Thread by @cblatts

Continuing the tradition of real time field work tweets:

This is Medellin, Colombia’s second city. Industrial & commercial hub. Also.... one of the densest and most hierarchical systems of organized crime in the country.

We’ve been studying the structures for over three years.

https://twitter.com/cblatts/status/1091381463969234944

The city holds ~400 street gangs called “combos”—groups of 15-50 armed young men who sell drugs, collect “security fees” from residents, and provide a degree of governing in some neighborhoods.

We’re completing a census of them. This is our best approximation of territories.

Nearly every low & middle income neighborhood has a combo. The borders are invisible but well known.

Some combos govern & hold legitimacy in eyes of residents.

Here’s a scatter plot from our survey of 7000 citizens. Averaging across 15 kinds of local gov, 0=Never 1=Always

The average reflects the degree to which a combo intervenes in disputes, polices neighborhood, stops domestic violence, helps collect debts, resolves noise complaints, etc. In these everyday kinds of problems, combo governance is about as common as actions by the city or police.

Today we’re visiting a neighborhood with one of the highest reported levels of combo governance in our sample. Lower-middle income.

The unofficial governors of this territory are unusual in that they’ve established a non profit organization that helps develop and maintain local infrastructure and work with youth.

They say that they have shifted out of drug dealing and begun selling yogurt, reclaiming scrap metal, and collecting recycling in the community.

They even built a community pool. They must sell a lot of... yogurt.

(Photo credit to me, but joke credit to my collaborator @BigBigBLessing)

It seems the city thinks the pool is not up to code. They want to shut it down. Not the best way to endear the community to the state...

Their organization supports work by local graffiti artists.

Some of this group’s activities are supported by cooperation and grants from the city and foreign donors.

It’s reminiscent of attempts in late 60s and early 70s in Chicago, when foundations tried to turn the city’s most powerful gang nations into community development orgs.

A famous example is the P Stones, who (as it happens) were born and operated just a few blocks from @UChicago. Vice Lords are another example.

Those gangs never exited the drug trade. In fact the organizations exploded in size and power with the 70s and 80s drug boom.

Critics call the foundation support naive and even counter productive—professionalizing and strengthening the gangs, giving them the veneer of a social mission, and facilitating their rise to power.

You could make the same argument here in Medellin.

At the same time, it’s impossible to deny the social power the gangs of Medellin possess. When they organize & establish hierarchies (as in the neighborhood we are visiting today) they can bring order and an end to conflicts & violence. Indeed, violence is way down in this area.

It’s a terrible dilemma for successive city govts: respect the power or even collude with the gangs, and bring down the homicide rate? Or arrest leaders & persecute gangs, & risk a spike in homicides?

Here’s Medellin’s homicide rate over time. A hint at what they’ve chosen.

All this is ongoing joint work with @SantiagoTobon @BigBigBLessing @gusduncan @MAranzazuRU and Pierre-Luc Vautrey, with help from @poverty_action and @EAFIT, including @JuanPabloMesaM and other amazing staff

Stopping for lunch on the street. The fish arepa was delicious. The deep fried hamburger one... not so much

We wear vests for security reasons that are somewhat unclear to me but make my research coordinator @JuanPabloMesaM happy.

I feel like a church group.

The @poverty_action cult

Local cult leader spotted.

Headed up to one of the highest mountainside barrios with special guest @MarakujaDRC

Maybe it’s a bad idea to put your entire project team in a single metrocable car.

Some very classic cars

A new project with Pierre-Luc and @MAranzazuRU is trying to understand gang entry among teenagers. Yesterday we visited the juvenile jail for the initial qualitative part of the study. It’s incredibly challenging in so many ways to interview children.

It is one of the hardest parts of the work. And one that is chillingly reminiscent of my dissertation work with child soldiers in Uganda and later Liberia.

chrisblattman.com/documents/rese…

chrisblattman.com/documents/rese…

To add to the depression, the jail is in one of the worst parts of the city. Surrounded by homeless persons and drug users.

Security hawk @JuanPabloMesaM tells us only to bring a notebook and pen and no wallet or phone. The night before he shows us the homicide heat map for the city. The center of the big red dot happens to be where they put the juvenile jail.

Surely middle aged wealthy people on a Friday morning face little risk? Or so we joke.

@JuanPabloMesaM sends us the homicide data for that demographic and time period for the past year. Again, it’s the brightest spot.

Here we are at a lookout in Comuna 1. 3y ago here, we were tailed by combo members listening in on us. Today it’s filled with tourists coming to see the famous library, perennially under construction.

You wouldn’t know this is the barrio with the highest combo gov in our data.

This is the Antioquia equivalent of American Gothic

@SebastianANicol displaying that he has more skills than just Stata and R

For those wondering about human subjects protections, I’m not exaggerating when I speculate that the @UChicago social science IRB spends about a third of their full board meetings just on these projects. My deepest gratitude to them as well.

SITVA violations
#ThingsEconomistsFindFunny

Thanks to donors @DFID_RED_GCSD @NSF @JPAL @Proantioquia1 for taking a chance on a very unconventional proposal