How not to build a state: Evidence from Colombia

Daron Acemoglu, Leopoldo Fergusson, James Robinson, Dario Romero, Juan F. Vargas

06 October 2016



A major problem in many countries today is the lack of state capacity, including the ability to control violence, enforce laws, tax and regulate economic activity, and provide public services. These limitations are rife in many poor countries, and Fearon and Latin (2003) have argued that this is the root cause of civil war. Despite the potential benefits, developing these capabilities seems to be remarkably hard, with many countries exhibiting persistently state weakness.

How can societies overcome these difficulties and successfully strengthen the state? How can states establish the legitimate monopoly of violence in their territories, something Weber (1946) regarded as the sine qua non of the state? A natural approach would be to focus first on a military strategy aimed at eliminating non-state armed actors and ensuring state dominance. This approach, sometimes referred to as the ‘state first’ or ‘security first’ view, is naturally top-down (generally without the consent or participation of society) and is historically illustrated using the state-building projects of powerful leaders such as Peter the Great, Louis XIV, Kemal Ataturk or Park Chung-Hee (e.g. Huntington 1968, Fukuyama 2001, 2014). This view reaches much farther than academic circles, and has become the guiding principle for US interventions in Afghanistan and Iraq in recent years and guides a great deal of international development guidelines (Grävingholt et al. 2012, World Bank 2012).

Top-down approaches, however, are typically unidimensional and prioritise military objectives ahead of all others. In a recent paper, we argue that these approaches may create significant negative consequences (Acemoglu et al. 2016). Moreover, they may not only fail to develop other crucial aspects of state capacity, but may also lead to deteriorations in these incipient capacities.

We study the consequences of efforts to establish the state’s monopoly of violence in Colombia following the election of Álvaro Uribe as president in 2002. President Uribe formulated a classic top-down state-building project, focusing on combating non-state armed actors, particularly left-wing guerillas. His ‘Democratic Security Policy’ comprised two main pillars: expanding the size of the military, and increasing their incentives to fight the guerillas. A report by Human Rights Watch (2015) describes the introduction of incentives after 2002, which “rewarded combat killings with vacation time, promotions, medals, training courses, and congratulations from superiors, among other prizes” (p. 29).

Figure 1 False positives by semester
Cases and casualties, 1988-2011

Notes: False positives between the first semester of 1988 and the second semester of 2011. Cases are the total number of events producing false positives, while casualties are the total number of people that were killed in these events. In both cases we depict the 3-semester moving average of the raw numbers.
Source: Acemoglu et al. (2016), with data from CINEP.

A major consequence of these higher-powered incentives was a surge in ‘false positives’, where the army killed civilians and falsely portrayed them to be guerilla combatants. These cases are depicted in Figure 1, which shows both incidents producing false positives and the number of people killed in these events. False positives had long existed in Colombia, but increased massively following President Uribe’s state-building project, and started declining only after policies where modified following media revelations of the extent of civilian killings in 2008. Figure 2 shows the distribution of false positives in the territory, revealing that the practice was widespread throughout the country, and not just driven by a few rogue military units.

Figure 2 False positives
Total executions per 100,000 inhabitants

Notes: False positives (per 100.000 people) per municipality over the entire sample period (2000-2010).
Source: Acemoglu et al (2006) with data from CINEP (false positives) and DANE (population).

A simple theory (in the spirit of Holmström and Milgrom’s 1991 multi-tasking framework) helps clarify the relationship between the structure of incentives, other dimensions of state capacity and the unintended consequences of top-down unidimensional state-building efforts. The theory has some simple predictions we can test.

  • First, more powerful incentives for military personnel to kill guerillas will increase agents’ effort leading to both false positives and real guerilla fatalities (what we call ‘true positives’).
  • Second, this effect will be more pronounced for brigades led by colonels because they have more powerful career concerns (the promotion from colonel to general is a difficult step in most armies, including in Colombia).
  • Third, the effect on false positives will also be more pronounced in municipalities where local judicial institutions are less able to investigate and hold accountable military units and their commanders. Crucially, weak judicial institutions will impact false positives but not necessarily true positives.

We show that these predictions, including the last asymmetry, are consistent with our data. These findings underscore the negative consequences of the one-dimensional approach to state building. The asymmetric response of true and false positives in weakly institutionalised areas also bolsters our interpretation that what we document is not just unavoidable collateral damage in the process of attacking real guerillas, but systematic actions taken by military units directed towards killing civilians and pretending that they were guerilla combatants.

It also falls in line with growing case study evidence from judicial and media investigations. UN Commissioner Philip Alston observed that the pressure to ‘show results’ and rewards for doing so is cited by experts, even within the military, as one of the causes of false positives. A soldier explained to him how a killing by his unit would be rewarded with 15 days of vacation: “When important holidays approached, he stated, soldiers would attempt to ‘earn’ vacation time” (Alston 2010, p. 11). Another soldier, who witnessed as many as 25 false positive cases occurring in 2007 and 2008, refers to government Directive 29 of 2005, and notes that to claim the monetary rewards it promised for killings and war material, army members would kill civilians and ‘plant’ weapons on them.

The evidence we present also shows that the type of top-down state-building strategy that Colombia adopted might not just create a human tragedy, it can backfire in terms of its intended goals. Imagine again those state agents facing strong incentives to kill guerrillas. It will be easier to get away with killing civilians instead when the quality of judicial institutions is lower. Under such circumstances they may take actions to further weaken the local judiciary. The empirical evidence indeed points to a deterioration in the quality of judicial institutions in areas with a high share of brigades commanded by colonels, and perhaps more paradoxically, a worsening of the security situation in such areas (with increases both in guerilla and paramilitary attacks on civilians).

The top-down, unidimensional state-building project in Colombia was therefore counter-productive even in terms of the goals it was trying to achieve. It weakened other aspects of state capacity in the process, and perhaps also further undermined the legitimacy of the state and its consensual strength, which may be central to its capacity (see Acemoglu 2005 for the theoretical argument, and Isacson 2012 on the Colombian case). The main lesson we draw from this analysis is that even if one’s goal is to attain a legitimate monopoly of violence, it is critical to try to build state institutions in multiple dimensions simultaneously and that high-powered incentives in the absence of accountability, and when state institutions like the judiciary are weak, can be highly perverse.

There are many other examples of top-down, unidimensional state-building efforts that might be expected to have similar perverse, unintended consequences. In Peru and Guatemala, for example, post-conflict truth and reconciliation commissions have also documented widespread killings of civilians. In Peru, the impetus for these murders came, according to the commission’s report, from a top-down, security-first logic: “by privileging a military approach, one of the main objectives of the counterinsurgent strategy was the elimination of members, sympathizers or collaborators of armed insurrection, even more than the objective of capturing them to be judged by the competent judicial authorities” (Comisión de la Verdad y la Reconciliación 2003, p. 146). The Guatemalan commission also reaches a similar conclusion to our study, stating:

“Militarization became a pillar of impunity. Moreover, in a broad sense, it weakened the country’s institutions, reducing their possibilities for functioning effectively and contributing to their loss of legitimacy” (Comisión para el Esclarecimiento Histórico 1999, p. 28).

It goes on to conclude: “The justice system, nonexistent in large areas of the country before the armed confrontation, was further weakened when the judicial branch submitted to the requirements of the dominant national security model.” (p. 36)

These issues are relevant beyond Latin America as well. Top-down attempts to recreate the monopoly of violence that the state had lost in Somalia, Afghanistan, and Iraq all appear to have backfired in recent decades. From the vantage point of our general approach, this might have been, in part, because they sought to create strong incentives for the security forces to combat rebels or insurgents, without much of an effort to build accompanying institutions or support from the local population.

Tolstoy’s famous opening sentence of Anna Karenina comes to mind: “Happy families are all alike; every unhappy family is unhappy in its own way.” Where many things must come together for something to work well, failure can be achieved in a variety of different ways. State building – just like relationship building – is inherently hard. And one thing that seems clear is that for it to work, you really must work on many dimensions (see also Besley and Persson 2011).


Acemoglu, D (2005) “Politics And Economics In Weak And Strong States,” Journal of Monetary Economics, 52(7), 1199-1226.

Acemoglu, D, L Fergusson, J A Robinson, D Romero and J F Vargas (2016) “The Perils of Top-Down Statebuilding: Evidence from Colombia’s ‘False Positives’,” NBER Working Paper No. 22617.

Alston, P (2010) “Report of the Special Rapporteur on extrajudicial, summary or arbitrary executions,” Mission to Colombia, United Nations, Human Rights Council.

Besley, T and T Persson (2011) Pillars of Prosperity, Princeton: Princeton University Press.

Comisión de la Verdad y la Reconciliación (2003), Informe Final, Tomo VI: Sección cuarta: los crímenes y violaciones de los derechos humanos. Available at

Comisión para el Esclarecimiento Histórico (1999), Memory of Silence: Report of the Commission for Historical Clarification: Conclusions and Recommendations, Available at en.pdf.

Fearon, J D and D D Laitin (2003) “Ethnicity, Insurgency, and Civil War,” American Political Science Review 97(1): 75-90.

Fukuyama, F (2011) The Origins of Political Order: From Prehuman Times to the French Revolution, New York: Farrar, Straus and Giroux.

Fukuyama, F (2014) Political Order and Political Decay: From the Industrial Revolution to the Globalization of Democracy, New York: Farrar, Straus and Giroux.

Grävingholt, J, J Leininger and C von Haldenwang (2012) “Effective statebuilding? A review of evaluations of international statebuilding support in fragile contexts,” Available at

Holmström, B and P Milgrom (1991) “Multitask Principal-Agent Analyses: Incentive Contracts, Asset Ownership, and Job Design,” Journal of Law, Economics and Organization, 7(0), 24-52.

Human Rights Watch (2015) “On Their Watch: Evidence of Senior Army Officers’ Responsibility for False Positive Killings in Colombia,” Available at

Huntington, S P (1968) Political Order in Changing Societies, New Haven: Yale University Press.

Isacson, A (2012) “Consolidating Consolidation,” Washington Office on Latin America, Available at Last accessed March 7, 2016.

Weber, M (1946) “Politics as a Vocation,” in From Max Weber: Essays in Sociology, translated and edited by H.H. Gerth and C. Wright Mills, New York: Oxford University Press.

World Bank (2012) “Guidance for Supporting State Building in Fragile and Conflict-Affected
States: A Tool-Kit,” Available at 285741-1343934891414/8787489-1347032641376/SBATGuidance.pdf 



Topics:  Development Institutions and economics Politics and economics

Professor of Applied Economics, MIT

Associate Professor of Economics, Universidad de los Andes, Bogotá

Professor, University of Chicago Harris School of Public Policy

PhD student in Economics, Columbia University

Research Economist, CAF-Development Bank of Latin America; Professor of Economics, Universidad del Rosario