I love this article, it's cracker-jack full of interesting stuff about a crime family who have industrialised identity document production in the US.
The dominant forgery-and-distribution network in the United States is allegedly controlled by the Castorena family, U.S. Immigration and Customs Enforcement officials say. Its members emigrated from Mexico in the late 1980s and have used their printing skills and business acumen to capture a big piece of the booming industry.
Nice colour, there. Actually the entire article is full of colour, well worth reading. We'll just do the dry facts here:
Federal authorities said that calculating the financial scope of document forgery is virtually impossible but that illicit profits easily amount to millions of dollars, if not billions. One investigation of CFO operations in Los Angeles alone resulted in *the seizure of 3 million documents with a street value of more than $20 million.*
"We've hit them pretty hard, but have we shut down the entire operation? I don't think we can say that yet," said Scott A. Weber, chief of the agency's Identity and Benefit Fraud Unit. "We know there are many different cells out there, and they are still providing documents."
Ouch. 20 millions divided by $3 millions is $7. Identity 7, here we come.
Illegal immigrants are often given packages of phony documents as part of a $2,000 smuggling fee. Others can easily make contact with vendors who operate on street corners or at flea markets in immigrant communities in virtually every city. .. . A typical transaction includes key papers such as a Social Security card, a driver's license and a "green card" granting immigrants permanent U.S. residency. Fees range from $75 to $300, depending on quality.
Identity is a throw-in for a $2000 package tour sold out of Mexico. Say no more. Obviously, these numbers are all screwed up as there is a big difference between $75 and $7. But, consider. Even at $300, it would be more cost-effective for the average American business traveller to travel on false documentation than to do the following:
Currently, individuals who want to clear their names have to submit several notarized copies of their identification. Then, if they're lucky, TSA might check their information against details in the classified database, add them to a cleared list and provide them with a letter attesting to their status.
More than 28,000 individuals had filed the paperwork by October 2005, the latest figures available, according to TSA spokeswoman Amy Kudwa. She says the system works. "We work rigorously to resolve delays caused by misidentifications," Kudwa says.
The TSA's lists are only a subset of the larger, unified terrorist watch list, which consists of 250,000 people associated with terrorists, and an additional database of 150,000 less-detailed records, according to a recent media briefing by Terrorist Screening Center director Donna Bucella. The unified list is used by border officials, embassies issuing visas and state and local law enforcement agents during traffic stops.
This programme is of interest because its identity keystone drives other programmes. We are looking at a 7% error rate as a minimum, which should come as no surprise - of course, there are unlikely to be more than a 100 people on the list that really qualify as "terrorists who are likely to do some damage on a plane" so if the error rate is anything less than 99% then we should probably be stopping the planes right now. About the best we can conclude is that the strategy of stopping terrorists by identifying them doesn't seem worth emulating in financial cryptography.
And Darren points out the statistical unwisdom of relying on such programmes:
Suppose that NSA's system is really, really, really good, really, really good, with an accuracy rate of .90, and a misidentification rate of .00001, which means that only 3,000 innocent people are misidentified as terrorists. With these suppositions, then the probability that people are terrorists given that NSA's system of surveillance identifies them as terrorists is only p=0.2308, which is far from one and well below flipping a coin. NSA's domestic monitoring of everyone's email and phone calls is useless for finding terrorists.
Sure. But the NSA are not using the databases to find terrorists. Instead, when other leads come in, they look to see what they have in their databases -- to add to the lead they already have. Simple. With this strategy, clearly, the more data, the more databases, the better this works.
But, again, it doesn't seem a strategy that we'd emulate in FC.Posted by iang at June 24, 2006 12:29 PM | TrackBack