Games of Mutual Extortion: Fuel Theft, Electoral Competition, & Electoral Integrity in Mexico

Over the past two decades there has been growing scholarly interest in understanding the relationship between organized criminal groups (OCGs) and state actors in developing democracies. Moving away from accounts that emphasize a strictly oppositional zero-sum game between OCGs and the state, an emerging literature on the state-criminal nexus focuses on the conceptual area where OCGs and state actors often coexist, collaborate, and collude (Arias, 2017; Barnes, 2017; Lessing, 2021; Trejo and Ley, 2020; Magaloni et.al., 2020). 

Most explanations regarding the formation of collusive arrangements between OCGs and political actors view OCGs as the principal while political actors as the agent in the relationship. In these cases, OCGs use violence or corruption to subdue political actors and obtain their compliance, the so-called silver or lead extortion of political actors. The key idea about this dynamic is that extortion emanates from OCGs towards state actors. The less commonly studied pathway in the relationship between OCGs and political actors is one in which OCGs act as political actors’ agents, especially in the electoral arena.

My dissertation contributes to this emerging literature on the state-criminal nexus by studying collusive arrangements in which OCGs act as political actors’ agents in electorally competitive regimes. In that endeavor, I first develop a theory about the logic behind the formation of such collusive arrangements that I call games of mutual extortion and then I analyze the effects of such arrangements in the electoral arena through a comprehensive examination of the criminal activity of fuel, gas, and oil (FGO) theft in Mexico.

 In my dissertation I first develop a theory of collusive arrangements that I call games of mutual extortion where political actors extort OCGs to obtain campaign cash from OCGs, while OCGs use extortion to obtain cooperation and protection from politicians. I argue that collusive arrangements in which OCGs act as political actors’ agents are more likely to form when the primary criminal activity that OCGs engage in is stationary as opposed to mobile, including illegal mining or pipeline theft.

 I argue that in places with stationary criminal activities, incumbent political actors have more leverage over OCGs because “exit” is not an option for OCGs if they wish to continue profiting from the illegal activity (Hirschman, 1970; Clark, Golder, and Golder, 2017). While OCGs who engage in non-stationary criminal activities can relocate if incumbent political actors choose to extort them for campaign cash, OCGs who engage in stationary activities are forced to either comply or resist to continue profiting. Because resisting is costly, compliance is more likely under such scenarios and thus more likely that collusive arrangements in which OCGs act as political actors’ agents form. What are the political effects of such arrangements?

Using the recent phenomenon of fuel, gas, and oil (FGO) theft in Mexico to understand the political effects of such collusive arrangements, I find that FGO theft reduces electoral competition in state and municipal elections in favor of municipal incumbents and their gubernatorial allies. I also find that unlike most studies on organized crime and local elections which stress violence as the causal mechanism mediating outcomes, I find clientelism to be a more relevant and robust explanation about why collusive arrangements reduce electoral competition in local elections than violence. Lastly, I find that FGO theft increases citizens’ perceptions of corruption, decreases the approval rating of incriminated local political incumbents, and increases the incidence of electoral malfeasance in subsequent elections.

FGO THEFT

Empirically, my dissertation contributes to the study of organized crime by documenting in great detail a specific type of criminal activity theorized to increase the likelihood of collusive arrangement formation – FGO theft. FGO theft is a complex criminal activity in which OCGs illegally tap underground pipelines to extract FGO products. It is a stationary criminal activity because it can only occur in specific localities that have pipelines. In Mexico, FGO theft is an enduring criminal activity that occurs in 178 out of the close to 2,500 municipalities that exist in the country. The process of theft is complex, as it requires having knowledge of transport schedules and the precise location of the pipelines. It also requires high-level engineering expertise to successfully install high-pressure valves and alternative pipelines, and it requires the collaboration and coordination of entire networks of individuals in the upstream and downstream production and distribution chains. In Mexico, while FGO theft was a relatively minor illegal activity in the 1990s, by 2018, this criminal activity on average commercialized 56,000 barrels of stolen FGO products daily – or a $3.315 billion per year industry (Perez, 2020).

 After conducting extensive field work in Mexico, the data collection process for my dissertation involved making freedom of information requests to the Mexican government, processing thousands of documents, and creating original geocoded datasets containing, among other things, the daily number of FGO illegal taps in Mexican municipalities since the 1990s, the location of FGO pipelines, and the daily volume of fuel stolen from pipelines. In addition, I also collected polling station level local election results, public opinion surveys, campaign finance data, and conducted interviews with high-level party officials, journalists, and political operatives. In that endeavor, I not only documented how this criminal activity evolved for more than two decades, but also where it evolved and its extent, which in turn allowed me to precisely estimate the effects of said illegal activity on a range of outcomes, including electoral competition, clientelism, corruption, legitimacy, and electoral fraud.

 ELECTORAL COMPETITION

In my dissertation, I test the effect of FGO theft on local electoral competition to understand how collusive arrangements affect local democratic outcomes. To that effect, employing a regression discontinuity design (RDD), instrumental variable regressions, and geographic analyses, I find that FGO theft reduces electoral competition in municipal and gubernatorial elections in Mexico, and that these results are most pronounced in precincts within 5km of the pipelines. While other studies have found that in Mexican municipal elections incumbents suffer from an incumbency disadvantage (Lucardi and Rosas, 2016; Klašnja and Titiunik, 2017), I find that in municipalities that experience FGO theft, incumbents are significantly more likely to win re-election than incumbents in other municipalities. Since robust electoral competition serves as a mechanism of democratic accountability, the results from this chapter suggest that FGO theft undermines democratic accountability in Mexican local elections. But what is the causal mechanism that mediates the effect of FGO theft on electoral competition?

CLIENTELISM

Recent studies on organized crime in Latin America argue that OCGs may use violence (Trejo and Ley, 2020), coercion, intimidation (Cordova, 2019), or a combination of these factors (Albarracin, 2018; Trudeau, 2022) to influence elections. These studies assume that OCGs have some degree of control over local populations which allows them to influence electoral outcomes through mobilization or de-mobilization strategies. In almost all accounts, elections serve to protect the interests of OCGs while politicians are either victims or passive beneficiaries of election results.

 I, instead, argue that political actors are often the principal in the formation of collusive arrangements in places with stationary criminal activities while OCGs the agent. Political actors seek revenue derived from criminal activities to finance campaign operations to win elections. In this way, I argue that FGO theft revenue is used by local political incumbents to ignite clientelistic networks and mobilize voters on election day which results in a significant incumbency advantage over the opposition.

To that effect, I first estimate campaign expenses in municipal elections, comparing the amount of money spent by candidates in municipalities with and without FGO theft. I find that while one third of the candidates in municipalities with FGO theft reported no campaign expenses, on average, the winning candidates reported spending 30% less than the winners in municipalities without FGO theft. This finding, along with information I obtained from interviews with high-ranking party officials and political operatives, suggests that candidates in FGO theft municipalities utilize revenue from FGO theft to finance local campaigns.

Moreover, to evaluate whether the illegal campaign funds are used to finance clientelism and test whether clientelism is a possible causal mechanism mediating the effects of FGO theft on electoral competition, I collected data from public opinion surveys from 2000 to 2020. Using a variety of statistical methods, I find a robust positive relationship between FGO theft and clientelism. I also find that while vote-buying has increased in Mexico since national democratization began, significantly more reported vote-buying occurs in municipalities that experience FGO theft than in municipalities that do not.

Moreover, using survey evidence from the 2018 election, I also find that part of the reason why more clientelism occurs in places with FGO theft besides access to illegal campaign funds is that clientelism is an effective electoral strategy in these places because brokers have a higher capacity to monitor voters and to enforce clientelistic exchanges.  In FGO theft municipalities, for example, voters report having significantly less confidence in their vote being secret than voters from municipalities where this criminal activity does not occur, and citizens who received vote-buying offers in municipalities that experienced FGO theft during the election/campaign period reported feeling more afraid of local authorities than citizens who also received vote-buying offers but who do not live in municipalities that experienced FGO theft.

 CORRUPTION, LEGITIMACY, & ELECTORAL INTEGRITY

The last part of my dissertation focuses on the way that political actors who collude with OCGs govern and the likelihood that they commit electoral fraud to remain in power. I argue that that as incumbent political actors come to power through corrupt and illegal arrangements, these actors don’t govern democratically by representing the broad and diverse interests of their constituents, but instead through the expansion of corruption. As corruption becomes generalized, this paradoxically reduces political actors’ legitimacy in the eyes of the citizens, which simultaneously increases the costs of losing future elections. To avoid accountability demands in contexts of decreased political legitimacy, political actors involved in criminal activities opt for committing electoral fraud to remain in power.

Using Mexico’s experience with FGO theft to test these propositions, I first use survey evidence to analyze citizens corruption perceptions and approval ratings across municipalities that either experienced FGO theft or not, before and after they experienced FGO theft. I find that that in accordance with my expectations, in municipalities that experienced FGO theft, respondents’ perceptions of corruption and levels of disapproval of local incumbents significantly increased after experiencing FGO theft.

In terms of electoral fraud, employing election forensic methods to estimate fraud magnitudes in municipal elections, I find that significantly more fraud occurs in FGO theft municipalities than in other municipalities. To analyze electoral fraud in gubernatorial elections, I use the case of Puebla and all of the state’s elections from 2004-2019 to assess electoral malfeasance across FGO and non-FGO theft municipalities. I find that in 2018, eight years after scholars of subnational authoritarianism considered the state to have undergone successful democratization (Gibson, 2013; Giraudy, 2015), the gubernatorial candidate who had an extensive relationship with FGO thieves and whose political group had governed since subnational democratization took place, committed electoral fraud to remain in power. In that regard, one of the implications of my findings is that future work on subnational authoritarianism and on democracy more generally, should carefully consider whether party alternation is a sufficient condition to consider cases to have undergone democratization. Non-procedural features of democracy, including the way that incumbents come to power and govern, may be more informative than party alternation alone. Nevertheless, my findings regarding electoral fraud in Puebla suggest that subnational authoritarianism is alive and well, as even the most basic procedural democratic conditions were violated in 2018.

References Cited:

Albarracín, J. (2018). Criminalized Electoral Politics in Brazilian Urban Peripheries. Crime, Law and Social Change, 69, 553-575.

Arias, E. D. (2017). Criminal Enterprises and Governance in Latin America and the Caribbean. Cambridge University Press.

Barnes, N. (2017). Criminal Politics: An Integrated Approach to The Study of Organized Crime, Politics, And Violence. Perspectives on Politics, 15(4), 967-987.

Clark, W. R., Golder, M., & Golder, S. N. (2017). The British Academy Brian Barry Prize Essay: An Exit, Voice and Loyalty Model of Politics. British Journal of Political Science, 47(4), 719-748.

Córdova, A. (2019). Living in gang-controlled neighborhoods: Impacts on electoral and nonelectoral participation in El Salvador. Latin American Research Review, 54(1), 201-221.

Gibson, E. L. (2013). Boundary Control: Subnational Authoritarianism in Federal Democracies. Cambridge University Press.

Giraudy, A. (2015). Democrats and autocrats: Pathways of Subnational Undemocratic Regime Continuity Within Democratic Countries. Oxford University Press, USA.

 Klašnja, M., & Titiunik, R. (2017). The incumbency curse: Weak parties, term limits, and unfulfilled accountability. American Political Science Review, 111(1), 129-148.

 Hirschman, A. O. (1970). Exit, Voice, And Loyalty: Responses to Decline in Firms, Organizations, and States. Harvard University Press.

 Lessing, B. (2021). Conceptualizing Criminal Governance. Perspectives on Politics, 19(3), 854-873.

 Lucardi, A., & Rosas, G. (2016). Is the incumbent curse the incumbent's fault? Strategic behavior and negative incumbency effects in young democracies. Electoral Studies, 44, 66-75.

 Magaloni, B., Robles, G., Matanock, A. M., Diaz-Cayeros, A., & Romero, V. (2020). Living in Fear: The Dynamics of Extortion in Mexico’s Drug War. Comparative Political Studies, 53(7), 1124-1174.

 Perez, A. (2020). Huachicol un Desafio de Seguridad Nacional. Atlas de la Seguridad y la Defensa de México. Casede. 

 Trejo, G., & Ley, S. (2020). Votes, Drugs, and Violence: The Political Logic of Criminal Wars in Mexico. Cambridge University Press.

 Trudeau, J. (2022). How Criminal Governance Undermines Elections. Unpublished Manuscript.