Krugman wonks out: what we talk about when we talk about money

Lately I've been getting a lot of mail from readers infuriated by my relative nonchalance about two issues: budget deficits and money growth. When I point out that the federal government is able to borrow at incredibly low interest rates, some retort that this is only because the Federal Reserve is buying a lot of debt. When I say that we shouldn't be worried about runaway inflation, some point to the rapid growth in the money supply and say that we're on the verge of becoming Venezuela.

These are actually related complaints — and both, I'd argue, reflect common misunderstandings about what's actually going on both with the Fed's balance sheet and with the "money supply." (Scare quotes explained shortly.)

So, what makes an asset money? There's no ineffable essence that makes green pieces of paper bearing portraits of dead presidents money, whereas vintage comic books aren't. Money is defined by what it can do — above all, serve as a medium of exchange, something you accept in return for what you have, and then hand over for what you want.

To play this role, an asset must also be a reasonably stable store of value — not losing or gaining 30% over the course of a day. And a widely used medium of exchange also becomes a unit of account; we calculate profits and debts, make financial deals, in dollars — not with promises to hand over, say, a certain number of sheep.

Since money is a role, not a thing, does it even make sense to calculate the quantity of money? Yes, under certain circumstances. Tracking the number of dead presidents in circulation sometimes helps predict inflation. Tracking broader "monetary aggregates," which include things like bank deposits that can also be used for payment, may also be useful.

In their landmark 1963 book "A Monetary History of the United States," Milton Friedman and Anna Schwartz convinced many economists that M2, a measure that included both checkable deposits and other bank deposits that could easily be transferred to checking accounts, was a powerful economic predictor — so much so that the Federal Reserve could basically cure the business cycle simply by keeping M2 growing slowly but steadily. In particular, they argued that the Fed could have prevented the Great Depression if it had prevented M2 from falling 30% from 1929 to 1933.

But here's the thing: The Fed doesn't directly control M2. All it controls is the "monetary base," the sum of bank reserves and currency in circulation. (It only controls the sum, not the individual components, since people can decide to withdraw currency from banks or put it back.) What Friedman and Schwartz asserted was that the Fed can control M2 indirectly — that it can push monetary aggregates up or down if it's willing to move the monetary base enough.

The 2008 financial crisis wasn't kind to that view. The Fed hugely increased the monetary base, but banks basically just sat on the additional reserves so that deposits and hence M2 didn't rise much.

This in turn very much calls into question the notion that the Fed could have prevented the Great Depression. But that's another story.

More to the current point, M2 has in fact soared during the pandemic.

And the Fed has indeed bought a lot of government debt. But is the Fed really financing the budget deficit?

Not really. At a fundamental level, households are financing the deficit; the funds being borrowed by the government are coming out of the huge savings undertaken by families saving much of their income in an environment where much of their usual consumption hasn't felt safe.

However, household financing of the deficit isn't direct. Instead, it has taken the form of a sort of financial daisy chain. Families are stashing their savings in banks. Banks, in turn, have been accumulating reserves — that is, lending to the Fed, which these days pays interest on bank reserves. And the Fed has been buying government bonds.

More or less, households have acquired $2 trillion in deposits; banks have acquired $2 trillion in reserves; and the Fed has acquired $2.5 trillion in government securities. Wait, what's that extra $500 billion? It appears that someone has been stashing away huge amounts of currency — probably mostly $100 bills, probably mostly outside the US I guess not all Russian gangsters trust Bitcoin.

Exactly why the process has been so indirect is an interesting question. A guess is that private players are worried about liquidity, about having quick access to their funds if necessary. So both families and banks want deposits they can draw down in a pinch, not Treasury securities that might be slightly harder to liquidate.

In any case, there are two points about this process. First, it says that low interest rates aren't the result of artificial manipulation; there really are a lot of savings with nowhere to go, which are being made available cheaply to the government. Second, because the daisy chain of lending runs through bank deposits, it shows up in the measured money supply. But it isn't really a monetary expansion in the sense many people imagine. The Fed isn't the Venezuelan government printing bolívars to pay its soldiers; it's basically acting as a financial intermediary for investors who want to park their money somewhere safe.

And while there are plenty of reasons to worry about what's going on in the US economy, Fed purchases of bonds and rising M2 aren't on the list. Chill out.

©2021 The New York Times Company