Great Leap Forward

Krugman Rediscovers the Wheel: Commercial Banks As Creators of Money

From a Tweet by Stephanie Kelton:

MMT: Often imitated, never duplicated

I’ve been lecturing in Bogota and Quito (more on that later) but several people sent to me a recent piece by Paul Krugman in which he attempts yet again to show that he understands banking.

However, as Scott Fullwiler argued after a previous such attempt, whenever Krugman describes the “intermediation” process that banks supposedly engage in, it’s actually like a “flashing neon sign” saying “I don’t know what I’m talking about”.

A year ago, he tried to argue that he, not Minsky, truly understands banking. Banks cannot actually create deposits “out of thin air” because they need reserves for clearing. This time, he rediscovers James Tobin’s classic paper, “Commercial Banks as Creators of Money” (a paper known by everyone who works in the field, so it is hard to believe he forgot about it). In Krugman’s interpretation, this paper proves that you really don’t need to know anything about the mechanics of bank lending. While banks could create deposits, since they won’t stay put on the creator’s balance sheet, in practice, “Banks are just another kind of financial intermediary, and the size of the banking sector — and hence the quantity of outside money — is determined by the same kinds of considerations that determine the size of, say, the mutual fund industry.”

Well, yes. And no.  But confused.

Last time, I argued that his earlier piece set economics back a century. This time, I suppose you could say he’s brought us up to mainstream theory of 1982. There’s been a lot of water under the money theory bridge since then. And, in any case, Tobin was trying to correct the foibles of mainstream theory—there were plenty of nonorthodox economists who didn’t follow orthodoxy down the “exogenous” money path.

As I discussed in my 1990 book, Money and Credit in Capitalist Economies: the endogenous money approach, the 1980s debate over “exogenous” versus “endogenous” money was in a sense a reprise of the Banking School-Currency School controversy of the early 19th century. The Banking School took the endogenous money side, while the Currency School was a precursor to Milton Friedman’s Monetarist exogenous money. The earlier debate was settled when Britain tried to follow Currency School doctrine; the later debate was settled with the adoption of Monetarism in both the US and UK in the late 1970s. In both cases, exogenous money flopped. Endogenous money emerged victorious. As it turns out, the central bank cannot control the money supply. No central bank in a modern developed country even tries any more. Exogenous money is as dead as a doorknob.

The 19th century debate didn’t really have the notion of the deposit multiplier that after the 1920s slowly made its way into US textbooks. Unfortunately, those who haven’t been paying attention don’t recognize that the death of Monetarism was also the death of the deposit multiplier. Endogenous money not only means that the central bank cannot control the money supply but also that it cannot control reserves. It accommodates bank demand for reserves. Hence there is no “deposit multiplier” but rather a “divisor”—there is an ex post ratio of reserves to deposits but it is not constraining.

Now, are banks unique, or are they just intermediaries like money market mutual funds—that take in “deposits” and lend them out?

The answer to both is “NO”. Banks are not unique, and they are not intermediaries. Nor are money market mutual funds, if you get right down to it.

As Minsky always said, “anyone can create money”, but “the problem lies in getting it accepted”.  Both banks and mutual funds “create money” in the sense that they issue money-denominated liabilities. Both have to “get them accepted”. Banks are more special than others because government gives them special protection. When times are good, this might not matter much—the “money” created by other types of financial institutions are “just about as good as” the deposits created by banks. In bad times, that reverses and the liabilities created by shadow banks are about as desired as a fork in the eye. Paul McCulley put it this way:

Over the last three decades or so, the growth of “banking” outside formal, sovereign-regulated banking has exploded, and it was a great gig so long as the public bought the notion that such funding instruments were “just as good” as bank deposits. Keynes provides the essential – and existential – answer as to why the shadow banking system became so large, the unraveling of which lies at the root of the current global financial system crisis. It was a belief in a convention, undergirded by the length of time that belief held: shadow bank liabilities were viewed as “just as good” as conventional bank deposits not because they are, but because they had been. And the power of this conventional thinking was aided and abetted by both the sovereign and the sovereign-blessed rating agencies. Until, of course, convention was turned on its head, starting with a run on the ABCP (asset-backed commercial paper) market in August 2007, the near death of Bear Stearns in March 2008, the de facto nationalization of Fannie Mae and Freddie Mac in July 2008, and the actual death of Lehman Brothers in September 2008. Maybe, just maybe, there was and is something special about a real bank, as opposed to a shadow bank! And indeed that is unambiguously the case, as evidenced by the ongoing partial re-intermediation of the shadow banking system back into the sovereign-supported conventional banking system, as well as the mad scramble by remaining shadow banks to convert themselves into conventional banks, so as to eat at the same sovereign-subsidized capital and liquidity cafeteria as their former stodgy brethren.

But that does not make “shadow banks” or “commercial banks” mere intermediaries. Both “finance” positions in assets by issuing liabilities. Their liabilities are variously called deposits or “NOW” accounts or MMMF shares. Ultimately, however, only insured deposits are guaranteed to never “break the buck”. For that reason, banks are special. This has nothing to do with them being “money creators” versus “intermediaries”—it has to do with sovereign government backing.

And if a bank suffers a clearing drain, it goes to the Fed Funds market to borrow reserves, or to the Fed’s discount window. Often shadow banks have relations with commercial banks to provide clearing for them. A recent—somewhat shocking—discussion at the Fed indicates that the Board is thinking about letting a wide range of firms, including nonfinancials, hold reserve accounts. If that happens, even the shadow banks and the nonbanks might do their clearing directly at the Fed.

Heck, then I guess we could say “anyone can issue money and the Fed will make sure it is accepted”!

I hope Paul Krugman will move on to a 21st century understanding of banking. As I pointed out recently, even the credit raters at S&P have joined us in explicitly throwing out the deposit multiplier. It’s about time.

On another matter, MMT critics often argue that we oversimplify—or are just wrong—when we “consolidate” the central bank and treasury for the purpose of opening our discussion of spending with a sovereign currency. According to critics, the Fed and Treasury are independent and never shall the twain meet. The Fed might go vigilante on the Treasury, refusing to do its banking.


Here are two relevant quotes, one from the Fed and one from McCulley:

Fed: “You have to remember that we are a legal creature of Congress and that we only have a mandate to concern ourselves with the interest of the United States,” Dennis Lockhart, president of the Atlanta Fed, told Bloomberg Television’s Michael McKee. “Other countries simply have to take that as a reality and adjust to us if that’s something important for their economies.”

McCulley (in the paper cited above): “We pretend that the Fed’s balance sheet and Uncle Sam’s balance sheet are in entirely separate orbits because of the whole notion of the political independence of the central bank in making monetary policy. But when you think about it, not from the standpoint of making monetary policy but of providing balance sheet support to buffer a reverse Minsky journey, there’s no difference between Uncle Sam’s balance sheet and the Fed’s balance sheet. Economically speaking, they’re one and the same.”

22 Responses to “Krugman Rediscovers the Wheel: Commercial Banks As Creators of Money”

Neil WilsonAugust 28th, 2013 at 11:43 am

"If that happens, even the shadow banks and the nonbanks might do their clearing directly at the Fed."

Already happening in the UK, where the shadow already has access to clearing accounts.

Check this list of financial institutions:

A list of building societies and private bankers – all with direct clearing capacity at the central bank.

The notion of a clearing bank, and other providers having to go through clearers, is starting to break down.

olly100August 28th, 2013 at 1:21 pm

"A recent—somewhat shocking—discussion at the Fed indicates that the Board is thinking about letting a wide range of firms, including nonfinancials, hold reserve accounts."

do you have a link to this?

OhMyAugust 28th, 2013 at 6:55 pm

I don't think Krugman will understand when you say that both banks and shadow banks create money. For him they convert customers' deposits into other assets to accommodate their "portfolio preferences" and stuff these deposits with somebody else, as "intermediaries". That the deposits come from thin air, from credit creation – this he seems to miss, for him these deposits seem to be like a physical commodity, like beads or chestnuts which then the banking system shuffles around. Maybe that is why he is hung up on the distinction between banks and shadow banks. How are you an "intermediary" when you help somebody expand his balance sheet?

pebirdAugust 29th, 2013 at 3:47 am

"A recent—somewhat shocking—discussion at the Fed indicates that the Board is thinking about letting a wide range of firms, including nonfinancials, hold reserve accounts."

Now I understand why Summers will become the Fed Chairman.

HepionAugust 29th, 2013 at 9:57 am

Ain't it the case that deposits are just mark-ups of banks debt to the deposit holder and thats why banks can create them, and banks use government issued currency to settle these debts, and our monetary system is actually a system of settling payments with 3rd party debt since we regular people use bank debt as our payment medium and banks use government debt as their payment medium?

I have to wonder why MMTers use mystical statements like "loans create deposits" that no-one knows what it means instead of explaining more clearly our monetary system.

L. Randall Wray L. Randall WrayAugust 29th, 2013 at 3:27 pm

Hep: It is not MMTers that came up with that. It is extremely well known, and very commonly used, by everyone who works in the endog money literature. “Loans create deposits and deposits create reserves” is short hand. It is NOT MMT. If you do not like it, complain to Marc Lavoie–who so far as I know popularized it.

PierreAugust 29th, 2013 at 11:24 pm

Sorry , i reformulate.
1.Main Street , views bank as a intermediation model:
A lend is savings to B, B pays 2% interest per year to A and 1%to the bank for the relation service.And the borrowed money ends on C account who will lend to etc…
2.If now we say Banks create money, that means that when Banks lend money they credit
"out of nothing" the account of borrower Y ,who will pay Z . Z keep the money on savings.
I suppose that Y pays 3% on interest to the bank and the bank give 2% to Z because Z savings deserve remuneration.
Deposit make credit or credit make deposit,with my two exemples ,we see that borrowers
Pay 2%interest to the lender or the saver (that the same at the end) and 1% to the bank as
Service for creation or intermadiation.
The problem with that concept "money creation out of nothing " is that push to think that bank get a "Droit de Seigneuriage" which means that banks gets the all interest from
The borrower . Which is not correct as in my Exemple n2
I say that because in France a lot of Leftistes people tell, ok gov is 100% in debt,
Let s create money with 0 interest to pay the bills.
But at the end the money ends on a saving account,and that saving account wants always interest!
So we can t have money without interest charge somewhere.

TomUsherAugust 30th, 2013 at 3:30 am

I submitted this comment earlier, but it never showed up. It then occurred to me that I had javascript off for this site. I turned it on and discovered the Intense Debate commenting system. Here's my comment again for the first time:

Hi Randy,

The Fed does still require 10% regular reserves against bank loans made/outstanding. What's the 90% if not a multiple?

Don't you think that the real reasons the Fed hasn't been able to cause inflation are 1) it's been paying interest on excess reserves and 2) the deregulators killed Glass-Steagall so that now the banks move their hot money where ever rather than concentrating on the traditional "conservative" lending they used to do?

If the Fed were to charge interest on excess reserves and if Glass-Steagall were reinstated, don't you think the banks would cause excess reserves to move into regular reserve accounts by way of lending newly created money that would then represent that 90% mentioned above?

I like reading your stuff and listening to your talks, but I've never really seen or heard you put this to rest.

I've read Steve Keen and others and like that too, but I'm never convinced by the arguments or points given that the multiplier doesn't exist. Until someone makes a really clear and convincing case that it doesn't exist, I will continue to maintain that it does for the reasons I mentioned above.

I'm open to proof it doesn't exist.

Perhaps we are speaking past each other and you mean in the practical sense only: in the sense that it is not now being utilized (much if at all).

Even knowing that, that's your position would be an improvement for me in trying to understand why you insist it doesn't exist.

At this point, I feel that the multiplier is still in use but is playing a very much smaller role than it probably ever has before by magnitudes of order.

Any light you shed on this whole subject will be appreciated.

Aside from this one aspect, I'm fully on board with MMT as the best explanation for how our mixed-economic system works.

Ideologically, I'm probably less patient than you are concerning what the government should do: full, permanent employment and high-skills training — just what the New Deal fell short on.



jsnSeptember 1st, 2013 at 2:50 pm

The extension of Fed access to shadow banks seems an awful lot like the Reich's bank honoring industrialist scrip in the run up to the Weimar inflation. I understand that the expropriation of capital plant was the proximate cause of these scrip issues and that reparations were behind that, so expropriation of plant and export of product generated capacity constraints in the real German economy, ergo inflation.

Should the Fed extend coverage to shadow banks I wouldn't anticipate immediate knock on effects in the real economy like Weimar, but instead an acceleration of the kleptocratic practices Bill Black so ably documents. My worry is that, having watched this process for twenty some odd years, the real economy is becoming so hollowed out as to represent a threat to real productive capacity.

I view the fiat dollar, as a system, as the score board of the productive capacity of that sector of the global economy denominated in dollars. At present the score is high indeed, but the kleptocratic efforts of the last several decades have off shored so much of real production and starved so much of real human investment and exchange at home, at what point does the tension between world power and domestic weakness blow the whole thing to bits? Not a hyperinflation, but a flash disintegration where all that real productive capacity abroad re-denominates into a different currency and an impoverished and ignorant polity here implodes or disintegrates?

L. Randall Wray L. Randall WraySeptember 3rd, 2013 at 2:26 pm

before the crisis it was more like 1% reserves against deposits. In any case what matters is causation, and you’ve got it backwards. We can certainly calculate an ex post ratio. So what? If the CB always accommodates, there is no constraint resulting from reserves. Which is the case.

Lake Erie LiberalSeptember 5th, 2013 at 8:41 pm

It is mere conjecture but I suspect that Paul Krugman might be a closeted MMTer but that he doesn't want to give up the argument that we MUST raise taxes on the rich in order to fund the type of society that we all want.

Maybe he thinks that is a more effective political argument than arguing as a point of morality or economic justice?

Or, maybe he's just as bad as the neoclassical economists he lampoons and is sincere in his dismissiveness about MMT?

olly100September 14th, 2013 at 4:10 pm

Hi Randall,

Lawrence Kotlikoff estimates future 'unfunded liabilities' of the US govt to be in the region of $202 trillion. Are his estimates nonsense, or is this figure accurate?


L. Randall Wray L. Randall WraySeptember 15th, 2013 at 9:24 am

olly: of course it is nonsense in numerous ways. Most obviously, federal govt commitments cannot be “unfunded” as it keystrokes to meet its commitments. It cannot run out. It has unlimited keystrokes so even if the commitments were unlimited, govt can meet them. Second place, what matters is future productive capacity, not keystrokes. and so on. Kotlikoff has no understanding of issues surrounding SocSec or govt finance; best advice: never listen to him.

TomUsherSeptember 15th, 2013 at 4:44 pm

Thank you, Randy, for your reply.

Electronic Code of Federal Regulations
TITLE 12–Banks and Banking
§ 204.4 Computation of required reserves.

(d) For institutions that file a report of deposits weekly, reserve requirements are computed on the basis of the institution's daily average balances of deposits and Eurocurrency liabilities during a 14-day computation period ending every second Monday.

(e) For institutions that file a report of deposits quarterly, reserve requirements are computed on the basis of the institution's daily average balances of deposits and Eurocurrency liabilities during the 7-day computation period that begins on the third Tuesday of March, June, September, and December.

[Reg. D, 74 FR 25637, May 29, 2009, as amended at 74 FR 52875, Oct. 15, 2009; 75 FR 65564, Oct. 26, 2010; 76 FR 68066, Nov. 3, 2011; 77 FR 21852, Apr. 12, 2012; 77 FR 65774, Oct. 31, 2012] <a href="

” target=”_blank”>
Obviously from that, express regulations allow lending before reserves are required. However, there are deadlines.

Essentially, more of the same:

§ 204.5 Maintenance of required reserves.

(b)(1) For institutions that file a report of deposits weekly, the balances maintained to satisfy reserve balance requirements shall be maintained during a 14-day maintenance period that begins on the third Thursday following the end of a given computation period.

(2) For institutions that file a report of deposits quarterly, the balances maintained to satisfy reserve balance requirements shall be maintained during an interval of either six or seven consecutive 14-day maintenance periods, depending on when the interval begins and ends. The interval will begin on the fourth Thursday following the end of each quarterly reporting period if that Thursday is the first day of a 14-day maintenance period. If the fourth Thursday following the end of a quarterly reporting period is not the first day of a 14-day maintenance period, then the interval will begin on the fifth Thursday following the end of the quarterly reporting period. The interval will end on the fourth Wednesday following the end of the subsequent quarterly reporting period if that Wednesday is the last day of a 14-day maintenance period. If the fourth Wednesday following the end of the subsequent quarterly reporting period is not the last day of a 14-day maintenance period, then the interval will conclude on the fifth Wednesday following the end of the subsequent quarterly reporting period. (cont.)

TomUsherSeptember 15th, 2013 at 4:45 pm


(c) Cash items forwarded to a Federal Reserve Bank for collection and credit are not included in an institution's balance maintained to satisfy its reserve balance requirement until the expiration of the time specified in the appropriate time schedule established under Regulation J, “Collection of Checks and Other Items by Federal Reserve Banks and Funds Transfers Through Fedwire” (12 CFR part 210). If a depository institution draws against items before that time, the charge will be made to its account if the balance is sufficient to pay it; any resulting deficiency in balances maintained to satisfy the institution's reserve balance requirement will be subject to the penalties provided by law and to the deficiency charges provided by this part. However, the Federal Reserve Bank may, at its discretion, refuse to permit the withdrawal or other use of credit given in an account for any time for which the Federal Reserve Bank has not received payment in actually and finally collected funds.

[Reg. D, 74 FR 25638, May 29, 2009, as amended at 77 FR 21853, Apr. 12, 2012];…
Then there's this:

§ 204.6 Charges for deficiencies.

(a) Federal Reserve Banks are authorized to assess charges for deficiencies at a rate of 1 percentage point per year above the primary credit rate, as provided in § 201.51(a) of this chapter, in effect for borrowings from the Federal Reserve Bank on the first day of the calendar month in which the deficiencies occurred. Charges shall be assessed on the basis of daily average deficiencies during each maintenance period.

(b) Reserve Banks may waive the charges for deficiencies based on an evaluation of the circumstances in each individual case.

[Reg. D, 74 FR 25639, May 29, 2009, as amended at 77 FR 21854, Apr. 12, 2012] <a href="

” target=”_blank”>
Are you claiming the Fed allowed waivers concerning up to some 90% of reserve requirements in the aggregate? I could ask many questions here but feel it better to simply ask you to clarify and elaborate.

For my part, let me say that I was aware of the compliance dates and that lending could, and does, occur before the lending banks have the reserves necessary to meet the requirements. I was also aware that the Fed can create, and has created, all the reserves it wants at its sole discretion. My point though is that much of the massive reserves created post-Great Recession onset have not been used by the banks against new loans, which would have moved excess reserves to regular reserves, but have rather been left in excess-reserve accounts earning the banks interest from the Fed. My understanding is that the Fed has entertained reducing the interest paid and even charging interest on excess reserves but in its mind, realizes the banks haven't had the borrowers who could meet the lending standards in order to put all those excess reserves to work.

So, where do we differ?