MMT ≠ QE
MMT is not and has never been about getting central banks to "print money" for the government
I have to teach a class soon, but I wanted to quickly comment on a media clip that drew some attention yesterday.
The clip is from Bloomberg TV, and it features the host of Bloomberg Surveillance, Tom Keene, talking with Bob Michele, the CIO at JPMorgan Asset Management. Tom starts off the segment by saying:
“Stephanie Kelton at Stony Brook [University] has changed the world. She came out with Modern Monetary Theory. We are in an MMT experiment of some type as well, and the financial media is not talking about this enough because everybody sort of wishes the theory would go away which is unfair to something that has had such an impact.”
The conversation continues with both men suggesting that the MMT “experiment” involves not just an embrace of substantial fiscal support (and large deficits) to fight the pandemic—which I accept—but also monetary support in the form of central bank bond-buying (QE) to keep the spending “affordable.”
While I’m happy to accept credit for helping to shift the terms of the debate in the realm of macroeconomics, especially with respect to the mechanics of government finance and the limits on government spending, it is Warren Mosler who deserves credit for “coming out with” MMT (although the MMT label came many years later).
My first encounter with many of the core tenets of MMT came from reading Mosler’s Soft Currency Economics about 25 years ago. In the years that followed, a small number of economists—myself included—worked to build on Mosler’s early insights. For me, one of the most fascinating arguments in Warren’s book had to do with the sequencing of the government’s taxing, borrowing, and spending.
Like almost everyone, I had been brought up to think of taxes and borrowing as two competing ways for the government to gets its hands on the money it needs to finance its spending. Taxing And Borrowing came first. Spending came last.
In my book, The Deficit Myth, I offered this simple pneumonic to describe the conventional model: (TAB)S
This model is affirmed in all mainstream macro textbooks, where students are presented with the concept of a government budget constraint. It teaches students that, much like a household, the government is financially constrained, so it must find a way to secure financing before it can pay the bills. There’s often a passing reference to a third financing option—”printing money”—but that one gets quickly dismissed as inherently inflationary, leaving students with the idea that the government must rely on either tax revenue or borrowed funds to pay for its spending.
Mosler flipped all of that on its head. He explained that not only did everyone have the sequencing backwards but that we were thinking about taxes and bond sales the wrong way as well. The funds to pay taxes and buy government bonds, he explained, come from the prior act of government spending (or lending). Here’s Mosler:
“The government spends money and then borrows what it does not tax, because deficit spending, not offset by borrowing, would cause the fed funds rate to fall.”
The mental model we should keep in our heads, then, is the one that sequences the spending first: S(TAB)
The point I’m making here is that from the very beginning, MMT has offered a (superior) descriptive framework, one that explains the actual mechanics of government finance. It was never a proposal to “print money” or to encourage central banks to engage in large-scale asset purchases (LSAPs). In fact, MMT scholars were some of the earliest skeptics of Quantitative Easing (QE).
The reality is that when it comes to covering the government’s bills, there is only one way to pay, as I explained in this recent Substack:
Unlike the rest of us, Congress never has to check the balance in its bank account to figure out whether it can afford to spend more. As the issuer of the currency, it doesn’t have to worry about running out of money. It can afford to buy whatever is available and for sale in its own currency. That might involve spending on roads and bridges, a military arsenal, or hospitals and schools. Finding the votes to pass a spending bill can be hard, but finding the money is never a problem. They just create it.
Here’s how it works. Whenever Congress and the president agree to spend more, the government’s bank—the Federal Reserve—works with the rest of the financial system to get that money into our accounts. Everything happens electronically, so there’s no physical “printing” of money involved. If you got a $1,400 check from the federal government earlier this year, or if your company received money to help cover payroll and other expenses, then you got some of the newly-minted digital dollars that were created to support our economy. No taxpayers were involved in the process. It was all done using nothing more than a computer keyboard.
There’s just no other way for it to work. It has nothing to do with QE ! So please, don’t conflate MMT with QE. And raise a red flag whenever you hear anyone present MMT as a proposal to “print money.”
Correct me if I'm wrong. Are you saying QE requires the government to pay interest because you are selling debt in the form of treasuries, and the Fed depositing (spending directly with key strokes) does not cost the government interest payments?
Please do enlighten me on the mechanics of Federal Government expenditures.
If government wishes to engage in the supply of some public good, how does it obtain the resources necessary to do the job? Does it wave its magic wand and the resources appear? How does it obtain the labour and cooperation of all those firms and workers with a bank account balance of zero at the Fed?