Benjamin Lessing: “The first thing I would say to politicians is that you shouldn’t fall into the temptation of an iron fist (mano dura) policy”.

University of Chicago’s political scientist and expert in criminal governance explains how Latin American gangs were transformed into Criminal Leviathans capable of deploying their power over vast territories. He warns that repressive, mano dura (“iron fist”) policies and practices do not stop this process but instead drive it. This interview is an academic-journalistic effort that is part of the series Violent Democracies by the Vio-Demos Millennium Institute and TerceraDosis.

Aquí, la versión en español. Here, the Spanish version

Translated by Emilia Guzmán/ Translation reviewed by Benjamin Lessing

Many believe that we would be better off if we punished more. Therefore, they think that legal systems with fewer guarantees and longer prison sentences would teach criminals their lesson. But even those persuaded by that rationale know that “mano dura” does little and harm a lot. In fact, not even the staunchest advocates of long sentences would feel safe next to someone who has just spent 10 or 15 years in a Latin American prison.

The Brazilian case is probably the best example of how increasing incarceration rates is counterproductive. Today, most poor areas of Brazilian cities are controlled by criminal organizations born inside prisons. Let’s repeat it so that it is well understood: the criminal organizations that control large areas of Brazilian cities today, popularly known as criminal factions– formed and consolidated inside prisons at a time when several states in Brazil began to increase “mano dura.”

Brazil’s criminal factions were formed in violent prisons, under inhuman and degrading treatment, where a punitive state even resorted to the killing of prisoners. In the end, the criminals did learn a lesson, but it wasn’t the one that was expected.

Benjamin Lessing, the first interviewee in this Violent Democracies series, has studied this phenomenon in depth since the 2000s. Lessing is an associate professor of political science at the University of Chicago and the author of a highly influential book, Making Peace in Drug Wars, on cartel—state conflict in Mexico, Brazil, and Colombia. His new book project, Criminal Leviathans, explores the counterproductive effects of “mano dura” policies including mass incarceration and violent policing in peripheral neighborhoods. Simply put, he argues that mass incarceration signals to the typical poor young man living in the favelas that he is very likely to be imprisoned at some point in his life, whether he has committed a crime or not. One natural response, a self-protection strategy, is to ensure one is on good terms with the neighborhood’s ruling prison gang so that when one arrives in prison, he is not killed or raped. Thus, mass incarceration itself encourages gang recruitment. In other words, this is a perverse mechanism through which the state pushed generations of poor young people into the hands of prison gangs.

However, the failure of punitive policies goes beyond gang recruitment. Perhaps the most relevant lesson we can draw in Latin America derives from understanding how these prison gangs came to rule over not only criminal actors in peripheral communities, but the millions of civilians living there as well. What are the causes of this phenomenon, and what is the role and responsibility of the state in this? “Criminal governance” is the field of study that addresses those questions in political science, and Lessing is one of the leading figures in this vibrant field.

Lessing’s contribution starts from the following assumption: coercion alone cannot explain how criminal governance emerges; the ability of criminal factions to develop self-governing structures is just as important as its coercive potential. Brazilian factions, Lessing explains, developed self-governing structures in prison. In the 1980s and 1990s, Brazilian prisons, especially those in states pursuing “mano dura”, were hell. Prisoners were killed and raped, there were no places to sleep due to overcrowding, and there was no food. Moreover, police guards lacked limits and accountability. The killing of 111 prisoners who had rioted in the prison of Carandiru, São Paulo, in 1992 ended up being a turning point. Repression by the Military Police did not provide a lesson, but an incentive. One year later, the First Capital Command (PCC) was formed and expanded to several prisons. The PCC managed to impose order inside prisons, which was the basis of its growing power. First, it established internal order, banning rape, assault, killing, and all inter-inmate violence not authorized by the faction. The PCC then leveraged inmate’s loyalty to pressure the state, by organizing revolts and protests. More and better food, as well as less violence by guards, resulted from PCC’s ability to provide order within prisons. According to Lessing, the power of the PCC lies in its ability to improve inmates’ lives by organizing them to fend off an abusive state.

The PCC then extended its grasp and “pacifying” capacity to the streets of São Paulo: in areas controlled by the faction, no one kills without the PCC’s authorization. Unexpectedly (and disturbingly!), while in the rest of the country homicides escalated during the same period, homicide rates in Sao Paulo fell after PCC’s rule consolidated. (Watch Lessing’s presentation on this subject.)

What lessons should be drawn from this narrative on the origins and implications of criminal governance? What are the public policy implications of such a narrative? We let Lessing speak for himself on that in the following exchange. His analysis shows how promising the field of criminal governance is for understanding the emerging structure of organized crime in Latin America, as well as the limitations it imposes on state actors and political leadership. On that last note, let us just anticipate that for Lessing, letting the peripheries be controlled by prison gangs is something that has been going on for decades. Indeed, Lessing argues that “criminal governance” of areas populated by the poor in Latin America could be considered Latin America’s implicit development model. That in itself turns folding criminal governance back unlikely and an extremely difficult task.

-You started on the issue of organized crime, trying to understand why the traffickers in Rio wanted to have military-grade weapons and why the war on drugs had become militarized. What did you find?

-I observed two dynamics. The first is the war between traffickers for territory, which points to a classic security dilemma: say I am the boss of the Rocinha favela, and you are the boss of Vila do Vintém. If I acquire arms to defend myself, you buy more weapons to defend yourself against me. That logic, of course, may explain gang wars, but it doesn’t explain why traffickers fight the state. To understand that I had to theorize and propose other logics. One strong component, not the only one, but a very important one, involves the corruption generated by drug trafficking. Corruption that is not exceptional, but regular, that traffickers can calculate as part of their business and that involves, for example, a fixed weekly or monthly payment to the police. Now, let’s add to that the fact that political leaders almost always promise to be tough because it’s popular. They put more police on the streets with more leeway to use their firepower, to kill suspects, or invade areas without search warrants. If the police are already extorting money from traffickers, then giving them more repressive power, more freedom to enter homes and so on, amounts to increasing their power to extort. For example: say a cop is about to fine me $200; I’d be willing to pay a bribe of up to $199 to get out of the fine. I’m not going to pay any more than that, because it would be better for me to just pay the fine. But say the police officer can kill me, because nobody cares about killing a bandit; then I’m going to pay a lot more than $200. So, the increase in the powers of the police creates an incentive for the criminal to arm himself, because if he has guns, the policeman will think twice about asking for so much money.

“The incarceration rate helps the groups inside the prison exercise criminal governance on the street”

-What we must understand, then, is that if we give more power to a corrupt police force, we will make the situation worse. Before the iron fist, should the police be cleaned up?

-That is an interesting question. Suppose the police are not corrupt but we still give them heavy-handed repressive power. Does that eliminate the backlash against cops? I don’t know. Maybe not, because if the police have the authority to kill me, as a criminal I may not care much if they are corrupt or not. I may still try to defend myself with bullets. If the police are going to kill me no matter what I do, I don’t have much incentive to give up. I would say that it is the violence of the police itself that creates incentives for criminals to respond with violence. And I do think that’s stronger when it’s mixed with corruption.

–In a presentation you said that if the state throws all its force against the drug traffickers, they will defend themselves with everything they have, but if it leaves the traffickers some space, they may stop being violent. Now, to leave space for the drug traffickers is to let them traffic, right? When I heard that presentation, I thought of your idea that Latin American societies face three problems that they cannot solve at the same time: drug trafficking, violence, and corruption. You propose that we have to decide.

-Yes, in my book I talk about that as a policy trilemma, an «unholy trinity.»

–I’d like us to delve into that to see what limitations you think public policy has to control violence.

-To understand the “unholy trinity”, it helps to start prior to the drug war, when you just have a product that a lot of people wanted to use. Illicit drugs constitute markets with strong demand, demand that is also quite inelastic and that has its negative externalities: it is a product that destroys lives, and therefore the State decides to prohibit it. Very well, then the state bans drugs and initiates repression of drug selling. The problem is that since it is a product with strong and inelastic demand, repression causes prices to rise. In fact, the very purpose of repression is to reduce supply and raise the price, to discourage consumption. But when the price goes up, it becomes a more lucrative business. This brings us to the present day, where repression has not been effective in putting an end to consumption and trafficking. This is very different from what has happened with other cases of global prohibition. For example, if we look at slavery, we will see that at one time it was also banned and repressed by states, mainly by England in the 19^th century. Eventually, the repression phase succeeded, and the practice became rare. Today it’s really very rare to hear of ships full of slaves in the Atlantic. What you see are ships full of drugs. The drug trade is stronger than ever. Prohibition and repression failed to make consumption rare. On the contrary, it made trafficking more lucrative, creating incentives for traffickers to do two things: arm themselves and bribe the police. And this failure also gave the police incentives to take those bribes. At the end of the day, the drug trade is still going to happen. So, if I’m a cop at the border and the drug trade is happening, I’d better get something, because I’m not going to be able to completely stop the traffic anyway.

Thus, today the State is facing all three of these problems at the same time. Regularized corruption; extreme violence by traffickers, both among themselves and vis à vis the State; and the flow of drugs itself, which was the original target of all this. My argument is that we can’t fight all three problems at the same time. The more you fight the flow of drugs, the more you encourage violence and corruption. Consequently, my recommendation is to think about what the most harmful problem is and focus on that. And for me, violence is the most harmful, especially when directed at the State. The war between the cartels also does a lot of damage, but in some contexts, when traffickers have incentives, manage to make pacts among themselves and more or less minimize that type of violence. For example, in low violence-contexts, the cartels understand that if a shootout occurs, it will attract a lot of attention, and so they avoid it. In Medellín, for example, drug gangs have been «domesticated,» in a sense. Gustavo Duncan shows that organized crime in that city has become much more peaceful over time, and indeed Medellín today has a very low homicide rate. It was the homicide capital of the world in the time of Pablo Escobar and today it is a fairly safe city. But it has not ceased to have very strong trafficking groups. So, I’m saying that it’s much better to have less violence and a larger flow of illicit drugs, than a situation with a lot of violence, an open militarized drug war in the middle of the metropolis, and maybe a little less flow of drugs, but still big enough to cause a lot of damage.

«Criminal governance and the coercive power of the state are a whole, an organic being that is split in two. It’s an ecosystem that is co-evolving.»

–From this point of view, the idea of «mano dura» (“iron fist”) or a «war on drugs» makes no sense at all. Is it a war lost in advance?

-Yes. However, in the new book I’m working on, I add another perspective that goes beyond the idea that the war on drugs is counterproductive. I believe that it is counterproductive, and that state repression is not going to put an end to drug consumption; in my view this is already well demonstrated. But if the war on drugs has long been known to be a failure, then why does it persist as a solution? That is what I’m trying to explain in this new project.

-And what did you think?

-A very common explanation is to attribute it to corruption. Since the drug market produces so much illicit profit, it is possible that the state is also gaining from that business and so it is in its best interest to keep up the war. I don’t think that’s 100% wrong. When I published my first book in 2018, my argument was that corruption played a fundamental role in cartel–state violence. Today I think it has an important role, but I don’t know if it explains everything. I don’t know if presidents, for example, think: “I’m going to keep up this drug war in order to get money out of the traffickers.” I don’t think Brazilian Presidents Lula or Bolsonaro thinks like that. So, I went looking for a better theory and took up a topic that I had set aside in my first book, which is the question of criminal governance. That phenomenon was not very strong in some of the cases I studied in that book, Mexico under Felipe Calderón case and Colombia in Pablo Escobar’s time. But in the case of Rio de Janeiro, criminal governance is essential. In fact, many of the first studies on criminal governance come from Rio.

-Can you define criminal governance in simple terms?

-Of course. The phenomenon has been called by many names and there are many theoretical frameworks for it. The basic idea is that in many communities there are armed groups that do not have political aims, not insurgents or paramilitaries, but simply gangs, traffickers, criminal groups. And these groups, though they do not seek to take over or topple the state, nonetheless impose rules on the community and provide social order. Generally, they prohibit robbery, rape, and settle disputes. Sometimes they provide public goods and aid for the poorest, such as medicine or food staples. And in these ways, some gangs more than others, they end up being a form of local government. They do not replace the state, which can have more or less presence. There is almost always some overlap between the state and criminal groups. But every community is different. In Rio alone, there are more than a thousand favelas, and each one has its own local dynamics. Imagine the variety within Brazil, or across South America.

However, in general criminal groups govern a civilian population so that the state does not need to enter that territory so much. The message is, «don’t call the police if someone stole your bike, I’m going to figure it out for you, I’m going to ban motorcycles from being stolen, I’m going to ban disappearing people, we’re going to find out who it was, so the police don’t have to come in here.» That’s not to say the police can’t get in. They probably can, but they often stop coming in because that gang-provided order is fine with them. This is what I mean by “criminal governance”. I’ve sought to make this concept broad and encompassing, to be a concept that can be applied to as many contexts as possible. In Rio, scholars initially talked about “parallel power”. But then people asked, “Is it really ‘power’? Is it really a parallel state?” It clearly falls short of a state. In other places, gang governance is much weaker than in Rio, but still exists. So, the concept of criminal governance is meant to be a little more comprehensive and help understand the variation between cases.

In the urban settings I study, criminal governance usually includes civilians who live with criminal groups that impose a social order, or impose a set of rules, or adopt other governmental functions, but who do not replace the state entirely. This implies that de facto there is a certain duopoly of violence. What do I mean by that? That there are two governing authorities, each with its own coercive force: the police and the gang. And I, as a citizen, have to obey both. I obey the rules that the criminal group imposes on me, and, at the same time, I remain a citizen, that is, I vote, I pay taxes. This reality exists in many peripheries of Latin America and, it seems, is becoming more and more common. The work of Juan Pablo and his students has detected this phenomenon in the countries of the Southern Cone – Uruguay, Chile – which are not typically seen as having very strong organized crime.

“Today we have a thing that is half state, half criminal group, and that thing, those two halves together, is what governs the citizens”

-At the beginning, Rio didn’t have the big cartels found in Mexico, in Colombia, but in recent years the Brazilian gangs have been getting stronger and are starting to become global players that are very important to understand the dynamics of drug trafficking in Latin America. What is the mechanism for strengthening these gangs and the role that the prison played in this?

-A peculiarity of Brazil and Rio is that criminal groups, such as the Comando Vermelho (Red Command, CV), start inside the prison and then expand into the streets and take control of the favelas. To explain this phenomenon, scholars began to talk about criminal governance in the 1990s. For a while it was thought that this happened only in Rio, but then the PCC was born in São Paulo and took control of the city in a frightening way. In the last 15 years, this «faction» model, as they call it, has spread throughout Brazil and neighboring countries. I think it’s important to distinguish two parts of this trajectory, two currents. One thing that is very important is the fact that the PCC, and also the CV, have reached a level where they function to some extent as international cartels. You see this in the news a lot: they buy cocaine paste or cocaine in the Andean countries and transport it through Brazil to Europe.

But along with that there is another dynamic, which is what I am studying, and it has to do with another transformation: a city’s favelas governed by a handful of large “faction” organizations, usually including the PCC and CV, but always based within the prison system. What was once a q unique characteristic of Rio is now widespread in Brazil. All major cities are divided between two, three, or four factions that are based within the prison system.

What’s interesting is that this expansion of this system of factional control over urban peripheries received a strong boost from mano dura and mass incarceration. Brazil is currently the third largest incarcerated country in the world and continues to increase its prison population. Only China and the United States have more people in prison. Today, if I am a young black or poor person from the favela, I have a high chance of being imprisoned even without doing anything, because the rate of pretrial detention is also high. So, I don’t even need to commit a crime to be thinking that at some point I’m going to have to live with the PCC, and thus realize that I’m going to need to be on good terms with them. So, the incarceration rate helps the groups inside the prison exercise criminal governance on the street. Or, to put it another way, the coercive power they have in prison is projected and transmitted to the favela by the state itself, because the state physically grabs the young people of the favela and takes them to the prison system. That is a fundamental, and at the same time highly counterproductive, dynamic of repression. Because the state says we’re going to invest in mass incarceration, we’re going to incarcerate all those bandits, and that ends up strengthening the criminal groups, helping them establish control on the streets.

-Incarceration, then, is like conscription.

-Recruitment and control. It allows organized crime to project its power outwards, because no one wants to have a bad time in prison. But this is not the whole story. Coercive power is fundamental to governing, but governance is not done by coercion alone. Governance is also made with the ability to bring order. These groups inside the prison –and this is the theme of my new book– arose in totally chaotic and violent contexts inside prisons, with lack of water, food, places to sleep; with murder and rape a constant threat. And those groups brought order inside the prison. They established a normative order: they prohibited robbery and rape, rape of inmates, but also of visiting family members. They set rules of coexistence and organized the inmate masses to pressure officials for concessions, improvements in meals, and an end to the physical and mental abuses of the guards.

The governance they exercise within the prison is fundamental to understanding how they grew so much. When members get out of prison and begin to project that power onto peripheral neighborhoods, they are already experts in consolidating and managing social order. Their identity is very much intertwined with the concept of providing order, of establishing a more just social order for the criminal underworld. That is why my book is called Criminal Leviathans, because the very identity of these groups recalls Thomas Hobbes’ Leviathan: a fearsome power that is used to impose an order that improves the lives of all its subjects. And these criminal subjects obey in part because of the gang’s coercive power, as we all mostly obey the State, in part, because of its power to punish. But they also obey the gang because its power actually improves their lives. They are better off together; they are better off without killing each other. They can sleep at night. They are not raped or robbed, and together they have been able to push back on the state. Or rather, by governing, they push the state out. They do this both inside the prison, where state agents withdraw from daily management of cell blocks; but also in urban peripheries, where police often withdraw to the perimeter of the favela territory ruled by the criminals. And it’s not that the state can’t enter. Obviously, it can enter the prison cells whenever it wants, and can usually enter most favelas. But the state doesn’t need to, because those spaces are well-governed. That’s the balance these groups have built. The state has no incentive to spend every day inside the prison or the favelas; it is much easier to “let them govern themselves.”

And that’s good for the gangs’ business: the state controls the perimeter, but there’s a flow of drugs, there’s a retail drug market, the police come in from time to time to arrest someone or to take their share of drug profits, but on a day-to-day basis doesn’t need to be inside the community. That’s where my argument goes. We want to understand the persistence of criminal governance, which in Brazil has been seen in urban peripheries for decades, this duopoly of violence in the heart of the metropolis. How can we explain this persistence, why does this duopoly not diminish over time, but expand? The answer is that it is in some way convenient for the state, because it is difficult to govern the periphery, and these groups that were in a sense born to govern –who are experts in governance– are governing the periphery for free, and the state does not have to pay them anything.

“The drug trade is stronger than ever. Prohibition and repression failed to make consumption rare. On the contrary, it made trafficking more lucrative, creating incentives for traffickers to do two things: arm themselves and bribe the police”

-I understand the calculation from the State: it benefits from the governance provided by criminals because it gets rid of the problem of having to govern complex places. But what about the population? Because if the population is doing well, one implication would be that we need better criminal governance, rather than a better state.

-That is a question that is as important as it is difficult to answer, and that has been present since the beginning of research on criminal governance. For example, Elizabeth Leeds has a very famous article published in 1996, in which she conjectures that the population of Rio’s favelas is better off with criminal governance because it protects them from police violence. It’s a conjecture because it’s hard to know. We don’t have a way to ask all citizens, or even a clear idea how to measure that. The citizens, I believe, and it seems to me that there is evidence of this, what they would like most is state governance. But fair, good governance, one that offers full citizenship. However, that’s not a realistic option. The choice is to have a partial presence of the state, a negligent presence that is not completely absent; a violent, discriminatory, corrupt presence. The police arrive, say, but they don’t solve the citizen’s problems. They come to extract money from the traffickers. It is an illegitimate presence of the state and against that, criminal governance does, in some way, improve people’s lives, or has the power to do so. That’s not to say that there aren’t variations between different areas. Criminal governance can be extremely violent as well. It can be “bad” and worsen the lives of its subjects. There is extreme variation from place to place and over time. I don’t think we can say once and for all it’s better or it’s not. However, we have to recognize that the worse state governance is, the more opportunities criminal governance has to improve people’s lives. And that is precisely what they are interested in: to govern civilians well in order to gain a certain loyalty from them, so that when police enter the favela, civilians protect them, or at least do not denounce them.

–Seen in this light, criminal governance would be a new manifestation of inequality. That is to say, there is state governance for the middle and upper classes, for neighborhoods where intellectuals, politicians, and businessmen live, while the margins of society remain under this other governance.

-I agree, it is a perverse system, it is an unjust system that condemns a good part of the population to live under a duopoly of violence that is not simple at all. Some have to navigate between what the trafficker tells them and what the state tells them through the police. And many of those who live like this are humble people, who just want to work, educate their children, try to build a future. The most difficult thing, from my point of view, is that the system of criminal governance is stable, as it has worked in some form in the urban peripheries of Latin America for decades, in some cases almost 50 years. Think about processes of migration and the construction of large cities, with large peripheries. There, people build their own houses and form informal communities, and in the beginning that is fine, there are community groups, local leaders, etc. But when that periphery grows and people start arriving from all over, providing basic order becomes a real problem for the state, and for larger processes of economic and social development. It seems to me that in some cases, the solution that emerged was to let criminal groups govern those populations.

Even worse, where there is drug retailing, not only do you have groups with the capacity to govern, but you also have immense fiscal resources for governance. They don’t even need to tax civilians. Elsewhere, criminal groups may charge a security tax to residents, businesses, shops. But where there’s drug retailing, they don’t need to do that. It’s like a country that has oil, it doesn’t need to collect taxes.

–What can you tell us about El Salvador and Bukele? This case is one of extreme punitivism, as it has implied a process of authoritarian drift, violating basic civil rights, but at the same time reaching unprecedented levels of popular support.

-The book I’m working on looks primarily at Brazil, but also at Medellín and El Salvador, because they are two places where criminal governance is deeply rooted, and we already have decades-long duopolies of violence in urban peripheries. And frankly, it’s hard for me to finish the book precisely because I want to know what’s going on with Bukele. Could it be that finally someone arrived with so much repressive force that he really managed to eliminate the maras? Could it be that the problem with all the previous «iron fist» policies was that they didn’t go all the way? Or will we see the same old pattern once again: a wave of mano dura, with unprecedented levels of state repression, and a few years later, the emergence of even stronger, better organized and more pernicious criminal groups? We don’t know.

We don’t even really know what’s going on now. Reading the experts in El Salvador, it seems that Bukele has really managed, at least in the short term, to change something. They say that there are no more maras in the neighborhoods and that citizens are happy because they can move freely, that there are no more invisible borders. And if it is true that he has put 60,000 people in prison, as Bukele says, then 1% of the population is imprisoned, which would mean that El Salvador now has the highest incarceration rate in the world, something truly unprecedented. Is it possible that with such extreme repressive force, never seen at this level, he has achieved something?

You have to look very carefully and see if this is really going to make a permanent difference. For me, a very important question is what is happening in prisons. Because Brazil also has its mini-Bukeles in some states, where they deploy lots of prison guards and very rigorously control every moment of the prisoners’ lives, to eliminate the power of the factions. There is even a new discourse, which shares with my argument the understanding that the root of the factions’ power is their ability to rule inside prisons, and which then says, «the state must govern inside the prisons, to take that function away from the factions.» That discourse, which sounds good, is being imposed with a lot of violence. There are many allegations of torture, which is difficult to verify, because authorities don’t let observers in.

Nonetheless, if it’s true that the state manages to govern inside the prisons and govern legitimately, not as an abuser, but legitimately, and not just for a day or to take a photo and post it on Twitter, but as a new equilibrium, it is possible that it could take power away from the maras. Still, I don’t see how Bukele is going to keep more than 60,000 people in prison and provide them with a decent life inside. I doubt very much that he has the wherewithal. Much more likely, in my view, is that over time these people will be abandoned and starved, left to kill each other. And there we return to the conditions that allowed the birth of the maras and the factions in the first place. Time will tell.

–A case that is talked about a lot is that of Medellín. Some authors say that the city’s low level of violence today is due to the fact that intensive policies associated with building high-quality state public goods were carried out, and in the worst places, to recover the ground that the gangs had gained. You have investigated this case along with Santiago Tobón, Chris Blattman, Gustavo Duncan, and a group of other colleagues. What would you say you can learn from Medellín and what do you eventually have to be more careful with?

-Yes, we are trying to understand what happened in Medellín, and also use it as a lens for understanding criminal governance more generally. In one study, we used a natural experiment that took place in the city in the 1980s, when district boundaries were redefined, causing exogenous changes in the presence of the state at the local level. To illustrate: imagine that today you and I live in the same district and that, if something happens to us, we both go to the same police station. But one day, a new demarcation separates our houses and now I have to go to a delegation that is farther away from my house, but you are still close to yours. Now, if you think that organized crime grows where the state is relatively weak or absent, the prediction would be that where I live, given that it became a little further away from the state, the gangs are going to proliferate more, whereas where you live, which remains relatively closer, the gangs are going to have less presence.

However, the opposite happens. Twenty or thirty years after the administrative changes, where I live (a bit further from the state) there are fewer gangs, and where you live there are more, or the gangs are stronger. That means, in our argument, being closer to the state in some way incentivizes the formation of criminal groups that govern or incentivizes criminal groups to govern more intensely. Now, what is the mechanism that makes that happen? Well, we’re studying it and trying to understand it. It’s a tricky issue because there are so many mechanisms that could be behind it. But the basic idea is that, for the growth of criminal governance, the presence of the state is just as important as its absence.

Personally, I would say that this dynamic explains another puzzle, which has to do with the question: where are the criminal groups? Again, the conventional wisdom would predict that criminal governance is stronger in places where the state is weaker. But not necessarily. Perhaps the criminal group that governs the most today is the PCC, and the base of its power is São Paulo, which is not a place of state weakness. On the contrary, it is probably the most developed place in South America. It is the largest city in South America, and its periphery is ruled by a criminal organization. This is anecdotal, but there are many other examples. I don’t mean to say that all strong states have strong criminal governance; obviously the problem is more complex. But it is increasingly evident that the idea that criminal governance and organized crime only arise where there is no state, simply does not explain the variation we see.

“The question is how to reintegrate the criminal world and its subjects into the formal and legitimate world in which we have the great privilege of living.”

–That’s an important message for politicians who often argue that this only happens where the state isn’t there. Is there anything you want to say to a Latin American citizen or a Latin American politician who is concerned about these issues that you think they should know?

-Well, the first thing I would say to politicians is that you shouldn’t fall into the temptation of mano dura (iron fist). Unfortunately, we know that wherever there are criminal organizations and violence, that temptation arises. But you have to know that this often only strengthens criminal groups. And the politician who promotes that iron fist doesn’t pay the cost. The politician has already been president, and when he sees that it doesn’t work, he says «we have to fight harder.» Meanwhile, politicians who try other approaches often do pay costs: they are accused of negotiating with traffickers, with criminals. So, I don’t know what to say to politicians, except that, if they want to leave a lasting legacy and improve the situation of their citizens, they should not fall into the temptation of the iron fist.

As for the citizens, I would say that we have to understand that criminal governance and the coercive power of the state are a whole, an organic being that is split in two. It is an ecosystem in which the two are growing together, they are co-evolving. Both are manifestations of social problems and drug prohibition, which is a mistake of monumental scale. Drug use is a basic human activity that cannot be eliminated by coercive force. And if the state criminalizes a basic human activity, it will obviously create criminal organizations and stimulate criminal governance. It’s like criminalizing religion: you’re not going to eliminate religion; you’re just going to criminalize a good part of humanity. And that’s what’s happened with drug markets. Today we have a thing that is half state, half criminal group, and that thing, those two halves together, is what governs the citizens. So, it seems to me that the problem is not how to eliminate the crime of trafficking, but rather how to think about a future in which those two things – governance and coercion – are not so antagonistic. The question, I think, is how to retain and bring those two sides together and stop forcing the poorest half to live under low-intensity citizenship. The point is how to reintegrate that criminal world and its subjects into the formal and legitimate world in which we have the great privilege of living.

-In this sense, what are the questions that should mobilize academic research on crime in the coming years?

-When I wrote my first book, I didn’t have a silver bullet in terms of policy, but I did still think there was a clear way to try to minimize violence. But now, examining the idea of criminal governance, I don’t have an easy suggestion to make, at least in the short term. I think that’s because this governance is something that took decades to build: these groups have been learning how to govern and perfecting their criminal governance techniques for decades. It’s generations of people who are born in the peripheries under criminal governance and generations of police officers who accept that as part of life. So, in my view, there’s no policy that’s going to reverse the situation quickly. It’s a very deep problem because at the end of the day it was a model for economic and social development. A model that has been little talked about. No country says, ‘My development model is to let criminal groups rule the periphery,’ but in practice, it has functioned as a development model, just as import substitution was. That’s why I think we need to look more at criminal governance. We need to know more about how it works and how it varies from place to place. And if it’s growing, where it’s growing. And if there are cases in which it has decreased, to know how people managed to reduce it. That needs to be studied in more depth. Today our censuses don’t ask: do you live in an area controlled by a criminal group? However, there are millions of Latin Americans who live this way.