There are two issues:
- How much we redistribute, i.e. how much goods and services the government offers to people without market income
- How we distribute, i.e. cash vs. in-kind benefits
Your last paragraph addresses (1), and here I agree that if we’re looking to maximize welfare per dollar, we should look to the poor in developing economies rather than the US, given they’re so much poorer. It’s why I give pretty much exclusively to GiveDirectly and Against Malaria Foundation, both of which focus on the extreme poor in Sub-Saharan Africa.
But concerns around how people will use cash, e.g. substance abuse, address (2). Here there’s no evidence that people use cash less responsibly than in-kind benefits, while there’s a mountain of evidence that in-kind benefits cost much more to administer, and in many cases people make better choices with cash than in-kind benefits.
The basic income debate often gets into the realm of (1), but that’s a distraction. UBI doesn’t have to change the amount of redistribution; it does have to change the form. Many UBI advocates (including myself) would like to see more redistribution, but that’s a separate issue that can also be addressed by increasing budgets for programs like food stamps and Medicaid.