development. High-resolution imagery even captures indicators like number of cars in parking lots or ships docked at ports to proxy economic activity.
Satellites also survey agriculture – crop types, planted area, yield predictions – on a global scale. This data feeds models assessing productivity dynamics and food security. Studies have leveraged satellite land cover classification to estimate informal economic activity not captured in official statistics.
Without satellites, such data would require exhaustive ground efforts. Satellites provide a consistent data stream, opening new areas of economic inquiry.
Granular Geospatial Analysis
The spatial resolution of satellite imagery far surpasses traditional data sources. Landsat offers 30 meter resolution, while newer satellites go down to 1 meter. This enables hyperlocal economic analysis.
Researchers can observe outcomes literally neighborhood by neighborhood. For example, one study combined satellite imagery with economic census data to estimate income in 1 km grid cells across five countries. Such granular modeling facilitates analysis of economic questions involving distance and geography – from estimating tax policy spillovers across jurisdictions to predicting effects of new infrastructure on surrounding regions.
Global Coverage and Consistency
Satellites capture standardized imagery spanning countries, regions and the whole planet. This facilitates comparative economic analysis across geographic areas that is difficult with ground surveys.
For example, satellite data enabled estimating a global GDP prediction model, with the results showing revised relative income levels for different countries. Other research combined night lights satellite imagery with survey data to predict poverty levels across countries.
Such globally consistent data provides the foundation for building extensive cross-country socioeconomic databases for macroeconomic modeling. This is a boon for development economics questions involving cross-country comparisons.
Challenges and Promise
Of course, satellites are no panacea. Interpreting imagery requires care – advanced methods combining computer vision and machine learning are expanding capacities. Combining satellite data with traditional sources can enhance validity.
Satellites also capture extensive data requiring intensive computation. Accessing historical archives can be costly. New cloud computing and data storage solutions are expanding access.
While challenges remain, satellites’ global view from above offers transformative modeling capabilities. We are only beginning to tap data abundant enough to give economists a detailed, dynamic render of human society itself.
