About the project

Whether poorer regions are catching up with richer ones and how the fruits of development are

distributed across different regions, groups or ethnicities is of paramount concern in development

economics and macroeconomics. Empirical research in the field of regional inequality and convergence was so far constrained by the partial availability of regional economic accounts. In rich countries, subnational data are usually limited to fixed administrative regions, which depend on political boundaries. In developing

countries, the data constraints are an even more serious problem: sub-national accounts are often

unavailable and their national counterparts unreliable. Localized spatial analysis of nighttime light emissions can help to solve these problems. This DFG project will showcase the potential of this type of data for research on regional inequality and convergence, and contribute to the development of best-practice standards in this emerging field. While the project is formally split into three parts and three locations, the three teams collaborate intensely on all research themes.

Part 1: Measuring regional and group inequalities with lights (Bluhm, Hannover)

The first part is concerned with measuring the relationship between nighttime lights, income and inequality. Apart from mixing up various definitions, the existing literature ignores sensor saturation in the satellite images. This leads to top-coding of bright spots such as cities and distorts estimates of urban-rural differences. We present novel approaches to deal with this issue and construct a saturation-corrected data set of luminosity around the world from 1992 to 2013. We then use this data to calculate regional inequality in lights in different countries and compare it to measures of ethnic inequality and historic differences between ethnic groups. 

Part 2: Convergence, uneven growth, and growth clubs (Krause, Hamburg)

In the second part, we analyze changes in lights over time to determine if poorer regions are catching up with richer ones. We exploit the flexibility of the luminosity data to look at different levels of aggregation beyond administrative boundaries, analyzing convergence of regions around the world, including changes in the so-called ‘world distribution of lights’. We develop a new notion of club convergence which allows us to identify regional growth clusters.

Part 3: Root causes of regional inequalities and development (Lessmann, Braunschweig)

The third part shifts the focus to why some regions and groups are poorer than others, and why the spread of this distribution is larger in some countries than in others. Building on the results of the previous two parts, we take a closer look at the geographic roots of regional income differences and try to disentangle them from man-made factors (such as political institutions). We propose a new ‘geography hypothesis’ which focuses on geographic heterogeneity within countries rather than geography itself. We investigate how geography affects spatial inequality and thereby potentially comparative development, and how these effects are moderated by political institutions.