The influence of “dark networks” on citizens’ confidence in democratic institutions in Mexico

Highlights

•	Direct knowledge of corrupt activity negatively and significantly influenced satisfaction with democracy among Mexican citizens.
•	Direct knowledge of a corrupt act negatively and significantly influenced confidence in democratic institutions and local government.
•	Right-leaning political ideology had a significantly positive influence on confidence in democratic institutions.
•	Younger Mexican citizens appeared to be more satisfied with democracy and other representative institutions than other groups.

Abstract

The Americas have the highest homicide rate of any region in the world, with adherence to rule of law being a major concern in “third-wave” democracies. Given the surge in violence in Mexico in the years after the country ended one-party rule, we utilize the country case to conceptually and empirically test the influence of violence and corruption as proxies for “dark networks” – covert actors working outside of the law – on citizens’ confidence in democracy and democratic institutions. In a largely empirically unexplored area, we operationalize the concept of dark networks and study the impact on citizens over critical junctures of three Mexican presidencies. Utilizing representative survey data, we analyze the impact of dark networks (homicides and direct knowledge of corrupt acts) on satisfaction with democracy and confidence in democratic institutions. Experience with corrupt acts had a significantly negative influence on satisfaction with democracy and confidence in democratic institutions and local government. Homicides had a significant yet only slightly positive influence on satisfaction with democracy and confidence in local government. Right-leaning ideology had a significant yet only weak positive influence on satisfaction with democracy and on confidence in institutions.

Keywords:

• Democracy
• Democratic institutions
• Public opinion
• Dark networks
• Violence
• Security
• Corruption
• Mexico

Notes

1 Democratic institutions in this study are public bureaucracies, including local and other levels of government; congress/parliament; the judiciary; and political parties. These institutions are charged with providing accountability “checks” upon one another in democratic societies and are considered to be representative of the public (Gawthrop, 1998; Gruber, 1997; Przeworski et al., 1999; Selden et al., 2016).

2 The Latinobarómetro survey, modeled after the well-established Eurobarometer and well-known World Values Survey, is an annual survey in 18 Latin American countries coordinated by the nonprofit Latinobarómetro Corporation (Jamison, 2011). According to the methodological description, a random, multistage cluster sampling is conducted among a representative voting age population in Mexico. A different sample is obtained each year, and thus is not a panel. Each year of the study, 1,200 participants were surveyed in face-to-face interviews by those trained for the research. The urban-rural distribution of participants is 77.5% for urban and 22.5% for rural. This nearly coincides with the World Bank’s report of the percentage of citizens in urban areas of Mexico from 2005 to 2013 ranging from 76.31% to 78.70% of the population. The sample was stratified by electoral sections, locality (urban or rural), and population distribution of sex and age across the country, according to the classification criteria of the Federal Electoral Institute. All 31 states and the Federal District were included in each year of the study unless otherwise noted. The margin of error ranged from ∓2.8 to 3.0% points from 2005 to 2013. Data from 2012 was not available.

3 Data was unavailable for this measure in 2006 and 2007.

4 The question wording for all three models presented is the translated wording. Original items in Spanish can be obtained from the Latinobarómetro website.