Will Mexico be nonviolent in 25 years?

 I just got back from a short and ridiculously busy trip to Mexico. (Next time, I want to make more unprogrammed time.) On the trip, of course, I spent some time thinking about the country’s future. Alejandro Hope thinks that by 2040, the place will have remarkably low crime rates. His reasons come down to three:

1. Mexico is aging;
2. Technology is disintermediating criminal markets;
3. The rising middle class will demand better public policies.

I agree with the third point, although the cynic in me has to admit that Mexico has spent many years failing to live up to the demands of its electorate.

The second point rather confuses me. It is true that the opportunities for nonviolent crime are increasing. It is also true that the internet makes it possible to sell narcotics retail in new ways ... but so did the beeper. (I remember New York in the 1990s.) Moreover, the wholesale shipments will still have to move, and in Mexico it is the wholesale trade that attracts the violence. So I am not sure why technology should reduce crime in this way. (As opposed to reducing crime by letting the authorities surveille much more. I am not sure, however, that crime-reductions engendered in that way will increase human happiness.)

As for the first argument, well, Mexico is aging, but it is not aging that fast. For example, in 2050, Mexico will have 27.9 million people aged 15-29. That will be down only a bit from 30.7 million today. Given that most violent crime is committed by and upon members of that age cohort, the incidence of violent crime will decline as young people decline as a percentage of the population ... but the absolute amount of mayhem will not fall by much as a result of population aging.

In addition, aging is not as good predictor as people might think, since age-specific violence rates rarely remain the same. Venezuela, for example, has become one of the most violent societies in the world over the past 30 years as the population in the 15-29 age group fell from 28% to 27%. That is admittedly a less dramatic proportional fall than the drop from 26% to 19% that Mexico will experience in 2015-50, but it should make us doubt predictions that demography is destiny.

Finally, Diego Valle did a great look at levels of cohort violence in Mexico. He found, as you might expect, that violence decline with age and by cohort: i.e., each group of (say) 25-years-olds was less violent than the one before. Until you reached the group born in 1985-89, the proportionately-smallest group in Mexico’s modern population history. At which point their propensity to violence exploded. In other words, levels of cohort violence are not driven by the relative size of the cohort.

In short, I think Mexico will likely be less violent in 2040 than in 2016. But I do not think that it will be due to technological changes in illicit supply chain management or population aging. 

Comments

[Randy McDonald]

Crude demographics ultimately mean very little. The size of working-age cohorts in eastern Europe might have fallen since the 1980s, but that shrinkage has nothing to do with the propensity of working-age eastern Europeans to migrate. At most, the numbers of eastern Europeans at all are smaller.

[Noel Maurer]

Hi Randy! I'm not 100% sure what you're getting at with the comparison. Could you explain a little more?

[Randy McDonald]

A smaller youth bulge may well contribute to falling crime rates, even in a non-linear manner. I would still argue that cultural changes, and economic changes, will have more of an impact.

[patrick]

If you buy the lead-as-driver of crime theory then Mexico should get a boost there. Don't quote me but I believe they banned leaded gas in like 1998.

[Noel Maurer]

Patrick: hmmm. Maybe I should see what kind of numbers are available on that for Latin American countries. Was the recent crime explosion across the hemisphere south of the Rio Grande driven by lead exposure, or does it punch some holes in the theory?

[Noel Maurer]

Randy: gotcha! That's a good example. Other factors swamp these effects. For example, violence continued to dramatically increase in the United States between 1970 and 1990, while the population in the 15-29 age group remained roughly stable at 24%. (It is now 21%.)

[Randy McDonald]

Could you E-mail me?

[patrick]

Noel not sure, I'll have to check into that. I expect Central America will punch some holes into the theory.

[Alejandro Hope]

Noel,

Thanks for your insighful comments. I offer two rejoinders:

1. The technological argument was mostly about the relative decline of cash-based transactions over the long run. My take is that in a society with little cash, some forms of crime will become almost impossible (robbing banks at gunpoint), others will become more difficult (kidnapping for ransom), and some will become far more prevalent and lucrative (e.g., identity theft). I speculate that the resulting crime mix might end up being less gun-intensive and more brain-heavy, while also extending the physical distance between victim and offender.

About the drug markets, this is my super speculative take: the rise of online markets might be able to cut the middlemen, even at the wholesale level. A small cocaine refiner in Colombia might be able to get in touch with a small-time distributor in New York and then find a way to send him the stuff without going through large-scale drug trafficking organizations. How? Licit trade, the postal service, a relative flying to the US, etc. The distributor would then pay via Bitcoin or other cryptocurrency. After all, any given producer would be sending small quantities that might be more difficult to detect (Here's an interesting factoid in support of that possibility: in Australia, something like two-thirds of all drug seizures take place in the postal service). If that happened, then Mexican narcos (middlemen in nature) would be gradually frozen out the picture. I may be way wrong, but I don't think the idea is crazy.

2. Yes, my demographic analysis is way too crude. Propensity for violence might change between different cohorts of the same age-group. But I do think there is something to the idea that gradual aging of the population might produce violence-repressing tendencies. I might be wrong about the timeframe (maybe this is something that plays out over centuries and not decades), but I would be surprised if a society with far less young men is not somewhat more peaceful.

Anyway, thanks for reading and commenting my stuff.