Abstract
Recent literature has argued that decentralization can create new forms of leverage for criminal gangs, increasing their territorial presence and levels of violence. In this article we reverse the causal arrow and analyze how gangs affect the performance of decentralized institutions. We study the case of El Salvador, a country with the ubiquitous presence of gangs. We find that the higher presence of gangs reduced municipalities’ fiscal revenues, thereby increasing their dependence on transfers from the central government. This result is mostly driven by mid-size municipalities, as the effect is not significant for small and large municipalities. In addition to depressing revenues, gang presence was also associated with lower municipal spending and less service provision, effectively hollowing out municipal governance. The drop in municipal spending is particularly strong in capital spending and in small municipalities. The article also finds that gangs undermine economic activity – not just in the municipalities where they operate but through spill-over effects in neighboring jurisdictions as well.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 For helpful comments on this article, we would like to thank Axel Radics and Alejandro Rasteletti of the Fiscal and Municipal Management Division of the Inter-American Development Bank. This publication does not constitute an endorsement by the InterAmerican Development Bank of the opinions expressed in it. Data and code available upon request at [email protected].
2 For a comprehensive review of the relationship between decentralization and development, see Faguet Citation2014.
3 For the argument that regional inequality in developing countries can also impede factor mobility, see Rodríguez-Pose and Ezcurra Citation2010.
4 The municipal code assigns a number of other key roles to municipalities, including trash collection, municipal policing, the construction of housing, the regulation of local transport, commercial activities and restaurants, and the promotion of industrial development, education, and health (Vargas, Citation2022).
5 In terms of local expenditure, the main functions of municipalities are: solid waste removal, public safety in local markets, parks and local streets, construction and maintenance of cemeteries, fostering local economic development, and supporting community housing services.
6 Panel data could not be used, given that the spatial autocorrelation models in any given period only function with a balanced panel, and the availability of data is limited for the variables considered.
7 Nighttime lighting was obtained from Google Earth Engine and the Administrative Map from OCHA Latin America and the Caribbean. This variable is constructed using satellite recognition of faint levels of brightness on the Earth’s surface during the night, which reveals a city’s level of nighttime luminosity. This variable differs from the measurement of energy consumption, which is an indicator of economic activity (which takes place mostly during the daytime). It is worth remembering, however, that both variables may be correlated to the existence of urban centers.
8 The effect of gangs in public service in other size municipalities is marginally significant with a level of confidence less than 10%.
9 It is important to consider that there are significant variations in electricity consumption among localitites. In aggregate terms, the average electricity consumption per department varies, with some departments reaching as high as 120 million kWh, such as the case of San Salvador (or 972 kWh per capita), while other departments like Morazan record a consumption of 2.5 million kWh (or 256 kWh per capita).
10 Decree 762. June 14, 2023. https://www.asamblea.gob.sv/sites/default/files/documents/decretos/4194112C-1F6E-4E24-808E-9854A3D081AD.pdf.