Three MMT fallacies
By Scott Sumner
I see three statements repeated by Modern Monetary Theory proponents, almost like mantras:
1. Money is endogenous
2. Banks don’t loan out reserves
3. There is no money multiplier
All three of these statements are either false, misleading, or meaningless, depending on how you define terms.
1. Endogeneity: Everyone has their reasons
When economists say a variable is endogenous, they mean it is explained by other variables in the model. Endogeneity is not an intrinsic characteristic of a variable, in the way that an apple is red or water contains hydrogen and oxygen atoms. Rather we find it convenient to regard variables as endogenous for the purposes of a certain analysis. Thus one should never say, “You’re wrong; money is endogenous”, rather you might want to claim that, “For the purpose of your analysis, it is more useful to regard money as endogenous.”
Monetarists often view the monetary base as being determined exogenously by the central bank, that is, at the bank’s discretion. At the same time, they understand that if the central bank is pegging some other variable, say exchange rates or interest rates, then the central bank has no discretion to adjust the money supply independently. They might still believe that changes in the money supply under that regime are very impactful, but there is no policy discretion for the quantity of base money. Base money is endogenous.
Keynesians often regard the monetary base as being endogenous during a period of interest rate targeting, although with the advent of IOER the central bank can target the base and the interest rate independently. In Singapore, the central bank targets the exchange rate, and regards both interest rates and the monetary base as endogenous.
Things change if the central bank stops pegging interest rates at a constant level and instead targets them at a level that is frequently changed. While in that case the base can still be viewed as endogenous for the period when rates are fixed, it’s equally accurate to argue that the central bank adjusts its target interest rate in such a way as to allow desired changes in the base. Thus a central bank intending to do an expansionary monetary policy might cut the interest rate target in order to increase the monetary base. In that sense, they still do have some control over the money supply. The money supply can be viewed as exogenous over a period of months.
All of this nuance is lost in MMT descriptions of monetary policy. Interest rates are viewed as exogenous and the base as endogenous. Any alternative approach is viewed as unthinkable.
To an omniscient God, everything in the universe in endogenous. Everyone has their reasons. Claiming that something is “exogenous” is equivalent to claiming that we don’t fully understand the process by which it is determined. Thus interest rates might look exogenous to one economist, while another sees them as being determined by the central bank’s 2% inflation target. Indeed, the entire “Taylor Rule” literature can be described as an attempt to model interest rates endogenously.
2. It’s a simultaneous system
When a bank makes a loan, it typically gives the borrower a bank account equal to the value of the loan. If the borrower withdraws the money and spends it on a new house, the seller typically takes the funds and deposits them in another bank. That’s the sense in which MMTers argue that banks don’t loan out reserves; the money often stays within the banking system. The exception would be a case where the borrower withdrew the borrowed funds as cash.
My problem with the MMT analysis is that it often seems too rigid, with claims that the banking system has no way to get rid of reserves that it does not wish to hold. That’s true of the monetary base as a whole (cash plus reserves), which is determined by the Fed. But it is not true of bank reserves in isolation. There are two ways for banks to expel undesired excess reserves, a microeconomic approach and a macro approach.
The micro approach is to lower the interest rate paid on bank deposits and/or add service charges of various sorts. This encourages the public to hold a larger share of its money as cash and a smaller share as bank deposits. On the other hand, it’s not clear that this process would constitute “lending out reserves”.
The macro approach better describes what economists mean by lending out reserves. Assume the economy is booming and people are borrowing more from banks. Continue to assume a fixed quantity of base money. If the borrowed money comes back to banks as increased deposits, then banks can make even more loans and create even more deposits. Over time, this will increase the aggregate level of both deposits and loans, putting upward pressure on NGDP.
In the 106 years after the Fed was created at the end of 1913, the currency stock grew by 516-fold, (not 516%, it’s actually 516 times as large.) NGDP was up 549-fold. The currency to GDP ratio does move around over time as tax rates and interest rates change, but clearly the demand for currency is at least somewhat related to the nominal size of the economy. When NGDP grows, currency demand will usually rise. Thus, in aggregate, a banking system that makes lots more loans will gradually lose reserves as NGDP rises, holding the overall monetary base constant.
As is often the case, Paul Krugman expressed this idea more elegantly than I can:
When we ask, “Are interest rates determined by the supply and demand of loanable funds, or are they determined by the tradeoff between liquidity and return?”, the correct answer is “Yes” — it’s a simultaneous system.
Similarly, if we ask, “Is the volume of bank lending determined by the amount the public chooses to deposit in banks, or is the amount deposited in banks determined by the amount banks choose to lend?”, the answer is once again “Yes”; financial prices adjust to make those choices consistent.
Now, think about what happens when the Fed makes an open-market purchase of securities from banks. This unbalances the banks’ portfolio — they’re holding fewer securities and more reserve — and they will proceed to try to rebalance, buying more securities, and in the process will induce the public to hold both more currency and more deposits. That’s all that I mean when I say that the banks lend out the newly created reserves; you may consider this shorthand way of describing the process misleading, but I at least am not confused about the nature of the adjustment.
MMTers have a bad habit of assuming that mainstream economists are clueless, just because we use a different framework.
3. There are a million money multipliers
Krugman’s explanation also helps us to understand the confusion over money multipliers. Injecting more money into the economy sets in motion forces that boost the nominal quantity of just about everything, not just bank loans and bank deposits. An exogenous and permanent doubling of the monetary base will double the nominal value of every single asset class, from one carat collectable diamonds to Tesla common stock to inventories of soybeans to houses in Orange County to rare stamps. And it will also double the monetary aggregates. That’s because money is neutral in the long run, so doubling the money supply leaves all real values unchanged in the long run.
So the money multiplier for any asset class is merely the nominal stock of that asset dividend by the monetary base. No serious economist believes the M1 or M2 money multiplier is a constant, and indeed textbooks usually explain it this way:
mm = (1 + C/D)/(C/D + ER/D + RR/D)
It’s one plus the ratio of cash and bank deposits divided by the cash ratio plus the excess reserve ratio plus the required reserve ratio. Then economists model the money multiplier by describing the factors that cause these three ratios to change over time. In my view, the money multiplier model is pretty useless, as I don’t view M1 and M2 aggregates as being important. Your mileage may vary. But there’s nothing “wrong” with the model; the question is whether it’s useful or not.
The one money multiplier that does matter is NGDP/Base. Unfortunately, both IOER and the recent zero interest rate episodes have made that multiplier more unstable. I favor a monetary policy where the NGDP multiplier (aka “velocity”) would be more stable. No more IOER and fast enough expected NGDP growth to assure positive interest rates.
4. Beware of “realism” and the fallacy of composition
Sometimes you’ll encounter an economist who is very proud that he or she understands how the financial system works in the “real world”. And obviously that knowledge can be useful for certain purposes. But the banker’s eye view often misses what’s most important in macroeconomics, the general equilibrium connections that Krugman alluded to in his “simultaneous system” remark.
If the Fed gave me a check in exchange for an equal quantity of T-bonds, I’d be no richer than before, no more likely to go out and buy a new car. And if I took that check and deposited it in a bank, that bank might be no more likely to make a business loan. They could simply buy a bond, or lend the reserves to another bank. But as everyone tries to get rid of the base money they don’t want, subtle changes begin to occur in a wide range of asset prices, which will eventually push NGDP higher. If wages are sticky then the extra NGDP will lead more people to go out and buy cars.
Just not me, not the person who first got the new Fed-created money.