Measuring Economic Growth from Outer Space

Here is a clever new idea from Henderson, Storeygard, and Weil:

GDP growth is often measured poorly for countries and rarely measured at all for cities. We propose a readily available proxy: satellite data on lights at night. Our statistical framework uses light growth to supplement existing income growth measures. The framework is applied to countries with the lowest quality income data, resulting in estimates of growth that differ substantially from established estimates. We then consider a longstanding debate: do increases in local agricultural productivity increase city incomes? For African cities, we find that exogenous agricultural productivity shocks (high rainfall years) have substantial effects on local urban economic activity.

Here is the paper. WSJ blogs added:

They also noted how data from night lights can be focused to provide
data on a local level. In Southern Madagascar large deposits of rubies
and sapphires were discovered in late 1998 near the towns of Ilakaka
and Sakaraha, leading to an economic boom. But the data from the
satellites tell the story of where the benefits were felt most deeply.
“Over the next five years there was a sharp growth in the number of
pixels for which light is visible at all, and in the intensity of light
per pixel,” the economists said. “The other town visible in the figure,
Ihosy, shows no such growth. If anything, Ihosy’s light gets smaller
and weaker, as it suffers in the competition across local cities for


This measurement is going to be highly dependent on the statistical constancy between the lighting intensity and whatever variables are implied from the measurements. If there is an advantage to be gained from the appearance of growth, and this measurement becomes a standard, then we should expect to see a significant demand for battery-powered floodlights pointing upwards.

Regards, Don

I always wondered by North Korea doesn't install a few huge spotlights in strategic places, just to prevent those embarrassing satellite images.

If this gets NSF or NASA grant money I'm gonna be super pissed!

Paul Romer uses light in much the same way in his "Theory of History". He makes some very interesting points and does a great job of developing the framework for his charter city, a pet project of his that he recently resigned from Stanford to pursue. On his blog he has posted the light diagrams that he used for his research. You can find them here, You can find his lecture entitled "Theory of History" here, Well worth the 50 min in my opinion.


As we switch to LED based signal lights and other higher efficiency methods, we might see a drop in lighting, or maybe a drop in the first time derivative of the increase in lighting as technology become more efficient.

OR... we get much, much more efficient at putting light only where we need it.

Linking GDP and light from space is a very clever idea, although there are some problems with the connection. Here is a great article about why the link may not be as strong as one might think:

Comments for this post are closed