Rob MacCoun, a professor of law and public policy at the University of California, Berkeley, who has studied marijuana use in America, said there was little doubt that blacks — particularly black men — bore the brunt of arrests for marijuana.
“The arrest statistics are disproportionate with respect to African-Americans and disproportionate with respect to use,” said Mr. MacCoun. “And that’s very hard to justify in any way.”
On the contrary, it's easy to justify. For example, if African-Americans who smoke marijuana are conspicuous in their dress, while caucasians who smoke are not. In this case, even if 40% of both African-Americans and 40% of caucasians smoked marijuana, we would expect African-Americans to be stopped more often on suspicion of possession. Why? Because African-Americans arise suspicion at a higher rate than caucasians due to (continuing with the example) their dress code. So conditional on suspecting marijuana use, a police officer will be correct more often with African Americans. Certainly the police arrest those who arise the highest suspicion first--otherwise they would be wasting more of their time on innocent people, and so with sufficient constraints on how many people they can arrest, we would expect police to arrest a higher proportion of African-Americans than caucasians.