A variety of elected officials are currently proposing that the federal government send cash payments to every person in the country (sometimes called a “universal basic income” or “UBI”) as a way of responding to the recession. These calls have resurrected claims that a UBI is inflationary. Such claims circulate widely among advocates of workfare and in certain niche internet circles. But these claims are nonsensical and deceptive. In this post, I walk through these claims to promote better understanding of the issue.
Many people who claim that UBI is inflationary also belong to the Modern Monetary Theory (MMT) community. One of the key rhetorical moves of that community is to talk about government spending as if it separate from government revenue (e.g. taxes). This way of speaking goes against convention and therefore tends to mislead people who are used to conventional ways of talking about government programs.
It is precisely this rhetorical move that is behind the claim “a UBI is inflationary.” What they actually mean when they say that is “disbursing UBI payments without collecting offsetting taxes is inflationary.” But they never explicitly say the “without collecting offsetting taxes” part because, as mentioned already, they have all decided to talk about spending as if it is separate from taxes.
This is a fun semantical game to play I suppose, but that’s all it is: a game. Everyone who advocates for a UBI as a permanent fixture of the economic system also advocates for using some set of taxes or other revenue to fund it (or, in MMT language, “offset it”). For those advocates, the taxes are packed into the word “UBI.” Thus when those advocates say “a UBI is not inflationary,” what they mean is that “a UBI with offsetting taxes is not inflationary.”
If you’ve followed along so far, you see now that everyone has actually wound up at the exact same spot. Both claims are true:
- A UBI [without offsetting taxes] is inflationary.
- A UBI [with offsetting taxes] is not inflationary.
Everyone is just talking past one another, primarily because the MMT people are using an idiosyncratic way of speaking while being opaque about doing so.
Transfer Payment Nonsense
In addition to the word games above, there is another strand of argument that circulates in this MMT world that is just straightforwardly nonsensical. Though they rarely spell it out so clearly, the argument goes like this:
- Transfer payments are inflationary.
- A UBI is a transfer payment.
- Therefore, a UBI is inflationary.
Right off the bat, you can see that this argument also contains within it the rhetorical move where you assume that “transfer payments” or “UBI” are being done by themselves with no offsetting taxes (i.e. being done with seignorage). But more than that is going on in this particular point.
Transfer payments are payments governments make (almost always in cash) to various populations as part of the welfare system. Social Security and SSI payments are transfer payments. Unemployment benefits are transfer payments. Pell grants are transfer payments. Food Stamps are transfer payments. And so on.
All transfer payments are “inflationary” in the same way that MMT people accuse the UBI of being inflationary. Many people don’t realize this because, though you can find this point buried in various MMT writings and presentations, the popular advocates are very selective about when they bring up inflation. A proposal to increase transfer payments via a UBI will always cause MMT people to talk about inflation. But a proposal to increase transfer payments via an increase in Social Security benefits won’t. There is no theory-based reason for this selectivity: all transfer payment increases cause inflation, so defined. But it’s a political calculation they make.
When accusing transfer payments of being inflationary, the MMT people very often also say that government spending on public employee wages is not inflationary. This becomes a key reason why you are supposed to favor a job guarantee, which pays people “wages,” over more conventional transfer programs like unemployment benefits, which pays people “transfer payments.”
This is completely nonsensical of course. Government spending on public employee wages is inflationary, so defined. It is no different than government spending on transfer payments in this regard. In both cases, the government gives an individual cash that they then go out and spend on goods and services. It makes no difference for inflation if the person who received the cash sat at home and watched TV (“transfer payment”) or instead raked leaves in the park (“job guarantee wages”). The inflation (again, so defined) comes when they spend the money trying to buy good and services.
The reason some JG advocates don’t seem to understand this is because their minds have been captured by a very literal interpretation of national accounting and the quantity theory of money. Put simply, there is a general notion that inflation occurs when there is more money chasing the same number of goods. With that notion in mind, it is then observed that public employee wages are counted in the national accounts as output, meaning they boost GDP. But transfer payments are not counted that way. So, the reasoning then goes that public employee wages increase output and therefore lead to “more money chasing more goods” (not inflation) while transfer payments do not increase output and therefore lead to “more money chasing the same number of goods” (inflation).
This is simply confused. To understand why, consider the case of public school teachers. There are a little over 3 million public school teachers in the US and they collectively receive a little under $200 billion in wages. In the national accounts, that $200 billion will be counted towards GDP and reasonably so: teachers probably do produce about $200 billion worth of educational services each year.
But the government does not charge students or their parents for those services. It does not collect any revenue from those services. And because it does not collect any revenue from the users of the service, those $200 billion of wages are not offset at all. They are purely inflationary just like an unemployment check would be and just like JG wages would be, which is why we need taxes to fund public schools (or “offset the inflation caused by teacher wages” to again use the special language of MMT).
If you press some of the high priests of the MMT world hard enough, they will sometimes admit this point. For instance, in 2016, Rohan Grey (a second tier MMT advocate) pushed Randy Wray (one of the high priests of MMT) to address this question directly. Grey raised the point this way:
If you put $1 trillion into solar panels and those solar panels were just installed — they weren’t sold, they were just given away. That $1 trillion is now out there in people’s pockets and presumably private sector demand is going to absorb that. It seems to be that the product of the job guarantee, unless its sold, is not necessarily absorbing the purchasing power demands being created by the income that’s being given for it.
Wray responds to this point, not by trying to show that through some magic JG wages won’t be inflationary, but instead by saying that a JG program would actually be very small, something like 1 percent of GDP. And so while JG wages would be inflationary (just like UBI payments are inflationary), the inflation is not something to worry about because the program is simply not very big.
Hyman Minsky, who is revered in MMT world, also acknowledged that the JG (or a public jobs program more generally like the WPA) is clearly inflationary unless its output is sold to generate government revenue:
If the WPA output is useful and is sold, WPA-type relief for the unemployed is less inflationary than unemployment insurance. If the WPA output is useful even if it is not sold in a market, then it makes a contribution to the well-being of those who find the output useful, and, presumably, a tax or user’s fee offset to all or part of the WPA spending could be collected.
So in summary, this particular version of the argument that the “UBI is inflationary” is really just an application of the broader argument that “transfer payments are inflationary.” And while it is true, under the way MMT people use words, that transfer payments are inflationary, it is also true that all other government spending is inflationary, including spending on public employee wages and including spending on JG wages. The whole line of critique is one big head fake.
The last kind of argument you see about UBI and inflation basically goes like this: if everyone received $1,000 extra each month, then real estate rents would go up by $1,000 each month and nobody would receive a net benefit.
There are two basic problems with this argument.
First, if you believe that rent inflation will gobble up all income gains, then that is not an argument against the UBI. That’s an argument against all income increases from any source. Increasing the minimum wage by $6 per hour will also increase the incomes of minimum wage workers by $1,000 per month. If the argument is true, then real estate rents will go up in areas with a lot of low-wage workers and simply gobble up any minimum wage increase. Similarly, if the JG hugely increases employment, then real estate rents will go up in areas that currently have a lot of unemployment and simply gobble up the JG wages. I could go on.
Second, it’s just simply not true in general that rents gobble up all income increases. If that were true then we would all be paying over 90 percent of our income in rent. But we aren’t.
It is possible of course for a certain area to see rents rapidly run up as area incomes increase. We saw this in North Dakota in oil boom towns when massive oil worker salaries corresponded with rent hikes so high that, at one point, Williston, North Dakota had the highest rents in the nation. These rents have since come down as the pace of apartment building in Williston increased ten-fold a bit after the oil rush arrived. But this is not a very typical situation.
I suspect the only reason this particular inflation argument has any traction at all is because so many people in the discourse live in the few high-income, high-rent areas where real estate rents really do march up pretty reliably with income (not so much that they gobble up 100 percent of income increases, but still a large portion of them). But the problem with these areas is not that the income is too high. It’s that their real estate policy is wildly dysfunctional. The solution to these area’s woes is not opposing transfer payments like UBI, but rather building out more units and imposing rent controls.
While I think it’s helpful to walk through the specific linguistic and conceptual confusions at the heart of these inflation arguments, you ultimately do not really need to understand any of that stuff to see how obviously nonsensical the arguments are. For those wanting a simpler way to think about all of this, it should suffice to simply observe that a “UBI” already exists in various forms without any sort of troubling inflation.
This is literally true in that Alaska has been running a UBI program for decades without a problem. It’s also approximately true in that the US government, and especially governments elsewhere in the world, already have UBI-like programs. The Social Security Administration sends out checks each month to 64 million people, mostly the elderly and disabled. Unemployment Insurance programs send out checks each month to 2.1 million people. In other countries, the government, in addition to paying these kinds of old-age, disability, and unemployment benefits, also pays out a monthly benefit to every child. In the US that would mean checks going out to the families of around 74 million children.
None of this leads to any inflation problems because, as mentioned in the very first section above, taxes are used to fund these programs, just as they would be for any other permanent welfare state expansion.