It's impossible to fully prove a negative (e.g. prove this code has no bugs, prove there isn't a teacup hiding behind mars), but each study gives us tentative confidence to do larger and larger studies.
If we really are nearing an automation revolution which requires us to become a demand-constrained society, then we will want to understand how people behave with UBI.
If not, then it's still good to understand the nature of unconditional cash transfer programs (even if temporary), for scenarios like the pandemic in the US or as an alternative to other forms of aid in developing economies.
Who said they need to prove a negative? I'm saying these guys are citing a study about apples while making claims about oranges.
They studied the impact of a temporary stipend. Yet the title of the article ("Universal basic income seems to improve employment and well-being") is not a claim about the impact of temporary stipends. It is a claim about UBI.
You cannot honestly infer anything about UBI or the incentives created by UBI from a study of something that is fundamentally different than UBI.
I agree that the results of the study should give us confidence to do larger studies. If studies of UBI result in the same findings, it would likely win me over to the pro-UBI camp.
But for now, I find it irksome that the article claims a study of apples has taught us something about oranges.
I guess the article should be called "2 year unconditional cash transfer program to 2000 people seemed to improve employment and wellbeing".
But I disagree that this doesn't allow us to infer anything about UBI. It tells us something about unconditional cash transfer over a 2 year period, which is a subset of UBI. Baby apples, if you will.
There are of course other facets that aren't tested here, like the duration of the study not being a lifetime, or the fact that not everyone in the society is getting it, or the taxes not manifesting.
Ultimately, no study can fully prove the duration effects of UBI in a useful amount of time. We could do lifetime studies and we still wouldn't know about multi-generational effects. And we could do multi-generational studies, but do we really have 200 years?
---
That said I agree longer term studies are a good idea, maybe 5+ years which would give people time to do career changes. That this study has worked out hopefully gives someone justification to undertake the next.
If we really are nearing an automation revolution which requires us to become a demand-constrained society, then we will want to understand how people behave with UBI.
If not, then it's still good to understand the nature of unconditional cash transfer programs (even if temporary), for scenarios like the pandemic in the US or as an alternative to other forms of aid in developing economies.