Well, here's the thing, if the level it is set at leads to output-reducing withdrawal from the workforce more than output-increasing effects, then it also leads to inflation which drives price levels up and drives people back into the workforce [0] -- its self-limiting.
The bigger problem is that the economy might not yet be able to sustain a level at which UBI isn't substantially worse for the neediest populations than existing support programs -- increasing capacity through automation will get us there eventually, but its far from clear we're there now.
Which is why, while I think UBI is a great idea to adopt and phase in, I don't think a big-bang short-term conversion is a good idea.
[0] unless it is pegged to inflation, then if its set initially in a bad place, its stuck there. So, if you are going to index it rather than manually adjust it, don't index it to inflation, index it to output/revenue measures, not price levels.
Its not a matter of which price measure you use, using any price measure as a control for UBI means if you set it outside the stable range the economy can support, inflation no longer acts to reduce the real value of the benefit down to the level the economy can support.
The bigger problem is that the economy might not yet be able to sustain a level at which UBI isn't substantially worse for the neediest populations than existing support programs -- increasing capacity through automation will get us there eventually, but its far from clear we're there now.
Which is why, while I think UBI is a great idea to adopt and phase in, I don't think a big-bang short-term conversion is a good idea.
[0] unless it is pegged to inflation, then if its set initially in a bad place, its stuck there. So, if you are going to index it rather than manually adjust it, don't index it to inflation, index it to output/revenue measures, not price levels.