The Mysterious Mexican Drug Guns

[Pangloss]

I've been following this for a couple of weeks now hoping to see if the truth leaps out from one sector or another, but so far it hasn't so I thought maybe it might be fun to bat it around here and see what we can come up with when we add a lot more smart people to ponder search parameters and overheard stories. Let's have a look, shall we?

 

The presumed fact is as follows: "90% of all guns used by the Mexican drug lords are acquired in the United States."

 

That factoid has been widely reported since mid-march, and spiked again in the last couple of days after it was again repeated by Mexican President Calderón during President Obama's trip to Mexico:

 

http://features.csmonitor.com/politics/2009/04/16/obama-talks-guns-immigration-in-mexico/

 

Calderón asserted that during his two years and four months in office, more than 16,000 assault weapons have been seized in Mexico – and that 90 percent of them have come from the US.

 

That 16,000 figure is important, as you will see in a moment. Back in March that number was being reported as 12,000 -- I guess Mexican law enforcement officials have been busy.

 

Now, hang on to your hats, because this is where it gets interesting. On March 26th CNN published this op-ed piece by NRA CEO Wayne LaPierre:

 

http://edition.cnn.com/2009/WORLD/americas/03/26/lapierre.guns.mexico/

 

In the piece LaPierre states that the 90% figure is wrong. And his source for this claim seems to be a pretty good one: The Bureau of Alcohol, Tobacco and Firearms, the very people charged with tracking exactly this sort of information. According to his opinion piece, BATF assistant director William Hoover, in answer to Senator Diane Feinstein's direct inquiry about this figure, answered: "The investigations we have, that we see, for firearms flowing across the border don't show us individuals taking thousands of guns a day or at a time flowing into Mexico."

 

Apparently Feinstein chose not to believe him, because she's one of the people who's been quoting the 90% figure since that hearing. Why? What evidence has she had to the contrary? I have no idea. Apparently neither does anyone else.

 

After LaPierre's op-ed piece was run, Fox News decided to chime in, running this piece on April 2nd, entitled "The Myth of 90%":

 

http://www.foxnews.com/politics/elections/2009/04/02/myth-percent-guns-mexico-fraction-number-claimed/

 

You've heard this shocking "fact" before -- on TV and radio, in newspapers, on the Internet and from the highest politicians in the land: 90 percent of the weapons used to commit crimes in Mexico come from the United States.

 

There's just one problem with the 90 percent "statistic" and it's a big one:

 

It's just not true.

 

In fact, it's not even close. The fact is, only 17 percent of guns found at Mexican crime scenes have been traced to the U.S.

 

Now at first blush that sounds like a simple reprint of the LaPierre (not to mention another case of bias by FNC). But apparently someone at Fox decided to at least try and do some actual reporting on this:

 

In 2007-2008, according to ATF Special Agent William Newell, Mexico submitted 11,000 guns to the ATF for tracing. Close to 6,000 were successfully traced -- and of those, 90 percent -- 5,114 to be exact, according to testimony in Congress by William Hoover -- were found to have come from the U.S.

 

But in those same two years, according to the Mexican government, 29,000 guns were recovered at crime scenes.

 

In other words, 68 percent of the guns that were recovered were never submitted for tracing. And when you weed out the roughly 6,000 guns that could not be traced from the remaining 32 percent, it means 83 percent of the guns found at crime scenes in Mexico could not be traced to the U.S.

 

That's cute, I love it when Fox correspondents act like real reporters. Unfortunately there's no follow-through at all. The information is just baldly stated, with no effort to get any sort of confirmation from another source, response from Federal officials, response from Mexican officials -- nothing (or should I say "nada").

 

Frankly it reads like a freaking press-release, hot off the fax machine. However, the New York Times, to its credit, did pick up and run with the same angle in this story on April 14th, just a couple of days ago:

 

http://www.nytimes.com/2009/04/15/us/15guns.html?_r=1&pagewanted=2&hp

 

In any case, the BATF states this to be true, so let's see what the other side can come up with in response. Well, it just so happens that a group called the Violence Policy Center released a study stating that the 90% figure is accurate. (Warning: It's a PDF.)

 

http://www.vpc.org/studies/indicted.pdf

 

This "study" has already been cited in numerous news stories as if it debunks the BATF's statements. The problem with this study is that it actually offers absolutely nothing to support that 90% figure. In fact the study only looks at 21 cases, citing only 1700 actual guns delivered into the hands of Mexican drug cartels. No evidence of more widespread trafficking is offered.

 

Even worse, the Violence Policy Center is a special interest group devoted to the cause of banning all gun possession of any kind in the US. They're known for making misleading statements about guns, and in fact most of this "study" is actually a rant about the behavior of the NRA and gun manufacturers.

 

1700 is not 90% of anything less than ~1900, right? But we already have seen several much larger figures cited in the stories listed above. The smallest figure cited for the total seized weaponry is 12,000, and I believe the largest is 16,000 (I read one story that said 39,000, and I may have linked it above, but I'm getting tired now and I think you get the point).

 

¡Ay, caramba!

 

Ok, so to summarize, there does not appear to be a factual basis for the 90% figure, and it appears to be uncertain what the actual figure might be. In my opinion that is not sufficient reason for ANYBODY to be quoting the 90% figure -- they need to stop doing that. And some reporter needs to get off their duff and find out what the actual truth is. I've written to several, including the editors of Annenberg FactCheck.

 

I'm curious what you all think. Can your googling skills produce better leads than what I've been able to dig up so far? Is this another example of our media failing in its duty a public investigator and defender of the truth? Why can't anybody seem to find out the real story here?

[CaptainPanic]

(Same quote as in Pangloss' post above) In 2007-2008, according to ATF Special Agent William Newell, Mexico submitted 11,000 guns to the ATF for tracing. Close to 6,000 were successfully traced -- and of those, 90 percent -- 5,114 to be exact, according to testimony in Congress by William Hoover -- were found to have come from the U.S.

 

But in those same two years, according to the Mexican government, 29,000 guns were recovered at crime scenes.

 

In other words, 68 percent of the guns that were recovered were never submitted for tracing. And when you weed out the roughly 6,000 guns that could not be traced from the remaining 32 percent, it means 83 percent of the guns found at crime scenes in Mexico could not be traced to the U.S.

 

The way I read this is that from an initial sample of 11000, 12000, 16000 or 29000, a subsample of 6000 guns was created for testing. From that sample, 90% was found to originate in the US.

I believe that the sample is sufficiently large, so the conclusion can be taken as valid (you can actually calculate the likely error, but I'm not in the mood for that)... But this is assuming that the subsample of 6000 guns was picked unbiased.

 

The fact that no conclusion was reached on the remaining thousands of guns does not mean that they're not from the US, and also does not mean that they are from the US. But assuming unbiased sampling and a sufficiently large sample, we can (with known error) conclude that the same percentage of the untraced guns originates in the US.

 

I don't know enough about guns: Would there be any reason to assume that guns that cannot be traced have a different probability to come from the US than guns that can be traced?

 

Completely unrelated to statistics: if so much drugs can travel into the US, wouldn't it be easy to move some guns out of the US?

Edited April 17, 2009 by CaptainPanic

[Sisyphus]

So basically, only a fraction have been successfully traced, and of those 90% are of American origin. So everybody is wrong. If you consider those as something close to a random sampling (which is very probably not the case), 90% is reasonable. If you assume that the highest estimate of total guns is accurate and that every single American gun was submitted and successfully traced (also almost certainly not the case), then FNC is right. In other words, there's a huge margin of error, and everyone is wrong. Par for the course. How about we stick with "thousands of American assault weapons, making up a significant fraction of the total, have found their way to Mexican drug lords."

 

Just curious: where else do they come from? Even if you can't trace to a significant owner, you should at least be able to tell if it was manufactured in the U.S., shouldn't you? If an American company makes a given weapon, then somebody must have bought it legally in America at some point, right?

[Phi for All]

Would the Mexican government stand to profit somehow by only turning over to the BATF those guns they were pretty sure were US in origin (or at least 90% sure)? Are we giving them guilt-ridden aid funds to help combat this inconvenient "truth"? The reporters should be asking the Mexican government why *all* the weapons seized weren't turned over to the BATF for tracing.

[Mr Skeptic]

The way I read this is that from an initial sample of 11000, 12000, 16000 or 29000, a subsample of 6000 guns was created for testing. From that sample, 90% was found to originate in the US.

I believe that the sample is sufficiently large, so the conclusion can be taken as valid (you can actually calculate the likely error, but I'm not in the mood for that)... But this is assuming that the subsample of 6000 guns was picked unbiased.

 

The fact that no conclusion was reached on the remaining thousands of guns does not mean that they're not from the US, and also does not mean that they are from the US. But assuming unbiased sampling and a sufficiently large sample, we can (with known error) conclude that the same percentage of the untraced guns originates in the US.

 

I don't know enough about guns: Would there be any reason to assume that guns that cannot be traced have a different probability to come from the US than guns that can be traced?

 

Completely unrelated to statistics: if so much drugs can travel into the US, wouldn't it be easy to move some guns out of the US?

 

My thoughts exactly. Now I would assume that Mexico would submit American guns to America for tracing purposes, but if you can't tell which gun is from America and which isn't, then they'd submit a largely random sampling. However, if you could easily tell which guns were from America to do this, then you could tell what proportion of them were from America without tracing them.

 

I too wish to know whether American guns are as easy to trace as any other.

 

Incidentally, where else would they get their guns? Its not like we have a shortage of guns or they have a shortage of secret routes into and out of the US.

[iNow]

if so much drugs can travel into the US, wouldn't it be easy to move some guns out of the US?

 

where else do they come from? Even if you can't trace to a significant owner, you should at least be able to tell if it was manufactured in the U.S.

 

where else would they get their guns? Its not like we have a shortage of guns or they have a shortage of secret routes into and out of the US.

This has been the sticking point for me all these weeks. I personally don't give a rat's hindquarters about the actual percentage or how it's calculated. As mentioned above, no matter what, there is going to be a lot of error in such a calculation and it's entirely unhelpful to us anyway. My perspective is that far too many guns are going from the US to Mexico, our existing laws are enough to prevent this from happening, and we simply need to get more enforcement of those existing laws to improve the current circumstances in places like Juarez.

 

With that said, though.... where the hell else do you think all of the guns are coming from? Sure, they get grenades and RPGs from South Korea through Guatemala, but... with the guns... it's not like Smith & Wesson has a huge factory in Honduras or something. Seriously... If you want to do the math, then at least ask the right question. Where are most guns made in this hemisphere? Right here in the US.

 

All of this 90% versus 17% nonsense is just another distraction preventing us from dealing with the actual enforcement issue and moving forward.

[Pangloss]

In terms of where else they come from, as I understand it all the major gun types have knock-off factories in Latin American nations. The tracing process presumably determines whether the gun in question comes from a knock-off factory or the actual American gun manufacturer. I agree that the subject is a distraction, but it's being used by both sides to further an agenda, so it has to be exposed.

 

In terms of the analysis above, the potential sample bias in the above quote is pretty obvious. Mexican authorities submitted guns to American authorities to determine whether they were made in America. Well that's kind of a no-brainer -- you look at the gun, you see if it has a serial number, you check your FBI-connected computer system, you see that it came from America, and you toss it into the "ship to the Americanos to show them a thing or two" pile. On the way out the door you grab a few from the "undetermined" pile and toss them in just to make it look random. The motivation being to shift some guilt north of the border for funding purposes.

 

Not saying that happened, of course, just speculating.

[waitforufo]

What difference does it make? Let's say 100% are made in or processed through the USA. It would only show that gun control does not work. It doesn't work in Mexico and it won't work here.

 

We have another thread on the prohibition of Marijuana. Prohibiting guns would work as well as prohibiting Marijuana. Such a prohibition would only provide another product for criminals to sell.

 

Just because some other country is screwed up does not mean my rights change. My rights are inalienable. A fact that my government simply recognizes.

[Pangloss]

My main interest here is actually what this says (or doesn't say) about journalism, fact-checking, and the use of information by special interest groups.

 

I appreciate all the replies thus far. Several interesting points have been made already.

---

Merged post follows:

Consecutive posts merged

Hey check it out! Last night I sent FactCheck an email asking them to look into this, and look what appeared on their web site today:

 

Counting Mexico's Guns

 

They've probably been working on it for a while, so I don't think Sean Hannity will be calling me for an interview any time soon (though it would be fun if he did; I could feign conservative leanings long enough to get on the air and then suddenly start talking about centrism and compromise, scientific inquiry and the value of non-religious governance).

 

Some relevant quotes from the piece:

 

What Obama should have said is that 90 percent of the guns submitted by Mexican authorities and traced by the U.S. come from the U.S. But the president confused the percentage of guns "recovered" with the percentage of guns that Mexicans choose to submit for tracing by U.S. officials. Mexico actually recovers more guns that it submits, and thus the percentage of guns "recovered" must be less than the percentage traced to U.S. sources.

 

The report also levies some criticism for Fox News:

 

But the Fox figure of 17 percent is based on a misreading of some confusing House subcommittee testimony by ATF official William Newell. The Fox reporters come up with a figure of 5,114 guns traced to U.S. sources in fiscal 2007 and 2008. That figures to 17.6 percent of the 29,000 figure for guns seized in Mexico, as given by the country's attorney general.

 

The 5,114 figure is simply wrong. What Newell said quite clearly is that the number of guns submitted to ATF in those two years was 11,055: "3,312 in FY 2007 [and] 7,743 in FY 2008." Newell also testified, as other ATF officials have done, that 90 percent of the guns traced were determined to have come from the U.S. So based on Newell's testimony, the Fox reporters should have used a figure of 9,950 guns from U.S. sources. That figures out to just over 34 percent of guns recovered, assuming that the 29,000 figure supplied by Mexico's attorney general is correct.

 

Even that number is too low. At our request, an ATF spokesman gave us more detailed figures for how many guns had been submitted and traced during those two years. Of the guns seized in Mexico and given to ATF for tracing, the agency actually found 95 percent came from U.S. sources in fiscal 2007 and 93 percent in fiscal 2008. That comes to a total of 10,347 guns from U.S. sources for those two years, or 36 percent of what Mexican authorities say they recovered.

 

And then, best of all, FactCheck takes a shot at trying to come up with an accurate number.

 

Given the lack of hard data from Mexico we can't calculate a precise figure for what portion of crime guns have been traced to the U.S. Based on the best evidence we can find so far, we conclude that the true percentage is probably less than half of the 90 percent claimed by the president and others in his administration, who base their claim on a badly biased sample of all guns seized in Mexico. But we also conclude that the number is probably at least double what Fox News has reported, based on their mistaken interpretation of ATF testimony.

 

Whether the number is 90 percent , or 36 percent, or something else, there's no dispute that thousands of guns being illegally transported into Mexico by way of the United States each year.

 

Good for them.

[CharonY]

Sounds to me that journalists are pretty bad with numbers. And having a bias (either way) obviously does not help.

[ParanoiA]

Yeah, I second that. And great follow up there Pangloss.

 

We know there are problems with the border and security. This is egregious neglect by congress as they've been well aware for close to a decade now how the populus feels about border security.

 

I guess I don't think of gun regulation when I hear about the number of guns getting to Mexico. I think of failed border security. I think they want us to think of gun regulation and how we need to be stripped of more of them and we're all southerners who drink beer, drag minorities behind our pick-em-up truck and shoot deer from the cab. But I don't. I want the border secure. For once and for all.

[Mokele]

We know there are problems with the border and security. This is egregious neglect by congress as they've been well aware for close to a decade now how the populus feels about border security.

 

Well, in all fairness, it *is* nearly 2000 miles of border, crossing some of the most inhospitable terrain in the continent, with one of the greatest possible wealth divides across it. The feat of making it even moderately secure is breathtaking in scope and difficulty, especially since it's trying to thwart the efforts of a highly intelligent species which excels at cooperation and tool use.

 

I mean, look at the Berlin Wall: incredibly heavily built and guarded, only 96 miles long, with guards who could shoot crossers, and still roughly 5000 people managed to cross in 41 years. Just for fun, if we extrapolate that to the US border (assuming it would cover the same distance and have the same rate of crossings per year per mile), we'd still have an influx of 2500 individuals per year, and I don't even want to *think* about the cost.

[ParanoiA]

Oh yeah, absolutely. But all we're getting is lip service and barely that, anymore. I'm not impressed with fences and bulky physical structures. I'm thinking of something more technologically advanced. I find it hard to believe that we can't electronically revolutionize the art of border security.

 

Maybe you're right and there's really no security to be had. I do understand the nature of the situation in that we're a rich country bordering an impoverished one - no matter what we do, people are going to risk it all to come here. I know I would.

[Mokele]

Going a bit OT here, but:

 

I'm not impressed with fences and bulky physical structures. I'm thinking of something more technologically advanced. I find it hard to believe that we can't electronically revolutionize the art of border security.

 

I'd generally agree, but the desert is pretty harsh on machines as well as people. High-tech would likely be an improvement, but sand gets into *everything*, especially moving parts, so it'd have to be pretty rugged tech.

[Pangloss]

We don't actually have to stop all illegal immigration in order to take control of the situation and manage it fully.

[CaptainPanic]

In most European countries, there is a free traffic of people, but almost no (illegal) guns.

 

I'm not saying that our system is perfect, but it might just be an example.

Of course, it would be nothing short of a revolution in the USA: we have heavy restrictions on gun possession, which seems a big no-no on the other side of the Atlantic...

 

In terms of where else they come from, as I understand it all the major gun types have knock-off factories in Latin American nations. The tracing process presumably determines whether the gun in question comes from a knock-off factory or the actual American gun manufacturer. I agree that the subject is a distraction, but it's being used by both sides to further an agenda, so it has to be exposed.

 

In terms of the analysis above, the potential sample bias in the above quote is pretty obvious. Mexican authorities submitted guns to American authorities to determine whether they were made in America. Well that's kind of a no-brainer -- you look at the gun, you see if it has a serial number, you check your FBI-connected computer system, you see that it came from America, and you toss it into the "ship to the Americanos to show them a thing or two" pile. On the way out the door you grab a few from the "undetermined" pile and toss them in just to make it look random. The motivation being to shift some guilt north of the border for funding purposes.

 

Not saying that happened, of course, just speculating.

 

Thanks for that. For me it's not a "no-brainer". I honestly have never held a gun in my hand. I didn't know they have a serial number. I didn't know that there's a computer system for it. (Possibly I have heard about it before, but I don't just believe anything I hear from Hollywood).

[The Bear's Key]

In most European countries, there is a free traffic of people, but almost no (illegal) guns.

And try getting a job in Europe if you don't have the required permit.

 

So much for Europe being a "liberal" wasteland (..speaking of reporting bias and politicized statements)

 

Of course, it would be nothing short of a revolution in the USA: we have heavy restrictions on gun possession, which seems a big no-no on the other side of the Atlantic...

Well, there's England.

[Pangloss]

In most European countries, there is a free traffic of people, but almost no (illegal) guns.

 

As far as I know ALL European nations control the number of people they allow to immigrate into their country. And as for illegal guns in Europe, the IRA never seemed to have any trouble getting AK-47s from Africa and Asia.

[CaptainPanic]

As far as I know ALL European nations control the number of people they allow to immigrate into their country. And as for illegal guns in Europe, the IRA never seemed to have any trouble getting AK-47s from Africa and Asia.

 

There's a difference between immigration (for work and permanent residence) and travel for holiday purposes. Lots of people from outside the EU can visit, although this is mostly determined by the average income in the other country and also by politics. Basically, if you come from a country from which many people want to immigrate, then that's bad luck and it's difficult to get in...

 

I think that the second point is a good one. The terrorist groups in Europe seem to have plenty of weapons... so yeah, I may have to take back my previous statement... although I hope that groups like the IRA and ETA do not have 29000 guns, because then they might actually outnumber the armies of the smaller member states of the EU...

[Pangloss]

We get tourists too, CaptainPanic. I don't think many of ours are gun-runners either.