## Satellite images at night and economic growth

## Source details

**Canonical URL:** [Satellite images at night and economic growth](https://www.imf.org/-/media/files/publications/fandd/article/2019/september/satellite-images-at-night-and-economic-growth-yao.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/fandd/article/2019/september/satellite-images-at-night-and-economic-growth-yao.pdf.md)
- [Structured JSON version](/-/media/files/publications/fandd/article/2019/september/satellite-images-at-night-and-economic-growth-yao.pdf.json)

---

### Overview and methodology
- Satellite "night lights" images have been used to study human activity and natural events for almost 30 years; economists have used them extensively in the past decade.
- Each pixel in a satellite night-light image represents an area of less than one square kilometer and is associated with a digital number measuring brightness at night.
- Aggregating pixel brightness over a country produces an indicator of that country's nighttime activity; comparisons across countries and over time serve as a barometer of economic development and fluctuations.

### How night lights reflect economic change
- Night lights capture two broad dimensions of economic change:
  - Spatial expansion or contraction: growing economies show more areas lit over time as more pixels start to record light; conflict-affected regions show more areas going dark as pixels lose light.
  - Intensification: urbanization, agglomeration, and infrastructure modernization increase brightness per pixel as the same night sky brightens.
- Empirical relationships:
  - Rich countries are generally brighter than less developed countries, though notable exceptions exist.
  - On a per capita basis, Nordic countries have almost always been the brightest spots on Earth.
  - Japan, despite being a rich country, appears scarcely brighter than Syria did before the Arab Spring, likely due to energy conservation habits and high population density.
  - In advanced economies, night lights grow only about half as fast as GDP.

### Usefulness where official statistics are weak
- Night lights are particularly valuable in countries afflicted by conflict where official economic statistics may be scarce or unreliable.
- Night-light–based GDP estimates can differ from official series:
  - They often indicate faster economic deterioration during conflict than official data.
  - They often suggest a stronger bounce-back after conflict ends.
  - The informal economy is a plausible factor behind stronger postconflict rebounds captured by night lights.

### Big data potential and analytic advances
- Viewing each pixel as a data point yields enormous datasets: an individual country like the United States can consist of hundreds of millions of data points; with more than 200 countries and regions, the Earth's land surface contains almost a billion data points for one early coarse-resolution satellite image.
- Technological trends:
  - More frequent satellite releases and ever-finer resolution will cause the number of data points to grow exponentially.
  - Advances in data science, including machine learning, enable analysis of patterns for decision-making; private firms (e.g., DigitalGlobe and Orbital Insight) already apply these techniques.
- Research and policy uses:
  - High-granularity data can study local effects, spatial spillovers, and economic activity in remote areas where satellite data are the primary reliable source.
  - Night lights are one among many emerging geospatial and text-based data sources that can provide new perspectives on economic questions.

### Key implications and policy recommendations
- Night lights offer a complementary and sometimes superior measure for tracking economic activity, especially where official data are missing or unreliable.
- Policymakers and researchers should:
  - Harness night-light data to reestimate and validate GDP and other economic indicators in data-poor settings.
  - Use finer-resolution and higher-frequency satellite data combined with machine learning to inform local and national policy decisions.
  - Recognize the different signaling of night lights across development stages: strong infrastructure-driven lighting growth in early development and weaker light growth relative to GDP in advanced economies.
- As the age of big data unfolds, seize opportunities to integrate satellite and other novel data sources to improve economic measurement, guide smarter policies, and enhance understanding of development dynamics.

*Article by Jiaxiong Yao; draws on IMF Working Paper 19/77 by Yingyao Hu and Jiaxiong Yao.*

---


_Source: https://www.imf.org/-/media/files/publications/fandd/article/2019/september/satellite-images-at-night-and-economic-growth-yao.pdf_