The US Courts system has just released its annual report on wiretap warrants. As you'll recall, warrants are those quaint permission slips that law enforcement officers occasionally seek before performing searches. (Although it looks as though they'll be doing it more frequently from now on…)
Not included in this report are outlying actions by agencies whose rogueness has been codified. The US Courts FAQ on the report points out that data from the FISA court is not included. Presumably, the previously illegal warrantless wiretaps deployed by the federal government aren't being tracked either, because if there's no warrant, there's no court. And if there's no court, there's no oversight.
Also not included are pen register/trap and trace orders, which are basically given the same amount of scrutiny that subpoenas and NSLs are given: very little, if any.
No report to the AO is required for the use of a pen register (a device attached to a telephone line that records or decodes impulses identifying the numbers dialed from that line) unless the pen register is used in conjunction with any wiretap devices whose use must be recorded. No statistics are collected on the number of devices used in conjunction with each order.
The US Courts system has handily provided ten years of data on wiretap warrants and everything in it points to one thing -- US law enforcement is almost exclusively focused on fighting the War on Drugs.
Of the 22,741 warrants issued since 2003, 21,838 (96%) were issued under the heading of "Narcotics."
And the warrants themselves are ridiculously easy to obtain (giving lie to law enforcement's complaints about the complexity of securing a search warrant). To obtain these 22,741 warrants, 22,748 applications were submitted. Only 7 applications have been rejected since 2003, giving law enforcement a batting average of .99969.
So, we know drugs are big business for law enforcement. At an average of $41,119 per intercept deployed, expenses for just the drug-related wiretaps in 2013 easily tops the $100,000,000 mark
, and that's assuming all of those adhered to that average (the average was taken from 2,069 wiretaps where costs were reported.) The cost per intercept seems to be coming down (it was over $62,000 in 2003) but the number of intercepts has doubled over the same time period. The US Courts' information page also notes that further expenses are often reported after the reporting period has ended. (It points to an additional $62 million in expenses reported in 2013 for wiretaps issued in previous years.) Drugs are not just a multi-billion dollar business for distributors. With this intense focus on one particular form of crime, law enforcement has become just as big a "customer" of the drug business as drug users.
That data may not be particularly surprising, but it does reveal just how much attention is being paid to a narrow range of criminals. Nowhere does the report indicate anything has been deployed to combat terrorism, which is often the stated reason
for obtaining bigger and better ex-military weapons and vehicles. It could be that wiretaps issued for counterterrorism are hidden under other designations or done without court approval. Or it could be that there's simply not nearly as much domestic terrorism activity as law enforcement officials claim.
The more surprising data, especially when coupled with the fact that US courts have only rejected .00031 of the wiretap applications that cross their desks, is the sheer number of people and communications being swept up by single warrants.
San Mateo County, California had only one wiretap warrant issued (probably related
to the investigation of Sen. Leland Yee
), but it was able to put a whole lot of people under surveillance with that single document. Its single intercept gathered communications from 588 people, with a total of 19,477 interceptions
… of which only 513 proved to be incriminating
. (It should be noted that intercepted communications can also include text messages, a form of communication that can easily cause these numbers to swell to ridiculous proportions.)
Likewise, Franklin County, Ohio also sought only one wiretap order, but it, too, had a huge payoff -- 551 people, 9,654 communications with 454 deemed incriminating
(most likely related to this drug bust
But with all these large numbers and expenses come massive amounts of arrests and indictments, right?
Well, not really. Even accounting for the fact that it can take years before the results of an investigation result in jail time, the percentage of convictions resulting from wiretap-related arrests seems to be hovering right around 45%. And the number of arrests is far smaller than number of people whose communications were intercepted during the course of the investigation.
In 2003, 167,272 people
had their communications tapped (with 4.3 million communications intercepted
) but this has only resulted in (to date) 5,705 arrests
and 2,523 convictions
. 2004 was even worse, with 215,460 people surveilled
(5.1 million communications intercepted
). The end result, a decade on? 6,717 arrests and 2,815 convictions
If these patterns hold (and there's no evidence they won't), the hit rate of wiretap deployments will continue to fall, at least in terms of arrests and convictions. The average number of people tapped by an order last year was "only" 97 (the number has hovered between 100-110 over the course of the last decade) but the average number of communications harvested from each wiretap has been increasing over the past several years. Last year, it was 4,558 communications per wiretap (2003/2004 were 3,004/3,017 per) and the number of wiretaps issued has more than doubled
in that same period. 2013's numbers are astounding: 346,872 people surveilled with 16,299,408 communications intercepted, with less than 3 million declared "incriminating."
Within those numbers is another interesting fact: encryption is rarely used and even when it is, it's rarely effective
The number of state wiretaps in which encryption was encountered increased from 15 in 2012 to 41 in 2013. In nine of these wiretaps, officials were unable to decipher the plain text of the messages. Encryption was also reported for 52 state wiretaps that were conducted during previous years, but reported to the AO for the first time in 2013. Officials were able to decipher the plain text of the communications in all 52 intercepts.
Contrast this minimal number with the thousands of devices tapped with the claim made by the US government (in the US v. Wurie warrantless cell phone search case) that requiring warrants would put law enforcement at the mercy
of tech-savvy criminals.
[S]earching an arrestee's cell phone immediately upon arrest is often critical to protecting evidence against concealment in a locked or encrypted phone or remote destruction.The numerous party and amicus briefs in these cases have not seriously undermined that fundamental practical point. Although the briefs identify various techniques to prevent the remote-wiping problem (none of which is close to perfect), they barely address the principal problem that the government identified: automatic passcode-locking and encryption.
There is no tech arms race. Encryption was encountered 41 times durings the states' surveillance of over 200,000 people. The encryption "held" only nine times. All fifty-two times the federal government encountered it during its surveillance of millions of people over the last several years, it was able to defeat it.
The US Court system has proven indistinguishable from the FISA Court with its approval rate that's only .03% away from 100%. Likewise, the surveillance it approves sweeps up tons of "incidental" communications from completely innocent parties. The end result, however, is possibly more futile than the War on Terror. The Drug War isn't headed for a win, nor will it ever be. It will never even be a tie. This has gone on for 40 years and the only "benefit" has been increased budget lines for law enforcement agencies and the steady militarization of local police forces. Billions of dollars in taxes are being poured into a battle many Americans don't feel is worth fighting, while the system itself remains largely insulated against public opinion. Judges are granting nearly anything as long as the word "drugs" appears on the request and law enforcement agencies know this.
If nothing else, this reports shows that law enforcement should never have a problem with seeking a warrant. Even large-scale surveillance efforts (like the sweeping up the communications of 500+ people with a single wiretap request) are flying through the court system without a hitch.