Graphs I've Long Wanted to See

Via Boz, Keith Humphreys has put together a couple of pie charts estimating the proportion of Mexican gangs' income that comes from different sources:

Like Boz, I imagine the second is probably closer to the truth. Humphreys writes:

The most difficult thing about getting current revenue streams correct is estimating how much the MOCOs are making from non-drug-related activities (e.g., smuggling immigrants across the border, pirating videos, kidnapping, extortion). Everyone I read or talked to agrees that this line of MOCO revenue is growing, but estimates of how large it is were divergent. The first chart here uses the low-end estimate among experts of 15% non-drug revenue. If this estimate is correct, the MOCOs are still largely one dimensional drug trafficking organizations, with principal profit coming for cocaine. Note that were we mapping drug markets more generally, cocaine would be a much larger share of the pie but because much of that money is made by Colombian gangs, the share for the MOCOs is smaller.

[Break]

A minority of experts (e.g., Sylvia Longmire, Edgardo Buscaglia) think that the MOCOs make a much larger share of their money today from non-drug sources. This second chart presents the revenue breakdown if these experts are correct. You can see the differing policy implications, for example even national methamphetamine legalization in the U.S. wouldn’t put much of a dent in MOCO revenue.

It's worth noting that the enormous ranges between the two graphs, combined with the enormous ranges regarding the gross total of revenues from each activity, make this little more than a visual exercise, not something on which you'd want to base policy conclusions. I also think that rather than the industry as a whole, it would be more interesting to look at different gangs' revenue streams. Again, reliable info would be a huge problem, but I suspect we'd see a much larger share of the non-drug revenue from the Zetas, and a much smaller one from Sinaloa. Such a case-by-case approach would make the information a lot more useful in terms of combating specific groups.

Lastly, I don't expect the term MOCOs to catch on in Mexico, for obvious linguistic reasons, but it would be kind of funny if it did.