The Migration Policy Institute held a press conference Monday to release a study that assesses the implementation and impact of Section 287(g) — a federal statute that allows local law enforcement agencies to enforce federal immigration laws — in seven communities. The study finds that the program should be redirected to better identify serious criminals, and makes a series of other recommendations on how to improve the law.

287(g) is implemented when a local law enforcement agency signs a Memorandum of Agreement with Immigration and Customs Enforcement (aka ICE). In 2010, the Office of the Inspector General of the Department of Homeland Security issued two reports with over 45 recommendations for ICE to improve the 287(g) program.

Since 2002, the Florida Department of Law Enforcement and three local law enforcement agencies have signed 287(g) agreements with ICE.

Muzaffar Chisti, one of the report’s authors, told The Florida Independent during the press conference, that “a well-implemented 287(g) is much more defensible” than an Arizona-style immigration enforcement bill. “In 287(g), there is an agreement, parameters, that don’t exist in S.B. 1070,” Chisti said.

In recent legislative hearings in Tallahassee on the issue of immigration enforcement, lawmakers mentioned 287(g) as an alternative to proposed immigration enforcement bills that mimic Arizona’s S.B. 1070.

Chisti said that even though 287(g) is only active in 72 jurisdictions throughout the U.S., the Migration Policy Institute considers it an important case study because other ICE programs like Secure Communities have similar issues to 287(g).

He explained that the institute decided to review 287(g) because it provides an important lens into the relations between state and federal enforcement. He also said increasing evidence that comprehensive immigration reform is not going to happen anytime soon was a factor in their decision.

According to Randy Capps, the report’s coauthor, they found concerns about whether 287(g) targets serious multiple offenders or many unauthorized immigrants.

ICE has implemented a three-level system to identify criminals. Level 1 is reserved for the most serious offenses: major drug trafficking, murder, rape and manslaughter. Level 2 includes minor drug offenses and property theft, while Level 3 covers misdemeanors and traffic violations.

Capps explained there are three different 287(g) models: Jail model, Task Force model and Hybrid model. The Jail model accounts for 90 percent of 287(g) programs, the Hybrid model 8 percent and the Task Force model only 2 percent.

As the name indicates, in the Jail model, officers screen detainees for immigration status and issue detainers to initiate a deportation process. Task Force model officers participate in immigration enforcement operations on community streets. The Hybrid model is a combination of both.

Capps said the report also shows that in the seven jurisdictions studied, 287(g) is implemented in different ways. The four counties located in the South (Virginia, Maryland and Georgia) favor a universal program that issues detainers for removal on virtually every detainee.

The study itself indicates that at the national level, about 50 percent of detainers are placed on Level 1 and Level 2 offenders, while the other 50 percent of detainers are placed on people who have committed misdemeanors and traffic offenses. This has led to cases where detainees were removed before conviction.

The report also shows that the law enforcement agencies of Los Angeles, Calif., and Las Vegas, Nev., have developed targeted versions of 287(g) — that is they place the majority of detainers on Level 1 and 2 violations.

According to Marc Rosenblum, the authors concluded universal versions of 287(g) have a larger community and cost impact, while targeted programs better fit ICE goals to remove criminal aliens who are a danger to the community.

“[287(g)] models are being determined at the local level, favoring a universal program due to political pressure,” Rosenblum said.

Capps pointed out that ICE should work with local jurisdictions to develop targeted versions of 287(g).

The report also shows that while 287(g) Jail model programs are run by a county sheriff’s office, the arresting officers from other police departments that issue the initial detainer for deportation are not being trained or supervised by ICE.

The authors recommendations include:

• Implement targeted 287(g) programs focusing on Level 1 and 2 offenders
• Place detainers for removal after conviction
• Investigate racial profiling
• Weigh cost benefits
• Expand programs to jurisdictions where a targeted version will be put in place
• Establish steering committees to establish real dialogue between ICE and local communities

The report indicates that across all study sites,

immigrant and civil rights groups, service providers, elected officials and community respondents expressed concern that enforcement efforts lead to racial profiling by police and instill distrust of police.

In three counties studied by [the Migration Policy Institute] there was short term drop in Hispanic non-citizen populations while neighboring counties experienced no such decline.