Following a period of increasing immigration enforcement under George W. Bush's administration, the Obama administration reversed immigration policies and issued strict new guidelines to relax enforcement in 2011. The purpose of this paper is to exploit this natural experiment in the enforcement of the immigration laws to study the effects of federal immigration policies on local enforcement, crime and policing efficiency. I use a unique and new data set obtained through a Freedom of Information Act request on several steps of the deportation process. I estimate how the drop in federal immigration enforcement affected county level enforcement, local crime rates and policing efficiency. My empirical analysis suggests that Democratic counties complemented federal policies, by reducing their immigration enforcement, whereas Republican counties tended to maintain higher levels of enforcement and to not react much to the guidelines. Employing a triple-difference approach, I find that Democratic counties with higher non-citizen population shares saw greater increases in clearance rates, a measure of policing efficiency, with no increase in crime rates. The results indicate that reducing immigration enforcement did not increase crime and rather led to an increase in policing efficiency, either because it allowed police to focus efforts on solving more serious crimes or because it elicited greater cooperation of non-citizens with police.