[librvf]

It seems to me that a limited, 5-year Basic Income experiment is fundamentally flawed and will, at best, fail to predict anything important about what system-wide BI would have on society and at worst will be disastrously misleading and have an incredibly adverse effect on public policy.

What I'm reading scares me.

Oakland is a city of great social and economic diversity, and it has both concentrated wealth and considerable inequality. We think these traits make it a very good place to explore how basic income could work for our pilot.

To me this screams complete lack of any meaningful experimental controls.

If the pilot goes well, we plan to follow up with the main study. If the pilot doesn’t go well, we’ll consider different approaches.

What does this mean? If the pilot gives you the results you want? What kind of science is this?

[skoocda]

There needs to be some sort of action, though. Pondering all the possible effects has been going on for some time now, and simulating a perfectly spherical BI in a vacuum is not going to be any more useful than a 'fundamentally flawed' empirical experiment.

>What kind of science is this?

It's the normal scientific method. They have a hypothesis. They're going to test it in the real world. And the results will be analyzed. What's the alternative? If the pilot doesn't go well, and they take the same approach again, that'd be the very definition of insane.

[librvf]

Well, it's "not the normal scientific method." Although it's arguably a part of it. My objection may have been a bit strenuous, as the key word here is "pilot study." This is a pre-study study. Their only "hypothesis" here is "is a basic income experiment with real people even at all feasible"

They don't yet have a real hypothesis about basic income, at least not that they have presented. The questions here (https://blog.ycombinator.com/basic-income) are rhetorical and vague, which make for very bad hypotheses.

Personally I do not believe this study, as I see it described now, will yield any useful scientific knowledge. The results, if they seem favorable, will be used to influence public policy without regard to actual scientific merit and society will be subjected to the real experiment.

I would be much less concerned if I saw someone with a rigorous background in mathematics and economics developing an actual predictive model that could be falsified with specific tests. When the inputs are vague ("people") and the conclusions are subjective ("are people happy?") it is not scientific method.