David R. Henderson  

More Gangs, Less Crime

What I'm Reading: 2... The Triumph of Organized Labor...
Our analysis suggests not that gangs cause violence, but that violence causes gangs. In other words, gangs form in response to government's failure to protect youths against violence. The surprising implication of our insight is that efforts to reduce gang activity could actually increase violent crime.
This is from the September Feature Article on Econlib titled "More Gangs, Less Crime," by West Virginia University economics professor Russell Sobel.

I live in a city near Salinas. Salinas has a substantial gang membership. The local government plus the state and federal government are going after gangs in Salinas. But it's not clear from the news reports whether they're going after gangs per se or after crime that gangs participate in. To the extent the various police are going after gangs per se, it will be interesting to see whether crime increases.

Comments and Sharing

COMMENTS (24 to date)
Prakhar Goel writes:

The decrease in violence that the researchers notice from gang formation is merely the transformation of theft into taxation. The actual amount of objectionable behavior has not decreased --- it has only gained a thin, very thin, veneer of civilization. This is exactly the process by which pirates and bandits turn into government officials.

An improvement, maybe. Acceptable in lieu of a genuinely effective police force (the kind that this country used to have considering the crime rate of, say, 1930 or 1950), absolutely not.

David R. Henderson writes:

Prakhar Goel wrote:
"The decrease in violence that the researchers notice from gang formation is merely the transformation of theft into taxation." Please explain; I don't understand.

size4shoe writes:

I agree. I understand that many children in gangs lack a safe and healthy home environment. Without a home, children wander the streets with only the government for protection. When government ceases to protect children, gangs emerge.

Prakhar Goel writes:

@David R. Henderson

Explanation: A gang comes into town (or is formed). They go to shops and demand protection money. Some shops refuse and are punished. Eventually, the people get the basic idea and there are no more (or at least, fewer) detractors. Along the way, some gangs may take steps to protect their territory to prevent the entry of freeloaders/other gangs. Voila, we have a reduction of violence. Theft is when the shops do not understand the implicit rules of the game. Tax is when they do.

If this sounds like the formation of a government, that's because it is exactly the process by which many (though of course not all, some inherit) form. The one thing that separates the sovereign of a nation from a common gang is that the sovereign is the biggest gang, the only major gang, is publicly recognized, and has some non-trivial expected survival time.

The reason I find gang formation objectionable is because it results in a drastic change of incentives. With only one gang, a sovereign, the incentives point to value (and possibly tax revenue) maximization. Not perfectly ideal but a sovereign has no incentive to tax above a certain laffer maximizing level and they have an incentive towards efficiency. With multiple gangs we have an n-person's prisoner's dilemma. If gang A doesn't tax an individual, gang B will and so we end up with what is at best an unstable equilibrium.

What's particularly annoying here is that if we were to limit our consideration to availability of resources, it is blatantly obvious that the US government could eradicate gangs permanently. Even gang activities with high profits like drug trafficking and hostage taking can be eradicated with the resources that the US government has at its disposal. With the performance of the US government during reconstruction after the US Civil war and in the Philippines, along with the actions of the USSR in the warsaw pact states, the performance of the French paratrooper's in Algeria against the FLN, and with General Fonseka against the Tamil Tigers, we have plenty of evidence that governments can eradicate entrenched gangs and that the US government in particular is (or at least was) capable of eradicating entrenched gangs/guerrilla terrorists.

That the US government allows such gangs to flourish is a sign of shortcomings in the structure of the US government and an indication of the incompetence of the people employed by the US government (which deserves more blame, I have no concrete evidence towards). This problem is fixable and yet is unfixed.

PS. I am not sure I answered your question as all of the above is just intro political science and history which your posts indicate that you are familiar with. Could you elaborate?

David R. Henderson writes:

@Prakhar Goel,
Yes, you did answer my question. You're assuming that these gangs shake down local merchants, etc. I don't know that that's true. And, by the way, intro poli sci and history don't establish that it's true. You seem awfully a priori. I'm an empiricist.

FC writes:

The empirical thing to do would be to ask your local government for the data rather than us.

Ricardo Cruz writes:

Provocative theory! ;-) Deserves deeper empirical work done.

Ed Bosanquet writes:

Very interesting theory. If this is true, does that mean the way to combat gang violence would be to provide protection and safety to citizens. That way no one would be compelled to join a gang.

This article seems to be isolated to violent crime. I wonder the correlation to other crimes. Black market profit would seem to increase violence since you can't complain to the police about your drugs or drug money being stolen.

It always seems to me cutting the head off a criminal organization only made things worse because without control splinters are much more free to innovate. A leader would want to limit innovation to maintain his control.

I would love to see more research in this area.

James writes:

D. Henderson:

What do you mean when you point out that you are an empiricist? Are you indicating that, like nearly every other human who ever lived, you believe that knowledge can be had via the senses? Or do you intend that knowledge can be had only via the senses?

Either one is just an autobiographical non-sequitur but while the former is simply boring, the latter sounds an awful lot like, "I disregard whole classes of arguments according to some unarticulated and undefended theory of knowledge and I expect other people to accommodate me on that."

So far as I know, empirical arguments are neither more nor less likely to be fallacious than other types of arguments. Given that you neglect to provide any evidence that empirical arguments are more credible than arguments from first principles, arguments from abductive reasoning, arguments from introspection, arguments combining multiple classes of evidence, and so on, do you think people should do anything differently when you announce that you are an empiricist?

Fundamentalist writes:

I envy you where you live Dr. Henderson. I attended the Monterey Institute of International Studies many years ago.

David R. Henderson writes:

I meant nothing more than that I can't say a priori, when looking at gangs, whether they will engage in extortion. Theory alone is not strong enough to resolve the issue.
It is a beautiful place. Did you get a Masters? What in? When?

Prakhar Goel writes:

@David R. Henderson

a priori? Most of my post is definitional. My distinction between tax vs. theft is entirely a matter of definition. The consequence of reduced violence when gang activity becomes "normalized" has been observed throughout history (see the formation of Normandy). I used protection rackets only as a convenient example. Nearly any illegal activity includes a much larger amount of violence when there isn't some organization capable of making long-term commitments and enforcing standards (pardon my vocabulary as it is a little out of place when discussing gangs). When there are fifty street dealers dispensing heroin, a potential buyer has no idea what the punishment for stealing will be. Drug users steal for their addiction and this shows up as violence when the street dealers beat up the thieves. Replace those with five to six organized gangs and these gangs can establish a reputation for "fair dealing" (let us ignore for the moment the question of whether the drug trade should be legalized or not. My example could just as easily have concerned human trafficking). My point is that greater organization (which is ultimately what a gang is) generally replaces violence with long(er)-term commitments and reputation signals. It does not eliminate the root crimes which were the cause of the violence.

Consider the case of kidnappings --- a tactic often used by the mexican cartel. With a large number of individual kidnapping groups, the gangs are usually disorganized and often cannot be relied on to return their hostages. With larger gangs, they usually take steps to protect the identities of the kidnappers (thus protecting the hostages) and can establish a reputation for whether they will return hostages when they get the ransom. Negotiators know this and can plan for it. Thus "violent kidnapping" gets replaced by "civilized kidnapping" but kidnapping is still kidnapping.

My assertion that the US government is capable of eradicating gang violence is backed by multiple well documented cases of governments eradicating entrenched criminal organizations with far fewer resources than what the current federal government has at its disposal.

My assertion of the change in incentives is likewise easily verified from history. How often do the states and the US federal government worry about cutting into each other's tax base? Favors for unions are a classic case of a lack of a unified sovereign resulting in higher than optimal rents. Finally, roving bandits have been observed throughout history with the same result: over taxation of individuals. If you want concrete examples, consider the vikings or Ireland before the English occupied the country.

As to my last paragraph, in carrying out any activity, there is the question of availability of resources and the ability to use those resources. When the US government is taken as a single whole, removing gangs is clearly in its favor. If gangs aren't removed, there must either be a lack of resources or a lack of the ability to use them. The former is not the case as I have already demonstrated. If the incentives of the individuals employed by USG are not the same as USG as a whole by such a drastic amount so as to allow gangs to flourish in nearly every major city, this is clearly a failing of the structure of USG. If the employees of USG are not properly able to use the vast resources available to USG, that is likewise a failing of the structure of USG. If the employees of USG can not eradicate gangs despite having the right incentives and the right resources, then clearly they are incompetent.

Where exactly am I lacking in empirical evidence?

Liam writes:

@Prakhar Goel:

I disagree that the US Government has the ability to eradicate gangs. The examples you cite are military actions against militant groups yet the US Military is rife with gangs. If the US Government did have that ability then they would start with their own military.

However there were Governments that were very able to control gangs and when those Governments changed ideology then gangs (such as mafia) quickly emerged. I am speaking of course of the former Soviet Union. So would you support that the Government crack down on the gangs using similar tactics?


You seem to be making a number of assumptions towards David. He said he's an empiricist but never claimed that it made his arguments more credible or that he disregards other's argument or opinions. You peg him as either boring or close minded which ironically seems rather close minded.

NZ writes:

Ever read Gang Leader for a Day by Sudhir Venkatesh? Although it's basically an anecdote, I found it very illustrative.

Hyena writes:

The experience of Los Angeles, so far as I can tell, is that gang violence is multifaceted. However, one feature of gangs is that they enable asymmetric capabilities. Violence against a gang member or someone close to one can bring swift and certain violence in response. This behavior pattern is already observed, so it might provide a mechanism for the effect.

The other issue is that people conflate gangs and gang warfare. Much of the gang violence is the result either of leadership collapse (often due to prison) or external competition. A couple of people have noted this as a reason why a lot of efforts to reduce gang violence failed.

size4shoe writes:

To reduce gangs, some groups have applied the theory the article references by providing at risk persons with pratical services, such as a homeless shelter, youth center, a job, a different (more socially acceptable) sense of community and belonging, transportation...

David R. Henderson writes:

Prakhar Goel asks:
"Where exactly am I lacking in empirical evidence?"
My answer: In your first comment, the one that I responded to.

Chandran writes:

On the thesis that government arises out of the stabilization of gang activity, see Charles Tilly's famous 1985 article, War Making and State Making as Organized Crime, which can be found here:


The observation that governments are the biggest gangs of all is not new. St Augustine notes as much in The City of God (ch.19). David Hume implies as much when he says that "camps are the mothers of cities", though he has in mind victory in warfare as the source of political authority.

More recently published, I would recommend James Scott's The Art of Not Being Governed, though it's more a study of state encroachment than of state formation.

fundamentalist writes:

Dr. Henderson, I just spent a summer in intensive Arabic. No degree. But I often dream of Monterey!

David R. Henderson writes:

Thanks, Chandran. And thanks for your hospitality when I was in Australia in 1999-2000.

James writes:

David: I've never before encountered the sentence "I'm an empiricist," used to communicate a position on nothing more than the availability of knowledge about gangs. I had mistakenly thought you were stating a very general philosophical position, consistent with the more common use of the term.

Rebecca Burlingame writes:

Prakhar's explanation helped me out a bit, when I first read the post all I could think of was a dog chasing his own tail.

Prakhar Goel writes:

@David R. Henderson,

Ah yes, that one was a bit sparse. I tried to be brief as it was a comment and not a blog post. Perhaps I have spent too much time reading two hundred year old books by professional historians and omitted some critical details (those things are incredibly long and now with Google books, entirely too much of a distraction).

Do you think there are any other areas that I should flesh out?

Daniel writes:

What do you guys think will happen to the economics of crime in Arizona due the legislature SB 1070?

Comments for this entry have been closed
Return to top