Are night-time lights a good proxy of economic activity in rural areas in middle and low-income countries? Examining the empirical evidence from Colombia

Abstract

The use of satellite imagery, particularly night-time lights, has flourished in the last 20 years for socioeconomic studies. The intensity of the lights captured through remote sensing is frequently used as a proxy of different socioeconomic indicators. While some studies found a high correlation between night-time lights intensity and Gross Domestic Product, there has been an inconclusive debate about the validity of this assumption that night-time lights can serve as a good proxy for the economic development to sub-national level studies, particularly in rural areas of middle and lowincome countries. We test the suitability of night-time lights from publicly available data sources for estimating Regional Domestic Product (RDP) across municipalities with different degrees of urbanization in Colombia. We use a series of cross-sectional regression models to compare correlation between municipality RDP and luminosity from different sources for 2012 (DMSP, VIIRS, harmonized DMSP/VIIRS, and harmonized DMSP/VIIRS masked with Global Urban Footprint), as well as multilevel regression models to estimate RDP time-series from 2011 to 2018. Our findings reveal that all compared night-time light products can serve as a good indicator of municipal RDP patterns, while VIIRS data presents the best model fit. Harmonized data that have extensive temporal night-time light records from 2011 to 2018 were significantly correlated with RDP time-series. For seven population size levels – from big cities (>500,000 inhabitants) to rural areas (<5,000 inhabitants), the results present a comparatively higher model fit for urban than rural areas. The use of Global Urban Footprint further improved model fit for large cities but worsened rural RDP estimates. Therefore, our analysis underscores that the use of night-time lights can be a very valuable method to estimate patterns of socioeconomic change at the municipal level in medium and low-income countries and thus may help to implement better sustainability-oriented policies.