We're probably in agreement, then. Though, with a caveat: sometimes it can fail in a town but succeed on a wider scale. For instance, a city offering generous,

We're probably in agreement, then. Though, with a caveat: sometimes it can fail in a town but succeed on a wider scale. For instance, a city offering generous, open-access homeless benefits is almost doomed to have all its benefits swamped as it attracts homeless from around the country, but that doesn't give us much information about how the same policies would work if implemented nation-wide. In the case of the UBI, though, those effects can be worked around by designing the experiment well.