VoxEU Column Politics and economics

Regional favouritism across the world

Political leaders sometimes favour their preferred regions. This column looks at regional favouritism in a large sample of countries, using information on the birthplaces of political leaders and nighttime light intensity. Being the current leader's birth region increases nighttime light intensity by around 4%, and GDP by around 1%. Such favouritism is most prevalent in countries with weak political institutions and poorly educated citizens.

Many political leaders favour their preferred regions. An extreme example is Zaire’s former dictator Mobutu. In his remote ancestral home town Gbadolite, he built a huge palace complex costing $100 million, luxury guesthouses, an airport capable of handling Concords, and the country's best supply of water, electricity, and medical services. But Mobutu is no exception. There is a large literature on distributive politics documenting regional favouritism. Golden and Min (2013) review the literature on redistributive politics based on an inventory of more than 150 empirical studies. They notice that most studies focus on a single democratic country, and a single policy outcome.1

Regional favouritism and nighttime light

In a recent article (Hodler and Raschky 2014), we complement this literature on distributive politics by taking a systematic look at regional favouritism in a large and diverse sample of countries that includes democracies as well as autocracies, and by employing a broad measure of regional favouritism that captures the aggregate distributive effect of many different policies. In particular, we use information about the birthplaces of political leaders and satellite data on nighttime light intensity to study whether subnational administrative regions have more intense nighttime light when being the birth region of the current political leader.

Our analysis is based on a panel dataset with 38,427 subnational regions in 126 countries, and annual observations from 1992 to 2009. The dependent variable is the logarithm of average nighttime light intensity, which is recorded by US Air Force Weather Satellites and provided by the National Oceanic and Atmospheric Administration. Henderson et al. (2012) document a strong relationship between nighttime light intensity and GDP at the country level, and propose the use of nighttime light intensity as a measure of economic activity at the subnational level. Using regional GDP data by Gennaioli et al. (2013), we find a similarly strong relationship between regional nighttime light intensity and regional GDP. Our main explanatory variable is a dummy that equals one for the birth region of each country's current political leader, and zero for all other regions.

We find this leader region dummy variable to be positively associated with nighttime light intensity although we include region fixed effects to control for time-invariant regional characteristics, and country-year dummy variables to control, in the most flexible way, for changes over time in individual countries. We argue that the political leaders are the reason why leader regions have more intense nighttime light, and that our findings provide evidence for regional favouritism. To address the potential endogeneity of leader regions, we look at regions that will shortly become leader regions, or have been leader regions until recently.

  • Our results suggest that being the leader region increases nighttime light intensity by around 4%, and GDP by around 1% on average.
Determinants and dynamics of regional favouritism

The dynamics of regional favouritism are such that political leaders need a few years before they successfully engage in regional favouritism, and then become increasingly better at supporting their birth region. The increase in nighttime light intensity accelerates after political leaders have been in power for around 12 years. This acceleration may be no coincidence as most democratic countries have constitutions that allow heads of states a maximum of two terms of four to seven years in office, while such rules are often absent or remain unenforced in autocratic countries. We further find that the effects of regional favouritism do not outlast the political leaders. Hence, regional favouritism typically does not lead to sustainable development in the birth regions of the political leaders. We use different units of observation, i.e., subnational regions based on alternative regional boundaries, to explore the geographical extent of regional favouritism. This exploration offers a somewhat less daunting picture: While some benefits of regional favouritism are fairly local, a considerable part of the benefits flow to relatively large geographical areas and, thereby, presumably many people.

Our large and diverse sample allows us to look at potential determinants of regional favouritism.

  • Better political institutions may reduce regional favouritism by constraining the political leaders.
  • More education may reduce regional favouritism as educated citizens are more likely to participate in the political process and to hold political leaders accountable.

Using Polity2 scores and years of schooling as proxies for political institutions and education, we find that both better political institutions and more education reduce regional favouritism. Our results suggest that being the leader region increases nighttime light intensity and regional GDP by around 30 and 9%, respectively, in countries with weak political institutions, and by around 11 and 3%, respectively, in countries with the lowest level of school attainment in our sample. Regional favouritism is also more prevalent in poorer countries, and countries where the political leaders may be more attached to their birth region because of linguistic heterogeneity or strong family ties. Once we include all these potential determinants of regional favouritism jointly, we find that regional favouritism is most prevalent in countries with weak political institutions and poorly educated citizens.

In the last part of our study, we use our approach to examine the distributive effects of foreign aid and oil rents. We, thereby, complement the literature on the effects of foreign aid and oil rents on governance and corruption, which typically relies on perception-based country-wide indices, by using an observable measure of economic activity in subnational regions across the globe. We find that aid and oil tend to fuel rent seeking and regional favouritism in weakly institutionalised countries, but not in countries with comparatively better political institutions.

Policy implications

These findings have various policy implications.

  • First, we confirm that sound political institutions and good education are both helpful to keep political leaders accountable, and we demonstrate their importance in constraining regional favouritism. In particular, the enforcement of term limits seems to be crucial.
  • Second, our findings suggest that donor agencies need to be very cautious when supporting countries with authoritarian leaders because such leaders mainly use foreign aid to the benefit of themselves, their family and clan members, and others living in their stronghold.
  • Third, and more generally, many interesting and policy-relevant questions on drivers and barriers to regional economic development could not be addressed until recently because regional data on economic activity was missing or of poor quality.

Our study illustrates that the use of a large panel data set on nighttime light intensity in subnational administrative regions from all over the world allows addressing such questions. In addition, academic researchers and donor agencies could also consider employing nighttime light data to evaluate large scale policy interventions.


Franck, Raphael, and Ilia Rainer (2012), “Does the Leader's Ethnicity Matter? Ethnic Favouritism, Education and Health in Sub-Saharan Africa.” American Political Science Review 106, 294-325.

Gennaioli, Nicola, Rafael La Porta, Florencio Lopez-de-Silanes, and Andrei Shleifer (2013), “Growth in Regions.” NBER Working Paper 18937.

Golden, Miriam, and Brian Min (2013), “Distributive Politics Around the World.” Annual Review of Political Science 16, 73-99.

Henderson, Vernon J, Adam Storeygard, and David N Weil (2012), “Measuring Economic Growth from Outer Space.” American Economic Review 102, 994-1028.

Hodler, Roland, and Paul A. Raschky (2014), “Regional Favouritism.” Quarterly Journal of Economics 129, 995-1033.

Kramon, Eric, and Daniel N. Posner (2013), “Who Benefits from Distributive Politics? How the Outcome One Studies Affects the Answer One Gets.” Perspectives on Politics 11, 461-474.

1 Notable exceptions include Franck and Rainer (2012) and Kramon and Posner (2013) who study favouritism in a sample of various Sub-Saharan African countries.

2 We are obviously not the first highlighting potential drawbacks of providing foreign aid to autocratic states, but we are arguably the first backing up these warnings with evidence based on observable measures rather than perception-based indices.

945 Reads