Usefulness of the article: Public policies may be biased in favor of groups close to political decision-makers, for example, their ethnic groups, regions of origin, or constituencies that share their political affiliations. This article discusses recent work by economists and political scientists on this subject. It shows that this phenomenon can be observed everywhere, particularly in France, and that political competition seems to limit it.
Abstract:
- Political favoritism is defined as politicians disproportionately allocating resources in favor of certain groups, particularly based on their ethnicity, party affiliation, or geographic location.
- Political favoritism is not limited to developing countries. It can be observed everywhere, particularly in France.
- Political competition seems to limit political favoritism.
Between 1970 and 1978, nearly twice as much was spent in Kenya on building roads in districts with a Kikuyu majority than in those with a Kalenjin majority. One explanation for this difference, highlighted by Burgess et al. (2015), can be summed up in a few words: Kenyatta (1894-1978), the country’s leader, belonged to the former of these two ethnic groups. Whether out of preference or political calculation (in order to remain in power), he would have favored his own group.
This example is a textbook case of political favoritism. This phenomenon is defined as the disproportionate allocation of resources by political decision-makers[1] in favor of certain groups, particularly on the basis of their ethnicity, party affiliation, or geographical location[2]. It is of great interest to political scientists and economists because of its electoral consequences and because it creates bias and injustice in public policy.
The previous example may give the impression that political favoritism primarily concerns developing countries. This impression is false; in fact, it is a quasi-universal phenomenon that does not spare developed countries. The aim of this article is to present some of the research that highlights the importance of political favoritism, particularly in France, and then to discuss the impact of political competition. The latter seems to limit political favoritism.
Political favoritism: a universal phenomenon
Political favoritism is not unique to developing countries. De Luca et al. (2018) demonstrated this in an article with the evocative title: Ethnic favoritism: An axiom of politics? They sought to measure, in the largest possible sample (139 countries divided into more than 2,000 ethnic regions [3] observed between 1992 and 2012), whether the regions of origin of political leaders were favored and benefited from greater economic stimulus than neighboring and similar regions. In order to construct a homogeneous variable measuring economic activity, De Luca and his co-authors used an original methodology. Using satellite images from the US Air Force[4], they used the nighttime light intensity of regions, a variable strongly correlated with economic activity. They then show that the regions from which a country’s political leaders originate are associated with greater light intensity at night. Furthermore, an important finding of this article is that the « rich world » (particularly Western Europe), although often associated with lower ethnic favoritism, is not immune to this phenomenon either. To address the issue of reverse causality—political leaders could disproportionately come from the most dynamic and therefore most illuminated regions—De Luca et al. (2018) show that the regions from which future political leaders will come are not associated with higher light intensity. For example, if a political leader is elected in a given year (say 2010), their region of origin does not appear to experience a difference in light intensity prior to that year (2010).
Previous studies have examined the ethnic dimension of political favoritism. However, this is not the only factor, and public policies may also be biased in favor of certain regions, whether these are electoral districts or the regions of origin of politicians. Curto-Grau et al. (2018) show that in Spain, regions redistribute more resources to municipalities whose mayor shares the political affiliation of the region’s leader. This result is also verified in a sample of cities where municipal elections were extremely competitive, which ensures that it is not the result of unobserved differences between cities. Similarly, in Italy, Carozzi and Repetto (2016) show that the cities of birth of members of the Italian Parliament tend to receive higher per capita funds (transfers from the state to municipalities) (from 2.1 to 2.8%)[6].
France is not immune to this phenomenon. In a working paper, Fabre and Sangnier (2017) studied the investment subsidies received by French municipalities as decided by the government. They show that municipalities where a local elected official (often the mayor) becomes a minister benefit from an increase of nearly 45% in these subsidies after the appointment[7]. However, unlike the article cited above on Italian municipalities, Fabre and Sangnier (2017) do not detect any bias in favor of the ministers’ municipalities of birth (only those where they were local elected officials). The authors then document several findings that help to interpret this increase in subsidies. In particular, the authors show that the effect persists after the politician leaves office. They suggest that part of the explanatory mechanism could be due to the accumulation of knowledge and the creation of a network within the various administrations. Once the municipality has benefited from more subsidies, it knows how to maintain this advantage.
How can it be limited? Political competition as a remedy
Political favoritism is a complex phenomenon that responds to incentives from policy makers. Political competition seems to limit political favoritism. Institutions that organize or promote this competition are generally associated with lower levels of political favoritism. In the above-mentioned article on road construction in Kenya, Burgess et al. (2015) show that biases in public policy in favor of the leader’s ethnic group are only observable during periods when Kenya was under autocratic regimes. Similarly, while the study by De Luca et al. (2018) suggests that ethnic favoritism is a universal phenomenon, its importance seems to be linked to the quality of institutions.
Within the same institutional framework (within a country, for example), variations in political competition can also limit political favoritism. For example, Carozzi and Repetto (2016) show that they no longer detect favoritism toward politicians’ hometowns in the years leading up to elections. In Spain, Curto-Grau et al. (2018) also show the influence of political competition. When regional elections are more competitive, political favoritism towards co-partisan municipalities is greatly reduced. The explanation would be as follows: regional politicians would first aim for re-election and secondly to increase their influence within allied municipalities. It is therefore only when the first objective is assured that they skew public policy in favor of allied municipalities. Although political competition is probably only one « remedy » among many[8], it appears to be effective in preventing, or at least limiting, political favoritism.
In conclusion: the incentives for elected political decision-makers
There therefore seems to be a « universal » tendency among elected officials and political decision-makers to favor « groups » that are close to them. These groups may be ethnic, geographical (the region of origin of certain politicians, or their current constituency), or related to party affiliation.
However, this « universal » tendency to favor « one’s own » is largely affected by electoral incentives. In particular, institutions that promote electoral competition seem to limit the scope of this phenomenon. Furthermore, within the same institutional framework, greater competition is associated with less political favoritism. However, more research is needed to better understand the links between political favoritism and political competition and, more generally, the impact of political competition on politicians’ incentives. While the latter generally seems to be associated with less biased public policies [9], it may also have negative aspects. Heidhues and Lagerlöf (2003), for example, point out that political competition can lead politicians to conceal more information from their constituents.
References
Burgess, Robin, Jedwab, Remi, Miguel, Edward, Morjaria, Ameet, Padro i Miquel, Gérard, (2015). The Value of Democracy: Evidence from road building in Kenya, American Economic Review, Vol. 105 – 6, pp. 1817-1851.
De Luca, Giacomo, Hodler, Roland, Raschky, Paul A., and Valsecchi, Michele, (2018). Ethnic Favoritism: An Axiom of Politics? Journal of Development Economics, Volume 132, May, Pages 115-129.
Curto-Grau, Marta, Solé-Ollé, Albert, and Sorribas-Navarro, Pilar (2018). Does Electoral Competition Curb Party Favoritism? American Economic Journal: Applied Economics, Vol. 10, No. 4, October, pp. 378-407.
Fabre, Brice and Sangnier Marc (2017), What Motivates French Pork: Political Career Concerns or Private Connections, AMSE working papers.
Carozzi, Felipe and Repetto Luca (2016). Sending the pork home: Birth town bias in transfers to Italian municipalities, Journal of Public Economics, Vol 134, pp. 42-52.
Heidhues, Paul and Lagerlöf, Johan, (2003). Hiding information in electoral competition. Games and Economic Behavior, Volume 42, Issue 1, January, Pages 48-74.
Solé-Ollé, Albert and Viladecans-Marsal, Elisabet. (2012). Lobbying, political competition, and local land supply: Recent evidence from Spain.Journal of Public Economics, Volume 96, Issues 1–2, February, Pages 10-19.
Svaleryd, Helena & Vlachos, Jonas, 2009. Political rents in a non-corrupt democracy, Journal of Public Economics, Elsevier, vol. 93(3-4), pages 355-372, April.
[1] Articles on this subject focus mainly on those in executive power (ministers, presidents, autocrats, etc.) or legislative power (members of parliament, etc.). In practice, the phenomenon could concern anyone capable of influencing public policy.
[2] This definition is largely based on that proposed by Golden and Min (2013), who provide a summary of the literature on the subject.
[3] In their analyses, countries/regions are sometimes grouped into large geographical areas: East Asia and the Pacific, Central Asia and Eastern Europe, Latin America and the Caribbean, South Asia, Sub-Saharan Africa, and finally Western Europe and North America. De Luca et al. (2018) include many « rich » countries in their analyses, including France, the United Kingdom, Japan, the United States, etc.
[4] The database on light intensity was compiled by the National Oceanic and Atmospheric Administration ( NOAA).
[5] Curto-Grau et al. (2018) use a regression discontinuity methodology to ensure the causality of their results.
[6] This result is obtained using « fixed effects » and control variables by municipality, i.e., by controlling for the characteristics of these municipalities.
[7] The authors use a difference-in-differences method in three different samples (municipalities with more than 3,500 inhabitants, those with more than 10,000 inhabitants, and a sample obtained after matching municipalities) in order to make the results as robust as possible.
[8] Political favoritism is greatly facilitated if policies can be changed at will. Using « technical » criteria (e.g., based on population, poverty level, etc.) to allocate funds may be another way to reduce this phenomenon. The role of the press and, more generally, of information available to citizens is also widely discussed in political economy and could limit political favoritism. The effect of citizen information on politicians’ incentives is discussed, for example, in Svaleryd and Vlachos (2009).
[9] Political favoritism is not the only phenomenon concerned. Solé-Ollé and Viladecans-Marsal (2012), for example, suggest that political competition can limit the influence of interest groups in the development of public policy.
