We can be sure addicts are going to get their drugs, and they're going to get the drugs whether they have a place to live and food on the table or not. The question is whether they're going to be committing other crimes that put society at risk in order to do so. I'd rather they spend my tax money than break into my house.
The problem we want to solve is that these people don't have food and shelter - or from a more selfish perspective, that these people are costing us money on prisons and emergency services. Now, would giving these people money solving these problems? Would giving them money be more effective at solving these problems than housing projects and food programs? Like I say, I'd like to believe giving them money works, but I want to see the evidence.
Not exactly. I'm assuming SOME people will take advantage of the system. Some people ALWAYS take advantage of a system with bad incentives.
Lots of people don't want to work, which is fine. However we don't want to reward people more for doing less.
Do some people take advantage of welfare? Sure. But if you're doing pretty ok making $30k a year, would you drag yourself into unemployment just so you can take food stamps? Doubt it.
In fact, I see it completely reversed. It's a short term problem, with a long term solution. If you look at Maslow's hiearchy of needs, people who have their basic necessities met (housing, food, friends, etc) will naturally strive for higher living standards. In the short term, people will become homeless to claim free apartments, but in the long term strive for higher living standards.