The warning is part of a report released by the Institute Monday that focuses on the 287(g) program, which provides local police with federal training to enforce immigration law.
“Indeed, Secure Communities may even be more susceptible to this problem since there are no formal agreements defining the activities of participating law enforcement agencies, and local officers do not receive federal training in immigration enforcement,” the report said.
As part of their study, researchers assessed the implementation, enforcement, costs, and community impact of the 287(g) program, looking at the program’s task force, jail, and hybrid models at seven study sites. Of the sites, most unique was Los Angeles County in California, which does not check immigration status of an individual until they are convicted of a crime.
Researchers reported that jurisdictions that had jail models, where officers screen arrestees inside prison or jail, had accountability concerns because it did not impose federal oversight on officers who make initial arrests.
“We have found that it is quite likely in the absence of that training and in the absence of oversight that the officers that bring those people to the jail could be doing it on pretexual context, could be engaging in racial profiling and that is problematic,” said co-author Muzaffar Chrishti. “Therefore a better mechanism has be placed in monitoring those officers that are not 287(g) officers.”
Ninety percent of the 48,000 detainers issued in FY2010 were from jail models, compared to hybrid models at 8 percent and task force models at 2 percent.
According to researchers, about half of all 287(g) activity involved non-citizens arrested for misdemeanor or traffic offenses, veering away from the Obama Administration’s mandate of targeting and removing “dangerous criminals.”
Researchers also found that state and localities were setting their own priorities for the 287(g) program, finding that some jurisdictions operated “targeted models” aimed at high level offenders, while others operated a “universal model” designed to identify as many unauthorized immigrants as possible.
For example, it found that Las Vegas placed 70 percent of detainers on Level 1 or Level 2 offenders, while Cobb County in Georgia and Fredrick County in Maryland placed 80 percent detainers on Level 3 or traffic offenders.
In sites, operating universal models, researchers found that 287(g) participants and ICE officials reported that they placed detainers on as many Level 3 and traffic offenders they encountered on account of ICE detention capacity and the increase in trained personnel.
The findings have implications for the Secure Communities program, which is currently implemented in 998 jurisdictions in 37 states and will be activated nationwide by 2013.
Secure Communities, a technology-touted program, relies on police in local jails to enter prints of those they arrest into a joint FBI and ICE database. Whereas in 287(g) where officers are trained to screen arrestees on immigration status, ICE retains authorization of detainers after there is a match in the system.
Like the 287(g) program, Secure Communities has primarily identified more Level 2 and Level 3 offenders than Level 1. Of 398,257 matches, only 60,457 were identified as Level 1 offenders, according to statistics from October 2008 to November 2010 from the Immigration, Customs, and Enforcement agency.
Researchers say that ICE’s interior enforcement programs like 287(g) and Secure Communities should be more targeted to Level 1 and Level 2 offenders, promote ICE-community dialogue, and work with the Justice Department to investigate racial profiling accusations.
“Unless ICE takes steps to ensure that it is a targeted program, Secure Communities may replicate many of the controversies and negative community impacts we observed in universal enforcement 287(g) jurisdictions,” authors said.