In communities throughout Latin America, criminal organizations provide basic order and security. While multidisciplinary research on criminal governance (CG) has illuminated its dynamics in hundreds of site-specific studies, its extent remains understudied. We exploit novel, nationally representative survey data, validated against a compendium of qualitative sources, to estimate CG prevalence in 18 countries, and explore its correlates at multiple levels. Overall, 14% of respondents reported that local criminal groups provide order and/or reduce crime, corresponding to some 77—101 million Latin Americans experiencing CG. Counterintuitively, CG is positively correlated with both respondents’ perceptions of state-governance quality and objective measures of local state-presence. These descriptive results are consistent with multiple causal pathways, including case-specific findings that state presence—rather than absence—drives criminal governance. We offer suggestions for both more precise data collection on CG itself and, given its pervasiveness, inclusion in broader research on economic development, demography, and politics.

What explains the geography and timing of contestation in civil war? We propose a theory of opportunistic rebel tactics, in which insurgent commanders react to temporary shifts in the local balance of power to attack the state. We argue that these opportunistic strikes are enabled by two jointly necessary factors: (1) negative fluctuations in local repressive state capacity and (2) the expectation of civilian compliance with rebel incursions. We evaluate this argument on data from the Colombian civil war. Leveraging exogenous variation in local state capacity caused by landslide-induced road closures, we find that short-term negative shocks to repressive capacity increase the likelihood of insurgent-state clashes. However, this effect does not hold when local communities harbor strongly anti-insurgent attitudes, suggesting that state capacity and civilian behavior jointly shape rebel strategy and that popular opposition can substitute for state strength.

Leaders who seek to build public toleration for democratic backsliding have a little-noticed strategy at their disposal: degrading their democracies in the eyes of their citizens. If voters can be induced to believe that their democracy is already broken, then nothing of value is lost when leaders attack the courts, vilify the press, or undermine confidence in elections. We call this strategy trash-talking democracy, and study it in the context of contemporary Mexico. We use text-as-data methods to show that President Andrés Manuel López Obrador spends more time trash-talking his democracy than he does trying to deepen partisan polarization. With a survey experiment we show that exposure to López Obrador’s trash-talking of the courts elicits anti-democratic attitudes among Mexicans – both among his supporters and among supporters of the opposition. Strategies to resist backsliding should include not just efforts at de-polarization but also at restoring confidence in democratic institutions.

Under what conditions do insurgents succeed in establishing functional governance institutions in contested areas? Canonical theories of state formation and rebel governance insist that coercive control of territory is a necessary precondition for the development of governing institutions. Yet this claim is belied both by the empirical record and classical guerrilla warfare doctrines. We argue that a lack of consolidated territorial control need not preclude rebel governance. Rather, we posit that low state governance responsiveness enables insurgents to establish institutions in contested areas. Low state responsiveness increases popular demand for insurgent institutions, decreases the costs associated with governing, and enables insurgents to collude with civilians in hiding their institutions. Process-tracing evidence from Ireland, South Africa, and Algeria illustrates our propositions. Our findings shed new light on the determinants of rebel governance and state formation, and cast doubt on the assumption that territorial control is the only path toward statehood.

How do armed actors affect the outcome of elections? Recent scholarship on electoral violence shows that armed groups use violence against voters to coerce them to abstain or vote for the group’s allies. Yet this strategy is risky: coercion can alienate civilians and trigger state repression. I argue that armed actors have another option. A wide range of armed groups create governance institutions to forge ties of political authority with civilian communities, incorporating local populations into armed groups’ political projects and increasing the credibility of their messaging. The popular support, political mobilization, and social control enabled by governance offer a means to sway voters’ political behavior without resorting to election violence. I assess this argument in the context of the Peruvian civil war, in which Shining Path insurgents leveraged wealth redistribution and political propaganda to influence voting behavior. Archival evidence, time series analysis of micro-level violent event data, and a synthetic control study provide support for these claims. These results have implications for theories of electoral violence, governance by non-state actors, and political behavior in war-torn societies.

What explains civilian victimization during civil war? Existing scholarship claims that violence against civilians is driven primarily by competition between armed actors. We argue that this explanation neglects a crucial cause of civilian victimization: in communities they rule, armed groups employ systematic violence against civilians to establish and sustain social order. Drawing on original microlevel quantitative data from Colombia, we show that areas controlled by a sole armed actor experience high levels of victimization, while places where multiple actors jointly govern exhibit significantly less violence. To explain this pattern, we draw on evidence from original interviews, focus groups, and secondary sources. We show that armed groups employ violence to govern areas they control and enact social order. But this violence is checked when multiple groups rule jointly: the factors that sustain pacted rule disincentivize victimization. These results have implications for theories of political order, violence, and rebel governance.

Television is an overlooked tool of state building. We estimate the impact of televising criminal proceedings on public use of government courts to resolve disputes. We draw on survey data from Afghanistan, where the government used television as a mechanism for enhancing the legitimacy of formal legal institutions during an ongoing conflict. We find consistent evidence of enhanced support for government courts among survey respondents who trust television following the nation’s first televised criminal trial. We find no evidence that public confidence in other government functions (e.g. economy, development, corruption) improved during this period. Our findings suggest that television may provide a means of building state legitimacy during war and other contexts of competition between political authorities.

Scholars and the general public have been struck by the norm-shattering rhetoric of President Donald J. Trump. His “rhetorical signature” is heavy with Manichean good-versus-evil messages, vilification of his opponents, and disdain for institutions and for evidence. But many politicians vilify their opponents and style themselves as uniquely able to solve their society’s problems. In fact, Trump’s Manichean discourse is typical of populist leaders, in the United States and around the world. Using text-as-data analysis of campaign rhetoric, we study the content and mood of presidential campaign speeches by a range of U.S. politicians, which allows a broader perspective not only on the uniqueness of Trump’s rhetoric, but also its continuities with the rhetoric of others. This analysis allows us to define Trump as a right-wing populist. Right-wing populists, like left-leaning ones, are anti-elitist and Manichean in words and outlook. However, the two versions of populism differ in the nature of the anti-elitism, with right-wing populists targeting political elites and left-wing ones targeting economic elites. Right-wing populists also define the “other” as ethnic out-groups, who threaten the ethnically pure “people.”

In democracies around the world, armed non-state actors often use force to influence the outcome of elections. These actors leverage the threat of violence to deter candidates they oppose and intimidate voters into turning out for politicians they favor. But we know little about when these attempts succeed or fail. I show that the effectiveness of coercive interventions in elections is inhibited by local party competition. Competitive electoral constituencies attract attention and investment from parties, political elites, and civil society, restricting the ability of armed actors to successfully coerce voters and politicians. I evaluate this argument against evidence from a paradigmatic case of violent intervention in elections: the attempted capture of the 2002 Colombian Senate elections by the Autodefensas Unidas de Colombia (AUC). I find that the AUC’s efforts to channel votes toward allied politicians succeeded in party strongholds but overwhelmingly failed in competitive constituencies. Mechanism tests provide evidence that this relationship operates through the channels I theorize. Robust democratic competition, these results suggest, may offer a bulwark against the violent capture of democracy.

Declining confidence in public institutions is a dilemma facing many democracies, more so when elected leaders deliberately do and say things to shake this confidence. With text-as-data analyses, we demonstrate the harshness of the discourse of one politician - Mexico’s Andrés Manuel López Obrador - against Mexico’s national electoral administration body, the INE, which he was attempting to weaken. Survey experiments demonstrate the president’s capacity to undermine confidence in the INE. In our experiments, the loss of confidence extends even to opposition voters. A more hopeful finding relates to the potential for restoring institutional confidence. Counter-narratives, which presented a more positive and accurate account of the electoral body, restored pubic confidence - surprisingly, even among the president’s supporters. The source of the rebuttals mattered: only ones attributed to organizations above the partisan fray improved confidence. Our findings suggest strategies for breaking out of the cage of intense partisanship and countering democracy-degrading rhetoric.

What explains variation in statebuilding across space? I argue that mass culture shapes the geography of state presence: in deciding where to invest limited governance resources, policymakers are influenced by prevailing societal values. I examine this argument in the context of the Andean coca economy. Leveraging a novel dataset of geolocated state institutions, roads, and natural crop suitability, I show that the relationship between coca cultivation and state service provision hinges on social conceptions of the coca economy. In Colombia, where coca is widely stigmatized, communities where conditions are suitable for coca cultivation are more isolated from state institutions. In contrast, coca-suitable communities in Bolivia, where the crop is broadly valued, have better access to state services. In Peru, where social conceptions of coca are mixed, I find no difference in state presence. These results underscore the importance of considering citizens’ values and preferences in models of contemporary statebuilding.

In urban areas across the Americas, police are confronted with multiple violent criminal groups. How do they choose which of these groups to repress and which to leave alone? We advance a theory centered on the police as autonomous political actors, who develop preferences over criminal actors based on economic benefits, social ties, and ideological alignment. We outline a typology of police-criminal group relations and identify the effect of these relations on two dimensions of police behavior: the allocation of policing resources and police violence. We evaluate this argument against fine-grained spatial data on criminal group control and police behavior in Rio de Janeiro, Brazil. Leveraging geographic variation across criminal group turfs, we find that police preferences over criminal actors have striking consequences for both the degree and type of policing citizens receive. Additional tests suggest that police behavior is not explained by electoral politics or citizen preferences. These results underscore the importance of understanding the police as an independent political actor in the provision of public security.