> Everything you say is entirely logical, but omits the reality that as a black person in america you are vastly more likely to be arrested for drug crimes. Not more likely to DO drug crimes, just more likely to get caught.
It is true that black people are more likely to be arrested than white people for drug crimes. 57% of arrestees are white, 37% black. That's about 3x the percentage that you'd expect if you just go by the racial distribution of drug users.
However, that's not actually really relevant here because most drug users, white or black, do not get caught at all. There are about 1.2 million drug arrests per year, so roughly 680k whites and 444k blacks.
We've got about 320 million people. Let's be generous and knock out people 0-14 (20%) because they are probably too young to have gotten into drugs, and also knock out 55+ (26%) because older people don't seem to get arrested much for casual drug use (perhaps if you are old and a regular drug user you've figured out how to not get caught...). So that leaves us 173 million people who are similar demographically to the people getting arrested for drugs.
That would be about 22 million blacks, 109 million whites.
So...we've got about 680k whites arrested for drugs per year out of 109 million, which means 108.6 million out of 109 million whites are NOT arrested. In percentages, 99.4% of whites are not arrested for drugs in any given year.
For blacks, we've got 444k arrested for drugs out of 22 million, which means 21.6 million out of 22 million blacks are NOT arrested. In percentages, 98.0% of blacks are not arrested for drugs in any given year.
98.0% is lower than 99.4%, and that is something that should be investigated--I'm not trying to argue that there is no bias. I'm just saying that by focusing on those arrested rather than those not arrested, you get a misleading picture.
This kind of mistake is common when people assess risks, and so people tend to overreact when something increases a very low probability risk. For instance, we read that some particular food might increase our chances of getting some particular disease from 1/10^6 over our lifetime to 3/10^6, and we think we need to immediately cut that food even if we like it and it does plenty of good things for us, because it increases our risk 3x and that sounds huge.
Oddly, we don't seem to do this with transportation. Someone trying to decide whether to drive or fly from Los Angeles to San Francisco will probably decide based on cost, time, convenience, things on the driving route they might want to visit, whether they want their car with them in San Francisco, and similar factors. For most people I don't think that the fact that flying (at least on a major airline) is safer than driving would be a factor. They recognize that the chances of dying in an automobile accident on a drive from Los Angeles to San Francisco are so low as to not be worth considering, and so that some other mode of transportation would have even lower chances is not relevant.