Although several previous studies have advanced the knowledge of how violence perpetrated by DTOs affects electoral outcomes, the study of how levels of criminal violence vary during local electoral contests remains scant. Stated differently, we know little on whether the local electoral cycle has an effect on the level of criminal violence. Employing the CIDE-PPD Database, we find that local elections do have an effect on levels of DTOs violence and that the greatest incentives to upscale violence occur shortly before election day. These fluctuations suggest that DTOs are actively seeking to influence local governance in their favor especially during the campaigns. Our analysis also suggests that candidates in local Mexican elections face a more precarious and dangerous situation compared to recently-elected authorities.
This is a preview of subscription content, access via your institution.
Buy single article
Instant access to the full article PDF.
Tax calculation will be finalised during checkout.
Subscribe to journal
Immediate online access to all issues from 2019. Subscription will auto renew annually.
Tax calculation will be finalised during checkout.
Electoral competition is defined by the number of electoral alternatives in electoral competitions (Sartori 1976).
Electoral competitiveness grows as the electoral support for all available electoral alternatives becomes more similar (Sartori 1976). In other words, this happens when the distribution of electoral support becomes more egalitarian.
By electoral cycles, we consider the period that starts before the registration of candidates and ends after the day of election.
By captured local governments, we imply that a relatively stable DTO has influence over the actions of local authorities or candidates. Local officials support DTOs’ demands in exchange for safety, protection, and other possible benefits.
Moral hazard is the risk that a party entered into a contract in good faith or that provided misleading information.
Since 2018, it is possible reelection of mayors in Mexico.
Online Appendix 1 presents anecdotal examples of moves made by DTOs during local election campaigns, including killings and threats to candidates and key authorities.
We acknowledge that there are other potential non-violent means to capture local governments such as providing illegal financial resources to local authorities. DTOs might attempt to use these strategies before employing violence. However, we focus on the occurrence of violent actions during the election period once DTOs have exhausted other means.
Unfortunately, it is not possible to distinguish which actor (DTO or public security force) started the fight in which homicides occurred, based on the information provided by the database.
The CIDE-PPD Database classify events according to three categories: 1) Executions: “any intentional homicide in which the victim or the perpetrator is an alleged member of a criminal organization. It is not the result of a prior confrontation or an aggression. Neither does it include the participation of Mexican authorities”; 2) confrontations: “…events in which public forces use firearms (against criminals), or those events involving clashes among or within specific criminal groups”; and 3) aggressions: “…as an attack by criminal organizations against government institutions, public officers, or any other institution related to the government to which authorities are unable to respond” (Atuesta et al. 2019).
Despite these advantages, we point out that the CIDE-PPD Database presents two disadvantages as the authors of the database -- Atuesta et al. (2019), 1778) -- acknowledge: 1) differences in the collection of the information, probably caused by different agencies gathering the data, and 2) spatial-temporal differences in information collection. There are no ways to correct these differences as we cannot identify their sources. However, as we discuss below, we include several control variables in our multivariate models to account for the effect of temporal and spatial differences in our dataset.
The average number of homicides employed to calculate these deviations is the annual amount of homicides occurred in confrontations among DTOs for each municipality
The panel data sets combine measures of time (weeks or months) and units of analysis (municipalities) (Wooldridge 2012).
The Hausman test (1978) also confirms the convenience of employing this model instead of one with fixed effects.
The average period between registration day and election day for all municipalities is close to half a year (4.9 months)
Online Appendix 2 displays descriptive statistics of variables used in the analyses. To verify the robustness of our results, we replicate in Online Appendix 3 the models displayed in Table 1 using state fixed effects to control for the possible influence of unobserved state characteristics. Online Appendix 3 shows evidence that also supports the hypothesis of this study. While in the first specification the dependent variable is the deviation in the number of homicides in fights among DTOs, in the second specification the dependent variable is the deviation in the total number of homicides related to organized crime. Online Appendix 4 also replicates the model specifications displayed in Table 1, but this time we add a trend to control for the growing tendency in violent events. We also include the amount of municipal expenditures per capita as a proxy of municipal capabilities. Relatively strong capabilities might prevent from significant fluctuations in violence as local governments could be in better shape to reduce criminal interference in local politics. Moreover, we include a dichotomous variable that indicates whether partisan alternation occurred. Trejo and Ley (2018) propose that partisan alternation in Mexico broke the informal networks of agreements between state actors and criminal organizations. Such interruption weakened the protection that authorities offered to criminals. To evaluate whether changes in partisan control of municipalities alter violence in the short term (during the months following the election), we multiply the variable indicating whether there was partisan alternation by the dummy variable indicating the presence of a period post-election. If statistically significant, we could conclude that partisan alternation increases criminal violence in the short term. In all cases, our results confirm the validity of those reported by Table 1. Likewise, we do not find evidence indicating that our control variables are relevant for explaining short-term fluctuations in violence. This does not mean that partisan alternation does not produce an effect on violence; it just implies that this does not cause fluctuations in violence or does so immediately after the election day.
Finally, we also put together all variables used in the first and second model specifications in Online Appendix 5. Instead of using 3 months for the binary variables indicating the pre-registration period and the post-election period as Online Appendix 4 does, Online Appendix 5 employs a periodicity of 6 months for both variables. Overall, the results remain very similar to those of all previous tables and they confirm the validity of our hypothesis.
Although the aforementioned problems and drawbacks of the CIDE-PPD Database could create distortions, the findings reported by our empirical analyses are still statistically significant and robust. Besides, the high concentration of criminal violence in a few Mexican municipalities (in approximately 71% of them homicides related to criminal violence do not occur) increases our confidence in our results.
For instance, if only one DTO committed one or more homicides during a certain month, these homicides lead us to count only one DTO. We keep the same logic if more DTOs are involved in violent events during the analyzed period.
In the third specification of Online Appendix 3, we verify the robustness of this result by introducing state fixed effects in the model shown in Table 2. In the third specification of Online Appendix 4, we add a trend to control for the increasing tendency in the number of homicides during the studied period. We also add the partisan alternation variable and the interaction between the partisan alternation variable and the dummy variable indicating whether the month under study corresponds to a post electoral period with a periodicity of 3 months. In both cases, we confirm the validity of the results reported in Table 2. Likewise, in Online Appendix 5, we put together all control variables together with our key independent variables. In Online Appendix 5, we change the periodicity of the binary variables; this time the post-election variable and the pre-registration variable employ periods of 6 months. We also confirm previous results.
Unfortunately, the dataset does not allow us to identify the exact position of the politician in the local administration. This time we do not employ the deviations with respect to the annual average to construct this variable, since the mean closes to the value of zero.
The coefficient for attacks occurred between the registration day and the election day (during the campaign) is also positive, and it does reach the 10% level of statistical significance.
The fourth specifications in Online Appendices 3 and 4 also provide additional robustness checks for this finding. In both models, as in Table 2, the impact of the variable pre-registration period is statistically significant at the 1% level. While the model specification in Online Appendix 3 includes state fixed effects, the model specification in Online Appendix 4 controls for a trend, municipal expenditures per capita, and the variables indicating the occurrence of partisan alternation and the effect of partisan alternation in the short term. Both Online Appendices confirm Table 2’s findings. In Online Appendix 5, we employ a periodicity of 6 months for defining the short-term. Again, we find evidence confirming those results displayed by Table 2. It is important to acknowledge that the number of missing observations in the section of the CIDE-PPD Database for registering attacks to politicians might be significant. For instance, if a politician of a government agency committed an extrajudicial execution or the agency began a confrontation, such information is unlikely to be specified in the description of the event (Atuesta et al. 2019). Given this informational lacuna, specific events could not be recorded in the CIDE-PPD Database. Overall, when dealing with data related to drug violence, a higher standard of scrutiny (or higher statistical significance) might be necessary. Despite this potential drawback, the high concentration of these events in a few municipalities, increases our confidence in our findings.
We employ Stata 15 to perform this non-parametric analysis.
The results remain the same if we extend or reduce the six-month periodicity used to elaborate these graphs.
Acemoglu D, Robinson J, Santos R (2013) The monopoly of violence: evidence from Colombia. J Eur Econ Assoc 11:5–44
Albarracín J (2018) Criminalized electoral politics in Brazilian urban peripheries. Crime Law Soc Chang 69(4):553–575
Alesina A, Piccolo S, Pinotti P (2016) Organized crime, violence, and politics. https://www.nber.org/papers/w22093.pdf. Accessed 15 Feb 2017
Altman NS (1992) An introduction to kernel and nearest-neighbor nonparametric regression. Am Stat 46(3):175–185
Anaya-Muñoz A (2012) Security and human rights in the framework of Mexico’s “war on drugs”. In: Bow B, Santa-Cruz A (eds) The State and Security in Mexico. Transformation and crisis in regional perspective. Routledge, New York, pp 61–76
Astorga L (1999) Drug trafficking in Mexico: a first general assessment. Management Social Transformations MOST, working paper 36, UNESCO
Astorga L (2007) Seguridad, Traficantes y Militares: el Poder y la Sombra. Tusquets, Mexico City
Astorga L, Shirk D (2010) Drug trafficking organizations and counter-drug strategies in the U.S.-Mexican context. USMEX working paper 10–01, Center for U.S.-Mexican Studies, University of San Diego
Atuesta L, Pérez-Dávila S (2017) Fragmentation and cooperation: the evolution of organized crime in Mexico. Trends in Organized Crime 21(2):235–261
Atuesta L, Ponce A (2017) Meet the Narco: increased competition among criminal organisations and the explosion of violence in Mexico. Global Crime 18(4):375–402
Atuesta L, Siordia O, Madrazo Lajous A (2019) The “war on drugs” in Mexico: (official) database of events between December 2006 and November 2011. J Confl Resolut 63(7):1765–1789
Bagley B (2012) Drug trafficking and organized crime in the Americas: major trends in the twenty-first century. Woodrow Wilson Center Update on the Americas, August. Washington: Woodrow Wilson International Center for Scholars
Bailey J, Taylor M (2009) Evade, corrupt, or confront? Organized crime and the state in Brazil and Mexico. Journal of Politics in Latin America 1(2):3–29
Beittel J (2013) Mexico’s drug trafficking organizations: source and scope of the violence. Congressional Research Service Report R41576
Cabrero E, Arellano D (2011) Los gobiernos municipales a debate. Un análisis de la institución municipal a través de la encuesta INEGI 2009. CIDE, México
Calderón G, Robles G, Díaz-Cayeros A, Magaloni B (2015) The beheading of criminal organizations and the dynamics of violence in Mexico’s drug war. J Confl Resolut 59(8):1455–1485
Chabat J (2002) Mexico’s war on drugs: no margin for maneuver. Annals of the American Academy of Political and Social Science 582(Cross National Drug Policy):134–148
Chabat J (2010) La Respuesta del Gobierno de Calderón al Desafío del Narcotráfico: Entre lo Malo y lo Peor. CIDE Working Paper 196, Mexico City
Chacón M (2013) In the line of fire: political violence and decentralization in Colombia. Working paper. NYU-Abu Dhabi. Available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2386667
Collier P, Vicente P (2012) Violence, bribery, and fraud: the political economy of elections in sub-Saharan Africa. Public Choice 153(1):117–147
Dal Bó E, Di Tella R (2003) Capture by threat. J Polit Econ 111(5):1123–1154
Dal Bó E, Dal Bó P, Di Tella R (2006) Plata or Plomo? Bribe and punishment in a theory of political influence. Am Polit Sci Rev 100(1):41–53
Daniele, Gianmarco, and Benny Geys (2015) Organised Crime, Institutions and Political Quality: Empirical Evidence from Italian Municipalities Econ J, 125 (586):233–255
Daniele G, Dipoppa G (2017) Mafia, elections, and violence against politicians. J Public Econ 154:10–33
De Feo G, de Luca G (2017) Mafia in the ballot box. Am Econ J Econ Pol 9(3):134–167
Dell M (2015) Trafficking networks and the Mexican drug war. Am Econ Rev 105(106–6):1738–1779
Dercon S, Gutiérrez-Romero R (2012) Triggers and characteristics of the 2007 Kenyan electoral violence. World Dev 40(4):731–744
Dickenson M (2014) The impact of leadership removal on Mexican drug trafficking organizations. J Quant Criminol 30(4):651–676
Durán-Martínez A (2015) To kill and tell? State power, criminal competition, and drug violence. J Confl Resolut 59(8):1377–1402
Durán-Martínez A (2018) The politics of drug violence. Criminal, cops, and politicians in Colombia and Mexico. Oxford University Press, Oxford
Felbab-Brown V (2019) Mexico’s out-of-control criminal market. In: Foreign policy at Brookings working paper. Brookings, Washington
Grillo I (2011) El Narco. Bloomsbury Press, New York
Guerrero E (2011) La Raíz de la Violencia. Nexos, November 1, 2010. https://www.nexos.com.mx/?p=14318
Hausman J (1978) Specification tests in econometrics. Econometrica 46(6):1251–1271
Hope A (2013) Peace now? Mexican security policy after Felipe Calderón. Latin America working group working paper. Available at: http://thedialogue.orgwww.thedialogue.org/PublicationFiles/Mexico_Security_Hope.pdf. Accessed 18 Feb 2018
Hsieh C-C, Pugh MD (1993) Poverty, income inequality, and violent crime: a meta-analysis of recent aggregate data studies. Crim Justice Rev 18(2):182–202
Lessing B (2012) The logic of violence in criminal war: cartel-state conflict in Mexico, Colombia and Brazil. Doctoral Dissertation, University of California at Berkeley
Lessing B (2015) Logics of violence in criminal war. J Confl Resolut 59(8):1486–1516
Ley S (2018) To vote or not to vote: how criminal violence shapes electoral participation. J Confl Resolut 62(9):1963–1990
Merino J, Zarkin J, Fierro E (2015) Desaparecidos. Nexos, February 1. Available at: http://www.nexos.com.mx/?p=23811. Accessed 15 Feb 2017
Moro F, Petrella A, Sberna S (2016) The politics of mafia violence: explaining variation in mafia killings in southern Italy (1983-2008). Terrorism and Political Violence 28(1):90–113
Morris S (2012) Corruption, drug trafficking, and violence in Mexico. Brown Journal of World Affairs 18(2):29–43
Morris S (2013) The impact of drug-related violence on corruption in Mexico. Lat Am 57(1):43–64
Osorio, Javier (2012) Las Causas Estructurales de la Violencia: Evaluación de Algunas Hipótesis. In José Antonio Aguilar (ed.), Las Bases Sociales del Crimen Organizado y la Violencia en México, 73–132. Mexico City: Centro de Investigación y Estudios de la Seguridad
Osorio J (2014) Democratization and drug violence in Mexico. http://eventos.itam.mx/sites/default/files/eventositammx/eventos/aadjuntos/2014/01/democratizacion_and_drug_violence_osorio_appendix_1.pdf. Accessed 28 Apr 2016
Osorio J (2015) The contagion of drug violence: spatiotemporal dynamics of the Mexican war on drugs. J Confl Resolut 59(8):1403–1432
Phillips B (2015) How does leadership decapitation affect violence? The case of drug trafficking organizations in Mexico. J Polit 77(2):324–336
Pinotti P (2012) Organized crime, violence, and the quality of politicians: evidence from southern Italy. Paolo Baffi Centre Research Paper 2012-124
Ponce A (2016) Cárteles de Droga, Violencia y Competitividad Electoral a Nivel Local: Evidencia del Caso Mexicano. Lat Am Res Rev 51(4):62–85
Ponce A (2019) Violence and electoral competition: criminal organizations and municipal candidates in Mexico. Trends in Organized Crime 22(2):231–254
Ríos V (2013) Why did Mexico become so violent? A self-reinforcing violent equilibrium caused by competition and enforcement. Trends in Organized Crime 16:138–155
Ríos V (2015) How government coordination controlled organized crime. The case of Mexico’s cocaine markets. J Confl Resolut 59-8:1433–1454
Ríos V, Shirk D (2011) Drug violence in Mexico: data and analysis through 2010. Trans-border Institute, University of California-San Diego. https://justiceinmexico.files.wordpress.com/2012/03/2012-tbi-drugviolence.pdf. Accessed 30 Mar 2017
Sánchez F, Palau MDM (2006) Conflict, decentralization, and local governance in Colombia, 1974–2000. Documento de Trabajo CEDE 2006–20, Universidad de los Andes, Bogotá
Sartori G (1976) Parties and party systems: a framework for analysis. Cambridge University Press, Cambridge
Sberna S, Olivieri E (2014) Set the night on fire! Mafia violence and elections in Italy. Paper prepared for the Annual Meeting of the American Political Science Association
Shirk D, Wallman J (2015) Understanding Mexico’s drug violence. J Confl Resolut 59(8):1348–1376
Skaperdas S (2001) The political economy of organized crime: providing protection when the state does not. Econ Gov 2:173–202
Snyder R, Duran-Martinez A (2009) Does illegality breed violence? Drug trafficking and state-sponsored protection rackets. Crime Law Soc Chang 52:253–273
Taylor S (2009) Voting amid violence. Electoral democracy in Colombia. Northeastern University Press, Boston
Toro MC (1995) Mexico’s "war" on drugs: causes and consequences. Lynne Rienner Publishers, Inc, Boulder
Trejo G, Ley S (2016) Federalismo, drogas y violencia. Por qué el conflicto partidista estimula la violencia del narcotráfico en México. Política y Gobierno 23-1:11–56
Trejo G, Ley S (2018) Why did drug cartels go to war in Mexico. Comparative Political Studies 51(7):900–937
Trelles A, Carreras M (2012) Bullets and votes: violence and electoral participation in Mexico. Journal of Politics in Latin America 4(2):89–123
Wantchekon L (1999) On the nature of first democratic elections. J Confl Resolut 43(2):245–258
Wooldridge J (2012) Introductory econometrics: a modern approach. South-Western Pub, Mason
Conflict of interest
There are no potential conflicts of interest.
Research involving human participants and/or animals
This research does not involve participants and/or animals.
This research does not involve participants.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
About this article
Cite this article
Ponce, A.F., López Velarde, R.V. & Santamaría, J.S. Do local elections increase violence? Electoral cycles and organized crime in Mexico. Trends Organ Crim (2019). https://doi.org/10.1007/s12117-019-09373-8
- Criminal organizations
- Municipal elections
- Local electoral cycle