My big questions that do not seem to be answered in the paper:
1. Did any participants drop out? Were there exactly 107 treated participants at both the start and end of the trial? Previous UBI experiments I've seen have huge biases where dropped out participants are excluded, but dropping out is affected by treatment.
2. The ~1700 participants were selected from the 20,000 applicants via a procedure that accounted for their opinion about UBI, "the sample was supposed to contain an equal number of proponents and opponents of a universal basic income". Why was this done? Can we see how these survey responses correlate with outcome? Maybe the participants are much more/less supportive of UBI than the German population.
3. The participation criteria required that participants "had a personal, monthly income between EUR 1,100 and 2,600" and "were not unemployed for more than one year". This seems like it would introduce a lot of bias, participants are all members of the 94% of Germans I would expect to have their labor participation least affected by UBI.
Seems quite a bit better overall than the studies in the Bay Area over the last few years
A far more interesting study would be to give its to a bunch of graduating high school students, guaranteed for the rest of their lives, and see how many are contributing meaningfully to the economy at the end of a few years.
1. Did any participants drop out? Were there exactly 107 treated participants at both the start and end of the trial? Previous UBI experiments I've seen have huge biases where dropped out participants are excluded, but dropping out is affected by treatment.
2. The ~1700 participants were selected from the 20,000 applicants via a procedure that accounted for their opinion about UBI, "the sample was supposed to contain an equal number of proponents and opponents of a universal basic income". Why was this done? Can we see how these survey responses correlate with outcome? Maybe the participants are much more/less supportive of UBI than the German population.
3. The participation criteria required that participants "had a personal, monthly income between EUR 1,100 and 2,600" and "were not unemployed for more than one year". This seems like it would introduce a lot of bias, participants are all members of the 94% of Germans I would expect to have their labor participation least affected by UBI.
Seems quite a bit better overall than the studies in the Bay Area over the last few years