That money has to come from some where. If you're paying janitors more and they're no more productive than before, someone else is making less.

Yes, that is an apt description of a reduction in income inequality, i.e. the intended effect.

No telling if that would be the only effect. You could reduce income inequality while also increasing poverty, so I don't see the reduction of inequality as an end in and of itself.

> You could reduce income inequality while also increasing poverty

So let's think about what it would mean to do that.

In one case you could have some people who are just above the poverty line and if you took a little bit of money from them and gave it to people below the poverty line then the first group could end up below it while the people already below it don't get enough to end up above it. In theory that's "increasing poverty" but only in a strict technical sense which isn't even inconsistent with the result still being a net positive. Also, that wouldn't be possible under plausible rate structures because anybody in the approximate vicinity the poverty line would be receiving more than they pay in taxes and the net payers would be making significantly more than that with no reasonable possibility of falling into poverty as a result.

The other possibility is that the program somehow causes such a dramatic amount of economic damage that more people end up below the poverty line than started there as a result of increased involuntary unemployment etc. But the poverty line in the US is $12,760, whereas the most commonly proposed amount for the UBI is $12,000, so that seems incredibly unlikely -- people would have to be unable to find a job making $760/year, in an environment where employees have greater leverage. And you could preclude the possibility entirely if desired by making the UBI e.g. $13,000, which would just outright directly eliminate poverty.

If this program causes inflation because the cost of it is untenable, then $12K can quickly become worthless. Supply and demand are laws of nature, you cannot escape them. If the economy suffers from lack of productivity as a result of this program, and money must be printed in excess of what goods and services are being produced...then nominal payouts matter not. Hyper-inflation is a real possibility and has ruined more than one empire in history.

The argument that a UBI would cause a significant amount of inflation continues to be nonsense. Necessities do not have perfectly inelastic supply and would not be consumed in significantly larger amounts than they are already. The large majority of the US population already has access to food, shelter and medicine and pulling the remaining <10% of people into the market is not going to cause hyperinflation. There is a pretty good guess that it could relieve a lot of the existing regional high prices by making it more attractive for people to move into lower cost of living areas and use the UBI to offset the lower average wages there, until enough people have done so to bring down the already existing high costs in major cities.

But also inflation.

Inflation refers to the value of the currency, which would not be directly affected.

If different people had money than currently do then they might buy different things with it, and then the demand for the things they want would go up and demand for some other things would go down. That isn't inflation, it's a demand shift. And for most types of products it would only cause increased production of those things rather than significant price changes. Meanwhile for other things (e.g. things which are artificially scarce), the price changes themselves would shift demand elsewhere -- who is going to pay $10,000 rent in San Francisco if it's <$1000 in Pennsylvania and the difference in local wages doesn't close the gap?

Yes, hopefully some executives whose salaries have been increasing far faster than the janitors' salaries for the past 50 years. That's the point!!