OK so for the sake of discussion I'm going to throw out two major roadblocks that will have to be addressed.
First, this kind of thing has been proposed many times before, in other states, and even enacted in FL, only to be ruled unconstitutional (violation of 4th amendment rights) and tossed out. So we really have to ask... what is so different about this proposal that will not just be another dead-end road and big waste of time? Perhaps if it is only food stamp collections and not all unemployment, or if it is clearly known up front so that the participant can make a clear, informed choice? I don't know.
Second is the thing everybody ignores - the "false positive" problem. All drug tests... and this is moreso the case w/ cheaper tests, so you have a financial tradeoff... incorrectly come out positive X% of the time. Its small, 0.01% or better. But if you do 100,000+ tests a year, you WILL have false positives. What happens in those cases? What is the provision for those people? Do you always follow up a positive with a 2nd test? That's more $$ (who pays that?) Or have some rule that if they fail within X% of the margin, they get a free 2nd test to make sure?
It only takes a few lawsuits for this to become very expensive for the state. Of course... the upshot is that by definition the applicants can't actually afford lawyers or to fund a lawsuit, so the state is safe
If the rule is not "everybody get a test", but is instead" you get a test if we "suspect" you are using (as this suggests) then the actual testing rate may be pretty low, only a couple thousand per year. But then the question is... who defines what makes you suspicious?
The general theory and idea behind this sounds great... hence it has "collaborative support"... but there's a lot that has to be worked out first to be implementable.