4 ms·

> As always in these situations, I ask what experiment and evidence would you accept that UBI works? Being dismissive is different from being a skeptic. A skept

> As always in these situations, I ask what experiment and evidence would you accept that UBI works? Being dismissive is different from being a skeptic. A skeptic also has an experiment and hypothesis that could disprove their position. That's exactly the problem. I'm not sure a small scale study is actually capable of showing that UBI works or doesn't work. I worry that the only way we're going to find out is actually implementing it. > If you don't have a position that can be disproven, you're just an annoying gadfly. Again, I'm pro-UBI. I just don't think that people's claims about UBI can be proven or disproven without it actually being done at scale, both in terms of the number of people given money and the length of time the money is provided. I'm a physicist. I'm not anti-science. However, I am against the mis-use of science in performing experiments that cannot show what they aim to show. There is no law of nature saying that for any system you wish to study, there is a simpler, easier to understand system you can study instead which should roughly replicate the results in your bigger messier system. Human economies are complex.

chillacy 6y ago
Agreed, economics is probably more like medicine or biology: complex systems often have 2nd or 3rd order effects that aren't apparent at first. These studies should be seen like phase 1 clinical trials are, promising results lead to larger and larger trials. There's potentially a lot to be gained if UBI works at scale, I'm optimistic to see more ambitious studies.
bsder 6y ago
> There is no law of nature saying that for any system you wish to study, there is a simpler, easier to understand system you can study instead which should roughly replicate the results in your bigger messier system. Human economies are complex. Perhaps true, but you don't have any evidence to support this assertion either. Physicists didn't launch ITER before doing a lot of experiments at a lower scale that they thought might scale up. Human economies do have models and rules--and often fairly simple ones model behavior quite well. From my point of view, the first hurdle is making sure that UBI doesn't do active harm. A lot of "means-based" support DOES do active harm. The hard cutoffs have fairly severe perverse incentives that are now well-documented. Most of the UBI experiments seem to be indicating that they don't cause similar perverse incentives (that is the whole point of UBI but it's nice to have some evidence). That's important because then the argument can be made that current "means-based" systems should be transitioned to either having a much larger limit or toward a UBI-like mechanism. Those have much more staying power than a UBI experiment. I don't do this to "prove" that UBI works to the dog-whistlers--they will shout "welfare queens" until doomsday and simply can't be reached by logical arguments. However, I would like to have some evidence that we are allocating the money that exists relatively effectively and that there isn't a better choice for doing so.

chillacy 6y ago

Agreed, economics is probably more like medicine or biology: complex systems often have 2nd or 3rd order effects that aren't apparent at first. These studies should be seen like phase 1 clinical trials are, promising results lead to larger and larger trials. There's potentially a lot to be gained if UBI works at scale, I'm optimistic to see more ambitious studies.

bsder 6y ago

> There is no law of nature saying that for any system you wish to study, there is a simpler, easier to understand system you can study instead which should roughly replicate the results in your bigger messier system. Human economies are complex. Perhaps true, but you don't have any evidence to support this assertion either. Physicists didn't launch ITER before doing a lot of experiments at a lower scale that they thought might scale up. Human economies do have models and rules--and often fairly simple ones model behavior quite well. From my point of view, the first hurdle is making sure that UBI doesn't do active harm. A lot of "means-based" support DOES do active harm. The hard cutoffs have fairly severe perverse incentives that are now well-documented. Most of the UBI experiments seem to be indicating that they don't cause similar perverse incentives (that is the whole point of UBI but it's nice to have some evidence). That's important because then the argument can be made that current "means-based" systems should be transitioned to either having a much larger limit or toward a UBI-like mechanism. Those have much more staying power than a UBI experiment. I don't do this to "prove" that UBI works to the dog-whistlers--they will shout "welfare queens" until doomsday and simply can't be reached by logical arguments. However, I would like to have some evidence that we are allocating the money that exists relatively effectively and that there isn't a better choice for doing so.