Great Leap Forward


I think the biggest policy failures in Washington over the past three or four decades are caused by confusion over the nature of money, and more specifically, the nature of a sovereign currency. So many of our policy makers simply do not understand that a sovereign issuer of the currency is not like a household user of the currency. For a sovereign issuer, there is never a solvency constraint. And a sovereign issuer spends by issuing its currency—not by borrowing it.

The whole Euroland mess could have been avoided if this had been understood—since countries would not have abandoned their sovereign currencies, or would have insisted on a different structure of the Euro make-up. Further we would have avoided all the stupid battles over US deficits and debt limits. And would have ramped up federal spending sufficiently to get out of the great recession by spring of 2009—rather than allowing it to drag on to the present.

And if they understood how a sovereign currency operates, there would not be all this Rogoff-and-Reinhart nonsense about excessive US government borrowing.

Indeed, there is no operational procedure that allows an issuer of IOUs to borrow his/her own IOUs. If you write an IOU to your neighbor (“IOU five bucks” or “IOU a cup of sugar”) you do not go borrow your IOU from that neighbor in order to spend. It would be a nonsensical operation. If you needed more bucks or sugar you could approach a different neighbor and write another IOU—but you would not go back to your original creditor to “borrow” the IOU in order to issue it to a different neighbor.

A sovereign government does not really borrow its own currency, nor does it really even spend its tax revenue (receipts of its own currency). The best way to think about sovereign spending is “keystrokes”: the sovereign government “keystrokes” its own IOUs into existence as it spends.

I actually thought Chairman Bernanke understood this. When quizzed about where the Fed got all the dollars it used in its ($29 trillion!) bailout of Wall Street, as well as its $2+trillion Quantitative Easing, Bernanke quite accurately and honestly said: keystrokes. More recently he put it this way:  (Look at his presentation beginning on page 17):

“Now, you might ask the question, well, the Fed is going out and buying 2 trillion dollars of securities – how did we pay for that? And the answer is that we paid for those securities by crediting the bank accounts of the people who sold them to us, and those accounts, at the banks, showed up as reserves that the banks would hold with the Fed. So the Fed is a bank for the banks. Banks can hold deposit accounts with the Fed, essentially, and those are called reserve accounts. And so as the purchases of securities occurred, the way we paid for them was basically by increasing the amount of reserves that banks had in their accounts with the Fed.

Yep, simple keystrokes. Something the Fed can never run out of. Whether the bailout of Wall Street or the QE was a good idea is another matter entirely.

However, more recently, the Chairman weighed in on Treasury deficits. Apparently he has completely forgotten anything he ever understood about sovereign currency:

“Having a large and increasing level of government debt relative to national income runs the risk of serious economic consequences. Over the longer term, the current trajectory of federal debt threatens to crowd out private capital formation and thus reduce productivity growth. To the extent that increasing debt is financed by borrowing from abroad, a growing share of our future income would be devoted to interest payments on foreign-held federal debt. High levels of debt also impair the ability of policymakers to respond effectively to future economic shocks and other adverse events. Even the prospect of unsustainable deficits has costs, including an increased possibility of a sudden fiscal crisis. As we have seen in a number of countries recently, interest rates can soar quickly if investors lose confidence in the ability of a government to manage its fiscal policy. Although historical experience and economic theory do not indicate the exact threshold at which the perceived risks associated with the U.S. public debt would increase markedly, we can be sure that, without corrective action, our fiscal trajectory will move the nation ever closer to that point.”

Uh…NO! Completely wrong.

With regard to the observation that a “number of countries” have seen interest rates soar “if investors lose confidence”, that is obviously a reference to Euroland. However, even Paul Krugman (who, as I’ve argued, doesn’t fully understand money) recognizes that Euroland’s problems are caused by abandonment of sovereign currencies in favor of the Euro. Each individual member nation effectively became a user of the Euro—much like a US household, firm, or state government. Ability to spend is thus subject to markets and the ECB’s tolerance—which in turn is determined by German voters’ willingness to bail-out periphery countries. Don’t bet on that! Germany will not be as charitable as were the other 49 states when New York got in trouble back in the early 1970s. After all, New Yorkers sort of speak the same language shared by most Americans. But the rest of America told New York to take a hike.

So what’s the problem with Bernanke’s claims?

As a fellow economist, I can say it is quite embarrassing to read the quote from Bernanke above. You can understand that our rightwing wingnuts and goldbugs make similar comments—they are not well-trained in monetary economics. (OK, yes, I know, I’m going to receive a lot of their comments—watch for them below.)

But Ben Bernanke happens to be the flipping Chair of the Federal Reserve! And he’s a well-respected academic economist. Yet his apparent (mis)understanding of monetary economics is stuck in the simplest 19th century expositions.

Doesn’t he realize that his organization—the Fed—sets the overnight interest rate in the nearly monthly FOMC meetings? What does he think they are doing when they meet for two long days?

The base interest rate is set by a vote of the FOMC. Period. It is not set by markets. It is not determined by the government’s “borrowing requirement”. Sovereign currency-issuing government budget deficits place no upward pressure on interest rates. Ever.

Indeed, he should understand that as government spends by crediting bank reserves, the pressure is DOWNWARD on the overnight rate—as banks offer excess reserves in the overnight market. That is relieved by the Fed—if it wants to—to hit its target.

The deficit CANNOT raise interest rates unless Bernanke & Co. decide to vote to raise rates. They can always “just say no”: no rate hikes in response to budget deficits.

Now, we do know that if budget deficits eventually spur the economy to recover, the FOMC will vote to raise interest rates. This is not due to market pressures that result from budget deficits. It is due to the double superstition held by the Fed that a) a growing economy tends to cause inflation and b) rate hikes reduce inflationary pressures. Now, I think both of these superstitions are false. But the bigger point is that rate hikes occur due to a vote of the FOMC—not to “market reactions” to deficits.

Further, if any “crowding out” occurs due to the rate hikes (that is, if investment falls) it is the fault of the FOMC that chooses to raise rates.  So, yes, it is likely that the Fed will raise rates eventually, and it is possible this could reduce some kinds of debt-related private spending (ie “crowding out”), but all that is the desired result of Fed policy.

Now, Bernanke or his supporters might respond that all this is true enough of the overnight interest rate—but the problem is in the longer maturities since the Treasury will sell long-maturity treasuries to borrow to finance its deficit. And that then pressures long term interest rates at which investment is financed.

Again, nonsense. Treasury does not sell bonds to borrow its own currency IOUs.

Look at this in two ways.

First, let us say that as the Treasury deficit spends it does issue long maturity bonds dollar-for-dollar, and that tends to flood the market with more of such securities than the market wants. Can the Fed stop long term rates from rising?

Can anyone say “QE”? Duh! The Fed has bought up $2 trillion of such assets precisely to push down long interest rates. Is there any limit to its ability to do so? No, for exactly the reasons Bernanke says. Is there currently any crowding out of private spending due to the government’s trillion dollar deficits? Of course not. The treasuries are flowing onto the Fed’s balance sheets.

Ok, yes, I know. Our wingnuts and goldbugs are whipping up inflation hysteria—all those Fed purchases will cause hyperinflation by pumping banks full of reserves. No. The treasuries are safely locked up on the asset side of the Fed, and the reserves are safely locked up on the liability side of the Fed. They cannot get out. No bank can lend reserves except to another bank.

You can think of all these reserves as imprisoned for life—they’ll never get out. Rather, eventually as the economy recovers the Fed will sell the treasuries back to bank and will debit their reserves by the amount of the sale. Presto—the reserves will be gone. Without ever causing any Weimar hyperinflation.

Second. The operational purpose of bond sales by Treasury or Fed is to offer a higher interest earning alternative to reserves. This is not a borrowing operation. It is part of monetary policy—draining excess reserves that would cause the overnight rate to fall below the FOMC’s target. Just enough bonds are sold to accomplish that.

There is no competition for a limited supply of “loanable funds”—rather, banks have reserves created through government spending and they can choose to hold them, lend them in the fed funds market (to another bank) , or buy treasuries. That is it. There are no other alternatives. It has nothing to do with lending to private firms or households. It has nothing to do with the lending rate to firms that want to invest. The reserves are created through government spending then drained through sales of treasuries. And that is done only to prevent the overnight rate from falling to zero.

The Treasury and Fed can always “just say no”: do not sell the treasuries. Let the interest rate fall to zero. There is never any imperative to create “crowding out” by pushing up market rates—by raising the fed funds target or by selling long maturity treasuries in excess of what the market wants.

Let me close with reference to a relatively good piece by Martin Wolf, summarizing a paper by Paul McCulley and Zoltan Pozsar (formerly of the NYFed) that does get money right. Here is the link to Wolf and a quote from their paper.

“Experience over the last four years (not to mention Japan’s experience over the past 20 years) has demonstrated that governments operating with a (floating) currency do not suffer a constraint on their borrowing. The reason is that the private sector does not wish to borrow, but wants to cut its debt, instead. There is no crowding out. Moreover, adjustment falls on the currency, not on the long-term rate of interest. In the case of the US, foreigners also want to lend, partly in support of their mercantilist economic policies.”

(Note Wolf says he’ll ignore MMT’s approach as it would take a whole lengthy analysis of its own. Too bad. Hopefully he’ll come back to that.) Let me just say that the McCulley-Pozsar analysis quoted above has got it right except that they seem to think this is “special case” analysis applying to a downturn where the private sector wants to cut its debt. But in truth, it applies in all situations in which a government issues the sovereign currency.

Wolf promises a Part 2. If his follow-up makes points relevant to my argument, I’ll also offer a Part 2.


NeilWApril 19th, 2012 at 2:52 pm

It means like the stock market the value of the currency wrt. other currencies will fluctuate.

If you can work out which way it will fluctuate then you know more than anybody else and should be writing articles from your retirement pad in the Virgin Islands having made a pile on the forex market.

NeilWApril 19th, 2012 at 2:56 pm

"Again, nonsense. Treasury does not sell bonds to borrow its own currency IOUs."

The issue is the desire of the non-government sector to save more IOUs than it wants to invest.

So it is not the Treasury that is *pulling* IOUs from reticent holders who have to be encouraged to let them go. Instead they are being *pushed* onto the government by eager savers in search of a better interest rate.

And that difference in viewpoint changes everything.

brolltheamericanApril 19th, 2012 at 3:09 pm

I really like the IOU to borrow sugar analogy. That really helps put the nature of money and the national "debt" into clear focus. The "keystrokes" terminology, while accurate, still makes non-MMTers bristle. Leads to "out of thin-air / hyperinflation" hysteria.

puzzledApril 19th, 2012 at 3:14 pm

Hmm, the implication is that it will adjust in response to "keystrokes" – are you saying that it's just as likely to appreciate as depreciate in response to "keystrokes".

Joseph M. FirestoneApril 19th, 2012 at 3:41 pm

Nevertheless, people have to accept the "keystroke." First, it's what happens. So, it's the truth. Second, it is "out of thin air," or at least out of the constitutional authority of the Government to issue money, a power of the Government established by the founders. The constitution makes the Government the only true monetary sovereign in the US. And Third, talking about "keystrokes" as the truth, then opens the way to the discussion we really need to have, which is the discussion about the causes and cures for inflation and hyperinflation, and the reasons why the Zimbabwe and Weimar cases don't apply to our situation.

Btw, Randy, thanks for a great post. Hope it appears at NEP too!

areopageticaApril 19th, 2012 at 3:41 pm

Mr. Wray-

Aren't you just making the point that a central bank can monetize its nation's deficit? Doesn't Bernanke know that, and the pros and cons about it? It seems to me that was exactly the point of his famous 2002 "helicopter" speech.

John CardilloApril 19th, 2012 at 4:02 pm

Another excellent article. I love the IOU analogy. As long as IOUs/currency is used to facilitate the real economy, there should be no limit to its creation. Consumption and production should take off and run at full speed. People should be able to consume as much as they are willing to to work for. What concerns me is that, the creation of IOUs/curreny has been hijacked by the investment community to inflate asset prices. This side of the equation contributes nothing to the real economy and is responsible for every debt fueled catestrophic investment bubble in past century. If policy can differentiate between these two drivers of money creation and discourage the latter, all would be well.

LR WrayApril 19th, 2012 at 4:02 pm

Aero: NOT a matter of "CAN" but rather that sovereign government ONLY spends by crediting bank accounts. And if that is true then his position as shown in the quote is completely wrong. This is not a matter of choice.

areopageticaApril 19th, 2012 at 5:01 pm

You lost me there. If the Fed stops buying, there is no sovereign crediting bank accounts. I do agree that the Fed and other central banks have much more leeway to finance deficits than is commonly understood, but they have to buy to make your arguments apply.

Adam1April 19th, 2012 at 6:11 pm


“Aren’t you just making the point that a central bank can monetize its nation’s deficit?”

It’s not that the FED can monetize the debt, but that it always does. The Treasury can not sell bonds/deficit spend without the FED monetizing the debt. It’s just not operationally possibly.

The FED can not have and defend a target interest rate without monetizing the deficit as it is created. For the FED to NOT monetize the debt it would have to abandon its target rate and put the US payment system at risk of collapse – which would be a direct violation of its charter.

paulie46April 19th, 2012 at 6:49 pm

"The "keystrokes" terminology, while accurate, still makes non-MMTers bristle. Leads to "out of thin-air / hyperinflation" hysteria."

Yeah, the "earth isn't flat" terminolgy had a similar effect on their non-MMT ancestors.

AetherApril 19th, 2012 at 7:56 pm

rather, banks have reserves created through government spending and they can choose to hold them, lend them in the fed funds market (to another bank) , or buy treasuries. That is it. There are no other alternatives.

Now that the Fed has MBS on its balance sheet can't banks uses reserves to purchase mortgage securities. At some point one can imagine the Fed returning to an all treasury backed balance sheet – this is required by law, at least by my interpretation of the Fed's charter.

Mary CattermoleApril 19th, 2012 at 8:16 pm

I have read this 5 times and I still don't get it. Maybe that's why no one (or very few) do get it. I guess you are saying that the Fed does not increase the supply of money, it only changes the cost to borrow it which it can do at any time. Then what was the point of QE? Why did it drive up stock and commodity prices? What does cause inflation?

Manuel Costa ReisApril 19th, 2012 at 8:17 pm

This assumes the FED wants to bailout the government.

And also, it assumes that the FED wants to say to the government that it can and will. If it ever said so the government would assume a free lunch and start spending like crazy which would in fact require the bailout.

Why would this be bad? Firstly because interest rates in the U.S. would climb, not because of unpayability of the debt but due to exchange rate risk and inflation reducing the real interest rate (reducing foreign investment). Secondly it would bring very high inflation to the economy which above a certain level is undesirable.

So the FED cannot say it will bailout the government. And should not.

llisa2u2April 19th, 2012 at 9:40 pm

He does. And thank you very much for your article. And in response to the question: What does he think they are doing when they meet for two long days? It's elementary "they" were running up a big tab. Does it even matter what they were doing? A good time was had by all, and they don't have to pay for, or assume any responsibility for any thing. Pass the buck and let somebody else worry about uh, what?

B110April 20th, 2012 at 6:37 am

It is hard to face up to the fact that one has been fed nonsense by teachers, journalists and officials who should have known better.

It is always more convenient living in a bubble (in more than one sense)!

areopageticaApril 20th, 2012 at 5:56 pm

Adam- It is entirely obvious that the Fed does NOT always monetize debt. They can choose to proceed or not with QE3, or raise their target rate or even sell the securities they already hold, which would do opposite of monetization, none of which would violate its charter. What do you think the entire debate about QE has been about? Absent monetization, treasury sells debt to the private sector and it does not simply credit bank accounts the way Fed does when it monetizes.

LRWrayApril 21st, 2012 at 12:50 am

Look. Here is what people do not understand. Reserves can be used for exactly 4 things.
1. Clearing with other banks or the Fed/Treasury
2. Cash withdrawals
3. Buying treasuries
4. lending to other banks in the fed funds mkt.

That's all she wrote, folks. Forget all other stories. They are wrong. Period. Banks cannot lend reserves to nonbanks. Reserves cannot be used to buy securities of any type except treasuries from the Treasury. It is that simple.

Michael E PicrayApril 21st, 2012 at 4:50 pm

Gosh. Long article that produces lottsa potential comments. ;-)

Like this one:
You said: "For a sovereign issuer, there is never a solvency constraint."

Completely wrong. For a recent example, try to tell that to Mugabe in Zimbabwe. I believe they said pretty much the same thing – that there were no constraints on how much "money" (actually currency, since fiat currency is NOT "money") they could print up. While technically true as far as the statement goes, it leaves out a significant part of the issue – which is although the "sovereign" can in deed print up as much as they want without constraint, that does NOT mean that anyone will consent to exchange it for anything of value, because such paper/currency HAS no intrinsic value. As a result of Zimbabwe testing the limits of the theory of Monetary Sovereignty, the nation now uses the US Dollar as its official currency. Eventually, no one would accept their currency. (Not even the European company who sold them the paper to print their currency on.)

As to your sugar analogy you left out a bit, but a rather important bit. The FED can indeed "borrow" a cup of sugar… and the rest of it. But you forgot to mention that it is the FED that determines the DEFINITION of how much sugar is in a "cup" – ie the FED determines the measure of the underlying value.

You also said the FED sets interest rates… not quite correct. Although the FED can set the FFR (Federal Funds Rate), the FED does NOT set the Prime interest rate – which is the interest rate the rest of us have to pay for things like mortgages & etc. Before the crash, prime consistently ran 3% above the FFR – but since a bank can't even pay expenses on a loan that only returns 3%, Prime now is greater than 3% over the FFR. Granted, not by much, but still not a fixed differential, so it cannot be said that the FED (by extension) sets Prime.…

So the FED does NOT "set interest rates."

But I have to thank you for confirming something that I could only guess at a year ago… (linked below) and I included a return link to your article here.…

Semper Gumby!

RonTApril 22nd, 2012 at 1:12 am


"You said: "For a sovereign issuer, there is never a solvency constraint."

Completely wrong."

No, completely right. Mugabe didn't run out of money, despite the spending going into gazzilions. Inflation is the constraint, once you reach full employment you should (but don't have to) stop. But the sovereign government cannot run out of its own IOUs, that is what it means.

Michael E PicrayApril 22nd, 2012 at 2:23 am

There was no "currency" constraint, but Zimbabwe was bankrupt, the money worthless. I'd say there was a very big SOLVENCY constraint!

"Solvency, in finance or business, is the degree to which the current assets of an individual or entity exceed the current liabilities of that individual or entity."
(Much as I hate to use wikis as reference sources, the wiki had the best general definition.)

There is NO WAY on God's Green Earth that Zimbabwe was even on the same planet with solvency!!!

I responded to what was said.

RonTApril 22nd, 2012 at 2:31 am


No, you didn't respond to what was said. What was said was, as you yourself quoted, "For a sovereign issuer, there is never a solvency constraint." That is true. There is inflation constraint, *not* solvency constraint. Zimbabwe still could honor all its Z-dollar debts, it didn't run out of Z-dollars despite having spent trillions, so it wasn't insolvent, it run into the inflation constraint which MMT recognizes.

AetherApril 23rd, 2012 at 10:50 pm

Prof. Wray,

You might be correct and I don't mean to nitpick. It just seems to me that if the Fed decides to sell its MBS assets, it will be draining reserves from the system. In the process, banks will be exchanging reserves for MBS, hence buying MBS with reserves.

Further, by my interpretation, the Fed was/is only allowed to purchase securities that are fully backed by the U.S. Gov't. If that interpretation is correct, the Fed's purchase of MBS was illegal. FNE & FNM securities were not and arguably still are not fully guaranteed by the U.S. Gov't.


Sean FernyhoughApril 25th, 2012 at 1:37 pm

The other "special case" aspect of the Martin Wolf article is the last sentence. Isn't there an implication that the US postion cannot necessarily be extrapoltaed even to other sovereign issuers of their own currencies.

Most Read | Featured | Popular

Blogger Spotlight

Håvard Halland Håvard Halland

PHåvard Halland is a natural resource economist at the World Bank, where he leads research and policy agendas in the fields of resource-backed infrastructure finance, sovereign wealth fund policy, extractive industries revenue management, and public financial management for the extractive industries sector. Prior to joining the World Bank, he was a delegate and program manager for the International Committee of the Red Cross (ICRC) in the Democratic Republic of the Congo and Colombia. He earned a PhD in economics from the University of Cambridge.