Molson Hart

that is not his point. his point is if you think you need police for murder/violent crimes you only need 1-5% of the funding because they spend 95-99% of their time on other (non violent) stuff. he did not argue that you don't need police for murder

or rather the point is not whether you need police to prevent crime or not but what constitutes as crime to begin with

The 95-99% of police that seem to be allocated only towards petty crime serve as a deterrent to serious crime. Suppose you have a cop directing traffic. A violent assault is much less likely to occur where that cop is, even if they're there for traffic, not assaults.

Interactions cut both ways though. Someone who got jailed for possessing weed might be more likely to do crime later because ppl won't hire him etc. So I agree there's a ton of added uncertainty but to both directions.

The solution to the problem you just described is not to reduce policing. It's to legalize marijuana possession.

Yes but then what do you do with the extra police capacity? You can reduce it or redirect it. And if you reduce proportionally you still have the same capacity to traffic murder etc.

You can reduce it after changing the law, if you think it doesn't deter other crime. Bottom line is, even if only X% of the force works on serious crime, 100%-X% deters serious crime. That seems pretty logical to me and I'm unclear why are continuing to argue about it.

What you are saying is logical but not the full picture. What i am arguing about is if we cite 2nd order effects we need to broaden the scope (because 100-X deters AND causes serious crime).