In recent years, a quiet revolution has been taking place in the field of economics, driven by a new type of data – information collected by Earth-observing satellites. Orbiting hundreds of miles above the planet’s surface, these sophisticated sensors are providing economists with unprecedented insights into human activity and natural phenomena around the globe. By harnessing the power of satellite imagery and related technologies, researchers are shedding new light on issues ranging from economic growth and inequality to agricultural productivity and environmental change.
The origins of remote sensing stretch back over a century, but it is only in the past decade or so that this technology has truly taken off as a tool for economic analysis. Several key developments have fueled this trend: a dramatic increase in the volume and accessibility of satellite data; rapid advances in computing power and machine learning techniques to extract meaningful information from images; and a growing recognition among economists of the unique advantages that remote sensing offers.
Seeing the Previously Unseen
One of the most important benefits of satellite data is that they allow researchers to observe things that would be difficult or impossible to measure through other means. Many economic activities and outcomes – from factory output and agricultural yields to urban expansion and deforestation – can now be tracked from space at a fraction of the cost of traditional ground-based surveys. Moreover, because satellites collect data continuously and systematically, they can provide consistent measurements across different countries and time periods, overcoming the limitations of patchy or unreliable official statistics.
A prime example of this application comes from Indonesia, where economists have used satellite imagery to study the political economy of deforestation. In a country where illegal logging is rampant but rarely reported, researchers were able to directly observe forest loss by analyzing changes in tree cover over time. The results were striking: each new administrative district created led to an 8% increase in deforestation, suggesting that local officials were colluding to allow unsustainable harvesting. Without the objectivity of satellite data, such findings would have been nearly impossible to document.
Satellite sensors can also shed light on otherwise invisible phenomena, from air pollution to economic inequality. One innovative study combined daily satellite measurements of airborne particulates with health records to estimate the impact of forest fire smoke on infant mortality in Indonesia. Another used night-time luminosity data as a proxy for local economic activity, revealing stark disparities in growth between ethnic homelands in Africa. As the resolution and variety of satellite-based measurements continues to expand, so too will the range of questions that economists can tackle.
A Bird’s-Eye View
Beyond access to novel types of information, another key advantage of remote sensing is its ability to provide data at an extremely high spatial resolution. Whereas traditional economic datasets are often aggregated to the level of countries, provinces, or counties, satellite imagery can distinguish features as small as individual buildings or plots of land. This fine-grained perspective opens up new possibilities for analyzing the spatial dimension of economic activity.
For example, researchers have used high-resolution satellite data to study the lasting impact of historical events on the geography of economic development. One study focused on portage sites in the American South – river locations where boats had to be unloaded and cargo transported overland before the advent of steamships. By digitizing satellite images, the authors showed that these early centers of commerce went on to become major cities like Washington D.C. and Richmond, demonstrating the power of path dependence at a local scale that would be invisible in standard datasets.
The benefits of spatial resolution are not confined to the study of history. In urban slums in Nairobi, a novel machine learning approach has been used to map improvements in housing over time, based on subtle changes in roof reflectivity that can be detected from space. At an even finer scale, some companies are using satellite imagery to monitor the number of cars in retail parking lots, providing a real-time indicator of consumer spending. As the level of detail in remote sensing data continues to increase, the scope for such micro-level applications will only grow.
A Global Perspective
A third key strength of satellite-based data is their broad geographic coverage. Unlike most economic datasets which are confined to particular countries or regions, Earth-observing satellites collect data that span the entire globe. This comprehensive viewpoint allows economists to study issues that transcend national borders, from the spatial distribution of economic activity to the impacts of climate change.
In the realm of agriculture, for instance, satellite imagery has been used to map crop yields and growing conditions at a planetary scale. By combining these data with climate model projections, researchers have been able to predict the potential impacts of rising temperatures on global food production and international trade. The results suggest that in the absence of vigorous adaptation efforts, climate change could reduce worldwide agricultural output by a staggering one-sixth by the end of the century.
The global nature of satellite data also permits economists to study connectivity and interaction between distant locations. Ship-tracking data has been used to construct a comprehensive map of international trade routes, while night-light imagery has been used to estimate the extent of cross-border economic integration in places like Central Asia. More recently, some researchers have even used high-resolution imagery to study the diffusion of agricultural technologies from China to sub-Saharan Africa by identifying the distinctive circular irrigation patterns associated with Chinese investment. The potential for such “telecoupling” analysis will only increase as the coverage and cadence of satellite observations continues to grow.
Challenges on the Horizon
For all the excitement surrounding remote sensing in economics, there are still significant challenges to overcome. One is the difficulty of extracting meaningful economic information from images that are often highly unstructured and complex. While identifying a building or measuring a field’s greenness is relatively straightforward, more subtle concepts like product quality or management practices cannot be directly observed from space. Overcoming this limitation will require continued advances in machine learning and artificial intelligence, as well as creative data integration approaches that combine satellite imagery with ground-based surveys and other conventional data sources.
Another challenge is the need for caution in interpreting and validating results based on remote sensing data. Like any novel data source, satellite imagery is subject to measurement error, sampling bias, and other potential pitfalls. Economists will need to carefully consider issues of precision, representativeness, and external validity, just as they do with more traditional datasets. At the same time, the use of remote sensing is likely to raise new ethical concerns around privacy and surveillance, as the line between public and private space becomes increasingly blurred.
The Road Ahead
Despite these challenges, the future of remote sensing in economics looks bright. As satellite data become ever more abundant and accessible, and as analytical techniques continue to advance, the range of potential applications will only expand. One promising area is the use of remote sensing to study economic activity in data-poor environments, from conflict zones and failed states to the world’s most remote rural areas. Another is the integration of satellite data with other novel data streams, from mobile phone records to social media posts, to gain a more comprehensive understanding of human behavior and well-being.
Looking further ahead, it is not hard to imagine a future in which real-time satellite data becomes a standard tool for economic monitoring and policy evaluation. Government agencies could use high-frequency imagery to track the effectiveness of infrastructure investments or the impact of natural disasters. Businesses could use remote sensing to optimize supply chains and distribution networks. And researchers could use the data to study the economic impacts of everything from technological change to geopolitical events, all at a global scale and with unprecedented detail.
Of course, realizing this vision will require more than just advances in data and technology. It will also require a new generation of economists who are comfortable working with large, unstructured datasets and who are willing to collaborate across disciplinary boundaries. But if the recent history of remote sensing in economics is any guide, there is reason for optimism. By embracing the view from above, economists have the opportunity to gain new perspectives on some of the most pressing challenges of our time, from sustainable development and climate change to the future of work and the fate of the global economy. The satellite revolution may have only just begun, but its impact on the field of economics is already being felt, and the best is yet to come.
