On Colombia and Mexico, for the Millionth Time

Francis Fukuyama and Seth Colby have a piece at Foreign Policy looking at the so-called Medellín miracle. It is not a simplistic piece, which many in this genre are, and it does acknowledge the spikes in violence over the past couple of years, but it does seem to me to be conjuring a miracle where, if there ever was one, it has disappeared. (Plus, it has a number of facts wrong.) Take the various comments on violence in the city:

In 1991, the city had an astronomical 381 homicides per 100,000 residents (by contrast, the murder rate in Ciudad Juárez, the bloody epicenter of Mexico's drug war, was only half that last year). But today Medellín has, incredibly, become as safe as Washington.

[Break]

The harder lesson here, however, is that there are no quick fixes in a drug war, and two steps forward are often followed by one step back. After bottoming out in 2007, Medellín's homicide rate has since doubled (though it is still one-fifth of what it was at the city's early-1990s nadir).

But by their accounting, that's a murder rate of roughly 80 per 100,000 people; that's really violent! (That's also not nearly as safe as Washington, which had a murder rate of just over 20 per 100,000 inhabitants in 2010. Not sure where that assertion came from.) I wonder if the mistake in this approach is assuming that the Medellín of the early 1990s was in some sense the natural state of the city, and anything better than that indicates some sort of success. If so, I disagree. Escobar's Medellín was an anomaly of violence. It may be better than that now, but we're talking about a city in which almost 3,000 people were killed in 2009 (good enough for a murder rate well above 100), which in 2010 was one of the three most violent cities in a particularly violent country. The city is certainly worth examining and there are surely some lessons to be gleaned, but the "miracle" label is plainly ill-fitting.

As far as Mexico, a couple of points: First, Juárez had 3,951 murders in 2010, according to state authorities. It's hard to pin down an exact murder rate because it's hard to pin down an exact population, but with 1.3 million (a common, pre-violence-induced exodus figure), that would be roughly 300 per 100,000 residents, which is to say, well above the figure of 190 alluded to in the article.

Second, the prescriptions:

That said, there are useful lessons Mexico can draw from the Medellín miracle. Like Colombia in the 1990s, Mexico is vastly underpoliced and has a weak judiciary, problems that can be solved in time with sufficient resources and will. But Colombia's drug war shows that the battle will not be won by military force alone. The government needs to bolster its legitimacy by offering people alternatives to crime and violence, as well as a renewed commitment to public services -- something Medellín's metro, its starkly beautiful new buildings, and civilized public spaces now do.

As indicated above, Medellín's miracle really isn't one. Nuevo León, for instance, would have to get a lot more violent to follow Medellín's miraculous path. Furthermore, Mexico is not, at least according to some sources, underpoliced; it's problem is competency of police forces, not the quantity of officers. I completely agree that Mexico needs a more effective judicial system and better police, but I don't see what special insight Colombia provides as a model in that regard. Every country could use a more effective judiciary and better police.