Milton Friedman wasn’t a supporter of a basic income — though you may have been told that he was. A lot of advocates for unconditional cash payments use Friedman’s name as a sort of inverted appeal to authority, drawing on the opinions of someone they dislike, to show that the policy is so very reasonable that even flinty, small-state libertarians are in favour.
Friedman was in favour of his own version of a Negative Income Tax (NIT) quite clearly laid out his system — and to be fair, there are some similarities, but I don’t think they’re the same thing.
Both a guaranteed basic income and an NIT can have a constant marginal tax rate, eliminating the huge marginal rates that often hit people moving from jobless benefits into low-paid work. If guaranteed income and post-break-even tax rate are high enough, the two can be effectively the same on a mechanical level. But suggesting that would ignore what Milton Friedman actually said about the issue. Take for example, a 1968 article in Newsweek by Friedman:
“By varying the break-even income and the negative tax rate, by adding the negative income tax to present programs rather than substituting it for them, it is possible to go all the way from the rather modest and, I believe, eminently desirable plan just outlined to irresponsible and undesirable plans that would involve enormous redistribution of income and a drastic reduction in the incentive for people to work. That is why it is possible for persons with so wide a range of political views to support one form or another of a negative income tax.”
Here, Friedman seems to be laying out quite specifically what he does not want his proposal to be confused with — a basic income system. For some, the whole point, or at least one of the major reasons to support a basic income policy is to guarantee redistribution, to offset fears of a shrinking wage share or increased robotisation or just because they believe it’s the just thing to do.
For others, the idea is to replace the existing welfare system entirely, with something more simple, and perhaps much more parsimonious. The figures used in Friedman’s 1968 Newsweek article suggest a $1,500 payment for households with no other income — for reference, just over three per cent of families had an income of less than $1,500 in mid-1969, and the median household income was around $8,632.
Friedman argues that “the precise floor set would depend on what the community could afford” in Capitalism and Freedom, but it seems reasonable to expect that he knew what the average income was in the year he was writing, and that he believed the quantities he referred to were roughly in line with his desire to incentivise work.
Though I’m a big Milton fan, I think glossing over what the level of the payments is a big mistake. Elaborating on details like how much a policy is going to cost is a pretty important part of a sincere campaign. Even an NIT programme designed to have exactly the same redistributive effect as a basic income policy would cost half as much to administer in the US. absence of important elements like this is why UBI advocates have made very little progress over decades of well-intentioned advocacy.
To my mind, there’s a simple test for how much Friedman’s NIT and basic income are alike: if you were asked to rank your country’s existing welfare system, Friedman’s system, and a more generous unconditional basic income system from most to least preferable, how would you do it?
If you prefer the existing system to Friedman’s, but would prefer UBI to the status quo, you may want to take a look at the absurdly broad caucus in favour of “basic income”— likewise if you would rank Friedman’s first and UBI last. Personally, I would rank both above the existing welfare system for their mechanical simplicity. Perhaps I am incorrect, and lots of advocates would do this, but I feel I am probably in a minority.
