Violence and Productivity Growth

Between 1996 and 2006, there have been almost half a million victims of homicide in Brazil. Last week’s widely reported “Map of Violence” reveals an increase in the crime for Brazilian cities with relatively low incidences of violence before the mid 1990s, such as Florianópolis. For cities with higher incidences of violence, a small fall occurred in the rate of homicides after 1999.

Even in São Paulo, where the rate of this particular crime fell significantly, there is still little to celebrate. Its current rate of 31 homicides per 100,000 inhabitants remains higher than that of Bogotá – a city notorious for the violence stemming from the drugs trade and, especially, guerrilla groups financed by cocaine. Bogotá’s homicide rate reached 80 per 100,000 in 1993, but by 2004, it had fallen to 21 per 100,000. Today, it is safer to walk the streets of Bogotá than those of São Paulo.

The homicide rates of Recife and Vitória, 90 and 87 per 100,000 respectively, are among the highest in the world and are comparable to those of countries devastated by civil war. In addition to the lives lost and the public’s safety fears, this criminal activity has an economic cost resulting from expenditure on security guards and protective walls surrounding factories, shops and condominiums.

The cost of safety multiplies because, by shunting investment from roads and tractors into prisons and armored cars, economic activity is transferred from productive works to non-productive ones. Consequently, productivity and sustained growth plummet. This history is documented in the work of Maurício Cárdenas and Sandra Rozo regarding crime in Colombia (

Between 1950 and 1980, Colombia (like Brazil) enjoyed high rates of growth. In the 1980s and 1990s, this growth disappeared. The increase in violence, linked with the growth of the drugs trade, explains the collapse in productivity growth. The evidence collected by Cárdenas and Rozo shows the preceding increase in violence and resulting fall in productivity. This reduction in growth then incurs a further increase in violence. The country enters a vicious circle of violence with low growth. Social pressures and discontent then become prevalent.

In Colombia, as in Brazil, the answer was a new Constitution that tied revenues to expenditure and left fiscal policy strict and procyclic. Inefficient taxes were created, whilst the lack of investment in infrastructure helped to shake the foundations of sustained growth.

Colombia’s society, unlike Brazil’s, managed to halt the escalation of violence. Brazil regained its rate of growth in 2004 thanks to a favorable international context. But its inflexible budgeting and political interests prevented it from taking advantage of these circumstances to reduce violence, which continues to harm potential growth.

5 Responses to "Violence and Productivity Growth"

  1. mark turner   February 11, 2008 at 12:42 pm

    Half a million in 10 years? That’s 50,000 per year on average!Is that true? If so, wow!

  2. Eliana   February 11, 2008 at 3:36 pm

    Yes. it is terrible

  3. Mary Stokes   February 11, 2008 at 9:28 pm

    Thanks for highlighting these important issues! The impact of violence on Brazil’s economy seems to be overlooked by many economists, especially those abroad. Living in Sao Paulo for a time and hearing from friends about the PCC gang’s virtual shutdown of the city in May 2006, I’ve frequently thought about this problem. I also remember reports of the Comando Vermelho gang’s shutdown of certain areas in Rio.Violence, along with Brazil’s lack of investment in infrastructure, are major social and economic issues. Why aren’t these issues getting more attention from economists?Beyond the increased cost of safety that you point out (which translates into an increased cost of doing business in Brazil), it seems like this type of violence has contributed to a brain drain to some extent. And while difficult to estimate, I’d proffer that it’s also limiting long-term investment in areas beyond just infrastructure (like tourism, for example).It seems like income inequality and lack of investment in education and opportunities are some of the reasons for this. Is there anything that can be done policy-wise to reverse this cycle you point out?I think Brazil (and the harmful impact of violence on its economy) provides a stark reminder to economists that there are costs to policies that promote and continue income/wealth inequality.

  4. Anonymous   February 12, 2008 at 9:00 am

    A better justice system would help to get Brazil in a better situation in all aspects. While people see that most crimes and corruption don’t have any punishment, many keep doing it. Of course I don’t agree with these people logic, but it’s their “economic” point of view: steal, get money and don’t go to jail. Easier than working 12 hours a day. It’s a pitty… And with “steal” I also mean the corporate corruption that doesn’t help companies to compete in a fair environment.

  5. Anonymous   February 22, 2008 at 1:21 pm

    I live in Rio. Where I live you can hear gunshots a few days a week at the moment. Recently one young man I knew died from gunshot wounds – he was killed by the police. He was a member of one of the drug-trafficking factions in the city. Maybe his story helps unpick some of the economic issues touched on in comments above – and I think this story is not unrepresentative. This guy joined the faction for economic reasons, he had family to support, medicine to buy. With minimal education, his choice was working for peanuts (try living off R$500, or less than $250 a month) in a 12 hr a day menial job, or possibly several times that selling drugs. There are also non-monetary considerations: as a trafficante he would have respect (we are talking about a community where the police enter only to kill, the state provides almost no services, and the faction in fact provides law and order much more efficiently inside the community than the state does outside), not to mention a certain allure with the girls. Working in the formal economy he would just be another semi-literate black skivvy. Against this you have to weigh up the extremely high likelihood of dying if you enter the traffic. Gang members can expect to die in their teens. Brazil certainly needs a better justice system. But it is by no means the case that low class criminals in Brazil “get away with it”. Prisons are horrendous, and many are killed by the police before getting anywhere near a prison. Police killings account for a good number of the violent deaths (I am guessing police killings are not counted in the homicide statistics – if we add them in we’d go well above that half million) – e.g. just in Rio state the police killed 1063 civilians in “acts of resistance” in 2006. (Source – CESeC — Centro de Estudos de Segurança e Cidadania. I would suggest the numbers are higher if we start to distrust police data.) So the decision problem for a guy like this: a menial badly-paid job and a reasonable life expectancy, or a high-status better-paid life with high risk of imminent death. The fact that so many kids like him choose the second option – and suffer the consequences – is an indication of the extreme poverty of opportunities open to kids from Brazil’s lower strata.