[dragonwriter]

> I worry that it would be a lot worse than that - that we could create a class of non-workers and workers.

UBI is intended to alleviate the problem with means-tested social welfare programs in doing that by reducing the disincentive to taking action which would produce additional income faced by those on means-tested benefits where any additional income results in substantial (sometimes even dollar-for-dollar or greater) reduction in value of benefits received.

And, in any case, economically its somewhat self-limiting (unless you explicitly inflation index the benefit, in which case you are just asking for disaster if you set it wrong initially -- which is why I prefer tying benefits to a defined revenue stream, rather than a set benefit level): if too many people drop out of the work force because the benefit level is too generous given the current level of productivity, it will accelerate inflation, reducing the real level of the benefit, leading more people back into the workforce.

[ultramancool]

> unless you explicitly inflation index the benefit

You can be 100% sure that unless this is specified you'll have people lobbying for it every year. Or just flat rate increasing the UBI, etc. I can envision a strong push from some political groups for this already.

> if too many people drop out of the work force because the benefit level is too generous given the current level of productivity, it will accelerate inflation, reducing the real level of the benefit, leading more people back into the workforce.

This is a good point, but in this case, everyone essentially loses value from savings to pay the UBI. Definitely not a hit I'm interested in taking.

[dragonwriter]

I actually agree with a lot of what you say, which is why I prefer to start a UBI very low (not directly displacing existing programs, but since the income from it counts as income, gradually reducing eligibility for them), tied to a dedicated revenue stream that is expected to grow with productivity (which, unless something else is wrong, should over the long run grow substantially faster than inflation), rather than using a benefit-first calculation.

I'd really prefer to take strong steps to eliminate the risk of overshooting a sustainable benefit level, but -- while there are certainly costs to that -- the "mass exodus from the work force because benefits are too generous" problem is self-limiting.

[ultramancool]

Yeah, that's a fair assessment. I think more research needs to be done in this direction, there's got to be a good way to determine initial parameters for this sort of thing.

But I can't help but think that it'll be too politically charged to follow any sort of logical regime that might result from research, which makes it sort of scary to approach. You'd basically need the public to agree on the terms and limits of a well-defined system without turning it into a partisan mess. Unfortunately I can't think of many examples of this currently in existence.

[comicjk]

Indexing UBI to a revenue stream is a bad idea because the revenue will tend to swing up and down in sync with economic cycles, making bubbles frothier and recessions worse. Constant spending by contrast acts more like an automatic stabilizer. https://en.wikipedia.org/wiki/Automatic_stabilizer

[dragonwriter]

Indexing to revenue is a bit of a simplification. what I really think is the revenue stream should be split between current benefits and a reserve fund, with the reserve fund drawn from to avoid decreases in cyclical downturns. The idea is that the long-term effect should be increasing benefit with productivity, while avoiding short-term benefit drops.

And the effect you describe is really one of overall domestic spending, not one program in isolation.