Page 3 of 4
In practice, though, the sheriff's program has been out of step with ICE's stated goals from the very beginning. For the 2009 review of the first two years of Hall's 287(g) program, it was found that 85 percent of those processed for immigration violations were arrested on misdemeanor offenses — most often for driving without a license.
This disturbed advocates who feared, despite public pronouncements, that 287(g) was being used to round up undocumented immigrants, regardless of the seriousness of the crime.
"Are people ending up in the sheriff's custody for offenses they might not normally end up in the jail for, because of some suspicion that they're not in the country legally?" Newbern asked.
Indeed, right after the program began, a Criminal Justice Planning analysis of Metro Police COMPSTAT data indicates the arrest rates for Hispanics caught driving without a license doubled. For everybody else, the arrest rates for that offense declined by 25 percent. This was particularly confusing, because Metro police officers don't participate in 287(g) — at least not officially.
In fact, one of Hall's big selling points to ICE was that a firewall would exist between the arrest by Metro Police and the immigration screening at Davidson County jails, because Hall's deputies can't make arrests. Yet the no-license arrest rates raised a logical question: If 287(g) is only being enforced in the jails, why would rates have doubled for Hispanics — unless Metro Police were using the no-license offense as a pretext to arrest someone they suspect could be deported?
Ozment has often said that 287(g) encourages racial profiling. Sheriff Hall, on the other hand, argues our system of General Sessions judicial commissioners — who determine whether an arrest is valid before booking — ensures that illegal immigrants aren't picked up on bogus charges or racially profiled. So who is right?
First, it's important to understand that ICE district counsel must prove to an immigration judge that a person is here illegally. In that way, it's not much different from the relationship between a cop and a district attorney, the 287(g) officer being the cop. If the person can prove he has a right to be here — as Renteria did — he should get off the ICE hold and out of jail.
But here's where the police/287(g) similarities end. If for some reason Renteria couldn't prove it — even if the 287(g) officer hadn't proved he's here illegally — he might end up detained anyway. Frequently, someone like Renteria, who has multiple criminal charges and can't post bond, will get whisked away to a prison in Alabama and then one or two more in Louisiana before ending up detained in a federal prison in Oakdale, La. That's where the immigration court is located for people who can't get or pay a bond — guilty until proven innocent.
Therein lies the danger of a Catch-22 with the General Sessions commissioners. If a commissioner finds probable cause for an arrest, Hall can use that finding as proof the suspect poses a danger to the community — a reasonable basis to recommend deportation. How does Hall know they should be deported? They got arrested, didn't they?
"Okay, I don't know of anybody unless they've been arrested," Hall said in a deposition taken during the Juana Villegas case. "So there is no case you can give me that doesn't include an arrest. ... And the reason that I know that person posed a risk is that they wouldn't be in jail if they didn't."
The problem is, we'll often never know whether the arrest even led to a conviction. That's because ICE doesn't require that 287(g) programs follow through on the original charge. In explaining the procedure to the Scene, Temple Black, an ICE spokesman, made resolving the initial incident sound like the critical first step: "Once the criminal charge is resolved, an ICE supervisor utilizes ICE's enforcement priorities as a guideline when making custody determinations. ..."
So the Scene called bail bondsmen around town to see if ICE adheres to its own stated expectation: that all local charges be pursued to completion before someone is detained in Louisiana. After all, nobody takes a bigger hit than a bondsman if a defendant doesn't show up in court.
To a one, the bondsmen said they wouldn't post bond for someone on an ICE hold. Too often, they said, the defendant is shipped off to a Louisiana detention facility, which causes the accused to miss his hearing — and leaves the bond company holding the bag.
"[The judges] make us pay it off even though the federal government puts their hand on it and ships 'em off, even though they're not supposed to ship 'em off if they got charges here," said a bondsman who spoke only on condition of anonymity.
By spiriting an undocumented person away even after they've posted bail, ICE may actually be violating a provision of the Tennessee Constitution that states anyone can post bail and be released, except on capital crimes. But if programs such as Sheriff Hall's are guilty of straying substantially from ICE's intent, ICE itself is culpable for not more closely monitoring local deputies who suddenly find themselves wielding the power of federal immigration authorities.
A January 2009 study released by the U.S. Government Accountability Office found that federal oversight of 287(g) was patchy across the country. According to the study, many programs were using their immigration enforcement power to process undocumented people for minor offenses, such as open container and driver's license charges. Yet the report found that many of those programs — unlike Nashville's — lacked specific language in their ICE agreements about the type of criminal 287(g) is intended to nab.
That same year, now working for President Barack Obama, ICE drew up a new tiered approach that emphasized 287(g)'s focus on serious criminals. The most serious felonies would be Level One. Minor drug offenses and property crimes would make up Level Two. The rest, small-potatoes misdemeanors, would be lowly Level Three.
From then on, according to federal mandate, 287(g) officers were to focus on the top level, the worst of the worst. And yet the new agreement simultaneously appeared to lower the bar. Instead of a state or federal felony conviction, the suspected undocumented immigrant now needed only an arrest for a local, state or federal offense, which could be just about anything.
Despite this fine print, the new 287(g) priority levels accomplished two things. They appeased immigrant advocates, who were infuriated at all the small-time offenses that had netted potential deportees. And they temporarily quieted conservative critics, who claimed Obama was soft on illegal immigration.
But there remained no accountability system in place to ensure the rules were followed. As of March 2010, hardcore Level One offenders — the program's ostensible focus — made up only 9 percent nationally of those processed through 287(g), according to the Department of Homeland Security's Office of Inspector General. Level Three misdemeanor offenses, however, accounted for a whopping 50 percent.
Last July, an internal ICE memo sent to its field offices identified 11 287(g) programs across the country that were recommending deportation for too many people brought in on minor offenses. Davidson County's was one of them. "Aliens convicted of the most serious crimes are the highest priority," the memo read.
The audit also indicated that from January to March 2010, 73 percent of the undocumented people processed through Davidson County's 287(g) program were arrested on petty charges. Interestingly, the office's 287(g) priorities were in much better alignment during the following three months, despite the fact that the number of undocumented people processed through the program remained fairly consistent. In fact, the percentage of Level Three offenses was halved.
ICE spokesman Black addressed this abrupt turnabout only obliquely: "ICE also may, on a case by case basis, exercise prosecutorial discretion and not make an administrative arrest." He went on to say that the buck stops with ICE supervisors, who are charged with ensuring that the program's priorities are followed.
What this does not explain is why, throughout the program's entire existence, 287(g) criminal priorities were rarely observed — and why so many Level Three misdemeanors ended in notices to appear before an immigration judge.
The answer may be found in sheer numbers. Davidson County's program turns out a relatively high number of deportation recommendations — nowhere near Los Angeles or Maricopa counties, but still among the top. In December, a Washington Post story showed that's where Washington's priorities truly lie.
The story reported that ICE brass were pushing field-office directors to deport as many people as they could before the end of the fiscal year 2010 — in order to break last year's record. Typically, even charges like drunken driving and misdemeanor assault would have required undocumented immigrants to go before an immigration judge. Now, ICE officials were going so far as to suggest giving undocumented immigrants with these kinds of offenses the ability to voluntarily return to their native country, without a ding on their immigration record.
They got the numbers they wanted. "It has been another record-breaking year at ICE," Homeland Security Secretary Janet Napolitano said, announcing the more than 390,000 deportations in 2010 — more than the Bush administration or any other before. In private, ICE was pushing quantity above all else. Publicly, Napolitano was touting the program's ability to post those blockbuster deportation numbers "in sensible, firm and thoughtful ways."
Particularly under 287(g), however, ICE's enforcement has been anything but sensible and thoughtful. Since the beginning, some 186,000 people have been identified nationwide for deportation or voluntary removal under 287(g). But according to ICE's own statistics, only 127,000 were actually deported.
Following a Freedom of Information Act request, Syracuse University obtained overall ICE deportation numbers, finding a similar pattern for the last three months of fiscal year 2010: Roughly one in every three attempted deportations by ICE was rejected by an immigration judge — either for humanitarian reasons, or because of some other wrinkle in the most complex system of laws in the United States. ICE declined to release more detailed information about which program, whether 287(g) or others, was responsible for most of the thwarted deportations.
In the roughly four-year existence of Davidson County's program, some 8,400 people have been identified for deportation or voluntary removal by Sheriff Hall's 287(g) deputies, who receive about four weeks of ICE training. Only 5,600 ended up being deported — a percentage slightly worse than the national 287(g) average. If you factor in the total number of people in Davidson County detained on ICE holds and interviewed — which is 13,500 — the percentage deported drops to 41 percent.