Measurement errors in popular night lights data may bias estimated impacts of economic sanctions: Evidence from closing the Kaesong Industrial Zone

Abstract

Satellite-detected night lights data are widely used to evaluate economic impacts of sanctions. Such data should be free from political manipulation. However, measurement errors in these data, from blurring and bottom-coding, are rarely considered. To study such errors, we use a difference-in-differences analysis of impacts of closing the Kaesong Industrial Zone in North Korea—a sanction South Korea imposed in 2016. Luminosity in the affected region declined by a precisely estimated 50 percent. When using the popular Defense Meteorological Satellite Program (DMSP) night lights data the apparent impacts are imprecisely estimated and far smaller. Measurement errors in DMSP data may distort evaluations of sanctions.