[szermer]

I read the Oakland trial as a quick, local platform to determine what the best cadence for delivery and data gather would be.

Identifying the little bugs early (Visa gift cards versus Amex gift cards prove to be a better option for folks that don't have bank accounts...etc) while streamlining the reporting structure will allow the larger, long term experiment to run smoother.

[BinaryIdiot]

You're correct but that isn't the issue the author is bringing up.

The author is concerned with the 5 year "long term" study. Consider the following:

- You know you're going to get, say, $2,500 a month for the next 5 years.

- You know you're going to get, say, $2,500 a month for the next 15 years

- You know you're going to get, say, $2,500 a month for the rest of your life.

Do you plan each of those the same? Differently? You'll likely plan those completely differently because two of them have a maximum amount of payout whereas the last one, for the rest of your life and what real basic income would be, would not have a maximum.

Unfortunately this means the experiment is flawed from the start. You won't be able to apply effective controls so that people who know the maximum payout want act differently than someone who expects it forever.

Basic income is almost impossible to test with proper controls unless you want a valid but very, very long term study of probably several decades.

[Spoom]

I know if I was given $30,000 a year for free with the understanding that it would entirely stop after five years, I would try my damnedest to not let it affect my lifestyle, essentially treat it as an early retirement fund source, and invest it. Perhaps most people wouldn't plan as I would, but I would imagine that the cut off would have a significant effect on at least some of the participants.

I don't necessarily agree that the study is unethical; the participants, one assumes, are told of the cut-off before the start. I do think that the outcomes might be less valid.

[pdq]

For a valid experiment, the term should be a control variable on subjects:

- 0 years (ie no income, as per current)

- 1 year

- 5 years

- 15 years

- life

For example, it's possible subjects knowing they will get zero or one year of BI could have better outcomes than ones who know they have a lifetime of BI. Or not.

[avs733]

exactly, this is entrepreneurial MVP (minimum viable policy) at its finest. It is not intended to take a purely rigorous, scientific, experimental, and deconvolved approach. It is intended to be a trial to learn something to move the project forward.