The Devil In The Details: LA Times Ignores Substance, Attacks U.S. Gun Manufacturers

The L.A. Times is parsing math.

If you were to not read Josh Meyer’s June 17 article very carefully, you might think that 90 percent of the weapons recovered from Mexican cartel raids originated in the United States:

The report by the congressional Government Accountability Office, the first federal assessment of the issue, offered blistering conclusions that will probably influence the debate over the role of U.S.-made weaponry as violence threatens to spill across the Mexico border.

According to a draft copy of the report, which will be released today, the growing number of weapons being smuggled into Mexico comprise more than 90% of the seized firearms that can be traced by authorities there.

Pay close attention, however, to the wording.  That’s 90 percent of the seized firearms – that authorities are able to trace.  This wording actually reflects the vagueness of the GAO report’s highlights:

While it is impossible to know how many firearms are illegally smuggled into Mexico in a given year, about 87 percent of firearms seized by Mexican authorities and traced in the last 5 years originated in the United States, according to data from Department of Justice’s Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF).

The actual report is much more specific about the limitations of the data:

In 2008, of the almost 30,000 firearms that the Mexican Attorney General’s office said were seized, only around 7,200, or approximately a quarter, were submitted to ATF for tracing.

In other words, the sample size used to manufacture the 90-percent statistic is roughly one-quarter of the actual number of weapons seized.  Of those weapons lucky enough to be chosen for ATF examination, 90 percent came from the United States.  Even more strangely, the L.A. Times overlooks an important issue with the GAO’s methodology:

However, as noted earlier in the report, because firearms seized in Mexico are not always submitted for tracing within the same year they were seized, it was not possible for us to develop data to track trends on the types of firearms trafficked or seized.

So the major statistical hang-up is that the Mexican authorities arbitrarily choose which weapons to submit for tracing – and when.

To sum up what we have so far: Of the roughly one-quarter of the total weapons seized that are arbitrarily submitted to the U.S. for tracing and can be traced with the limited available data, 90 percent come from the United States.

And here’s the best part:

As is inherently the case with various types of illegal trafficking, such as drug trafficking, the extent of firearms trafficking to Mexico is unknown[...].

There is a footnote immediately following this statement, which reads:

While there is no data on the total number of firearms trafficked to Mexico, the Government of Mexico maintains data on the number of firearms seized in Mexico. Information, such as serial numbers, on many of these seized firearms is submitted to ATF’s National Tracing Center for tracing. ATF’s National Tracing Center attempts to trace the firearms using the information submitted. Between fiscal years 2004-2008, around 52 percent of trace requests from Mexico that were submitted to ATF’s National Tracing Center identified the first retail dealer. Furthermore, according to ATF, the identification of the country of manufacturing origin of a firearm does not depend on identifying the first retail dealer but rather on the initial description of the firearm.

Holy bureaucracy, Batman – these statistics are based on an incomplete knowledge of sample size?  Not only that, but the much-ballyhooed ‘tracing’ of these weapons is apparently done by looking at the firearm and saying “Yep, that looks like an American weapon.”

One might wonder if anyone at the L.A. Times bothered to read this report.   

A summary of the GAO report can be found here, and the full GAO report can be found here.  Also, here's Fox News' analysis of the 90-percent statistic – it was released well before this GAO report, but this same parsing of facts occured previously as well.