Another factor people take into account when deciding whether to take a job is what the effect is on their total benefits. Higher taxes for instance decrease t

Another factor people take into account when deciding whether to take a job is what the effect is on their total benefits. Higher taxes for instance decrease the amount of benefit in taking a job (relative to the value to your employer of you having taken that job). But taxes aren't the only thing that decreases the benefit of taking a job, there is also the potential loss of other benefits. When we have "means tested" programs (i.e., anything for "the poor" or "the unemployed") then the marginal benefit of a job (or doing more work) U(base or base job income + new job income) - U(base or base job income). When you factor in needs-based benefits U() can be a very peculiar function, and the marginal benefit can be small relative to the new job income, or even negative. If we take the social safety net as a given, then we shouldn't abstract away the existence of that net. A basic income has the potential to remove the need for more sophisticated implementations of the safety net, ones that can lead to poor U() functions. In fact I think that the bottom of the concave U() function cannot work as in theory, because of what is considered socially acceptable. Your first $1000 in income is very valuable because you are living in terrible deprivation. But some people are not able to make that income, no matter the utility. It may be job availability, it may be a disability, or an injury, or addiction to drugs, a criminal history. It could be a parent that has succumbed to any of these. It's hard to enumerate or easily define what any of these are, so to some degree we are willing to offer help simply because someone is experiencing deprivation. Which is to say, the part of the U() curve that a basic income chops off is already really messed up.