Hyperinflation – Some Lessons From a Scenario Analysis
I’ve been reading a lot of Leonardo Da Vinci lately and what I love about his approach to solving puzzles was his desire to understand what made things work. Da Vinci wasn’t always so interested in solving the world’s problems. Rather, he was interested in answering the questions about the way things worked. I’ve adopted some of this in my approach to understanding the monetary system and I think it’s useful here in expanding on a concept that is very important.
In the “Ask Cullen” section, Chuck asked a question about debt monetization and why it’s different when the Fed buys debt on the secondary market as opposed to the Fed buying directly from the US Treasury. This can be a bit confusing so it’s helpful to perform a brief scenario analysis to highlight some key points. Often, the most extreme environments teach us the most about the way a system works. In this case I will use a hyperinflation. I think this is crucial in understanding why we haven’t been in a hyperinflationary environment and more generally regarding a basic understanding of money demand and operational realities of the monetary system.
The US government finances itself in one of two ways – taxing or selling bonds. It sells bonds when it cannot procure enough funds from tax payments. This is what we refer to as the “budget deficit”. US government bond auctions are well orchestrated events that are designed not to fail. The US government essentially harnesses banks as funding conduits by requiring them to bid at government bond auctions in exchange for being a Primary Dealer. The Fed works in a symbiotic fashion with the Treasury to ensure that banks can settle the auctions while essentially serving as an implicit backstop. That is, even in a worst case scenario the banks can always sell to the Fed or the Fed can buy directly (although this is illegal now). In this regard, the US government can’t “run out of money” or face a solvency constraint like Greece (the US government always faces an inflation constraint), who operates in a monetary system with what is essentially a foreign central bank.
So what would happen in a hyperinflation? First, we must understand that money is ultimately only stable because its demand is based not only on the government’s ability to procure funds ( legal enforcement, taxation, selling bonds, etc) but also because the public is willing to utilize that money (MR calls this quantity value and acceptance value – see here for more). In the case of a hyperinflation what happens is that the demand for money becomes very tight. But its utility collapses. So people stop paying taxes as profits collapse and the utility of money declines. So people actually end up hoarding more money. When the government has to cover its tax receipt short-fall they sell more bonds. But the Primary Dealers can’t be expected to take on all these bonds in a hyperinflation, can they? Why would they collect endless amounts of bonds whose prices are collapsing? They most likely won’t because they’ll weigh the benefit of being a Primary Dealer versus the benefit of staying in business. So the Fed has to step in and buy the bonds. The Fed is like any bank. It can credit accounts at will. So they gobble up all the bonds directly from Treasury. But who cares? The money is already hyperinflated and worthless. That’s REAL monetization. It means the utility of money has collapsed and as a result the demand for bonds by the currency users has cratered.
The difference between this alternate environment and the one we currently face is that demand for government debt is extremely high. Treasury prices are at record highs, yields have collapsed, bond markets remain deep and highly liquid, bond auctions are seeing very strong demand, etc. In other words, there’s no need for the Fed to be buying bonds to fund the Treasury’s account. This form of QE is just good old fashioned monetary operations in an attempt to alter interest rates and other facets of the economy (it obviously hasn’t worked out all that well).
To me, this is a crucial scenario analysis to understand because it highlights the difference between a hyperinflationary collapse and the QE we’ve been seeing, which to me, is just boring old monetary ops and nothing remotely similar to real monetization because the private demand for debt (and US dollars more generally) is still very strong.
Some of the key insights from a little scenario analysis like this:
- Understanding quantity value and acceptance value matters enormously to your understanding of the monetary system.
- QE, as we’ve seen it, is not “debt monetization” since that implies a lack of demand for government debt (or an inability for the government to procure funds from anyone but its own central bank).
- Getting the precise operational realities of the monetary system will help you avoid colossal investment mistakes.
Recommended reading in addition to this:
Understanding the Modern Monetary System
Understanding Quantitative Easing
Cullen Roche
Mr. Roche is the Founder of Orcam Financial Group, LLC. Orcam is a financial services firm offering research, private advisory, institutional consulting and educational services.
Follow Me:
35 Comments
It’s in fact extremely simple. If the FED buys debt from the banks then the banks can use the money to make new loans to creditors that want to speculate on rising asset prices. But then – at least – one asset classes need to go up in price.
When the FED buys the debt directly from the Treasury, then it won’t be used for speculation but it will be used for some (useful or wasteful)government expenditure.
Agree. Although I would put it a bit differently. Banks create loans/credit “out of thin air” and then borrow the money to create deposits. (Steve Keen). And that’s where the checking account at the FED can be used for.
But in the currenct system the FED creates checking accounts for the banks. Not for the gov’t. If they would create account for the gov’t then the banks would be cut out of the loop.
‘Demand is extremely high,’ you say.
Could it be that demand is extremely high because the primary dealers are more or less obligated to buy the bonds, as you say?
In your example, the primary dealers only stop buying the bonds when we have hyperinflation, which effectively rules out anticipating that we will have problems.
Another point with this ‘high demand.’ Could it be that the primary dealers are themselves buying bonds with borrowed money? Let’s say, I’m a primary dealer and you come to be selling $1 trillion in bonds. I don’t have $1 trillion to buy them, but if I put them on my balance sheet I can use the value of the bonds to cover what I owe (that is, as long as I know the Fed will pay full price for them.)
It seems to be an entirely artificial market in which the participants have made a deal to believe that everything is just fine.
There are (maybe) plenty of buyers for T-Bills at the moment because people (and institutions including other governments) don’t like the alternatives. (The percentage of T-bills bought by the Fed is pretty high right now as I understand it.)
What happens to all those T-Bills when the economy recovers and this is no longer true?
Seems like eventually they will be converted to cash, one way or another.
Or do you assume we will never return to normal?
When the economy recovers the equity base upon which all these treasuries rest will soar in value. Bonds are a hedge or a facilitator of growth in real wealth. It is not just the liability side we care about. It is ultimately the bottom line. If equities surge who cares about what happens with bonds? The country will be much much wealthier as a result of the increase in equities (not that some bond holders won’t be hurt, but it is what it is).
Not quite right. Personally, if I am a bondholder, I absolutely do not care about “country being much wealthier”. Then, please check current UST (and, for that matter, corporate too) holders base in the US. What political power do they yield. Can you imagine what turmoil can happen if their interest are damaged?
Also, surge in equities would not compensate for loss in bonds – check the most recent Mauldin piece about P/E contraction during inflation.
Very good explanation of the current mechanism, thank you. The question now is what might be the catalysts to push the current quite stable system into hyperinflation mode.
I am still struggling with that. Can only imagine M1/M2 levels reaching certain threshold as a % to total USD denominated debt, and then money velocity suddenly accelerating, and money sterilization not working….
Your hyperinflation scenario sounds a lot like certain markets which collapse under stress: asset-backed & auction-rate securities. The liquidity is an illusion, until it isn’t. (Da Vinci? I hope this isn’t a jump the shark moment ….;-> )
Small financial websites don’t jump the shark because the sharks don’t care about them!
Sorry, I forgot to make the added point that esp. with the auction-rate securities (and asset-backed CP depending on the country), the dealers are *supposed* to supply liquidity, analogous to the relationship between the PDs and the Fed.
PS mean to reply to my own post …
This is a really good post. I love the idea of acceptance value and demand value. Really nice additions to the MR thinking.
Up until now all other article posts on pragcap reiterated how the US Gov’t doesn’t really need to finance itself because it is the monopoly supplier of the currency and can print money at will, the bond market was a rate-setting mechanism, and taxes were a drain to the “tub” which helps curb inflation, etc. This article states “The US government finances itself in one of two ways – taxing or selling bonds” – Holy 180 degrees turn (so it seems) Batman! This is completely the opposite of what’s been preached… I’m now confused. What did I miss while on holidays? Hopefully this is a symantics thing where I’m just missing the connecting link.
Hi Sixx,
I think you’ve missed the evolution of Monetary Realism. What I’ve done in collaboration with several other people is break-down the institutional design of the current system for what it is and describe the operational realities as they currently exist. For example, the US Tsy must procure funds into its account at the Fed before it can spend. I had fallen into a trap of using many of MMT’s loose analogies (like the scorekeeper analogy) that led to a false understanding of the precise way things work. We’ve done a deep dive into the details and corrected these errors. The result is a cleaner and more accurate description of the operational realities.
The big ideas are all still there. Like, the US govt isn’t “running out of money”, deficits are not inherently negative, sectoral balances are crucial to understanding, banks don’t lend reserves, etc. But we’ve gone in and tightened up the institutional description to better describe the way the system, as it is currently designed, actually works.
I’ll be posting my revised primer in a series over the next few days. Stay tuned. Or if you want to read more please see here:
http://pragcap.com/hyperinflation-some-lessons-from-a-scenario-analysis
BTW, somewhat geeky point. I read the links and the original article on MMR.com way back, but can’t really work out how one can prove that funds are procured before spent. Since the Treasury account sees flows (in and out), you could never prove the operation sequence unless the account gets close to zero (i.e. close to the check bouncing). ?
Of course you can. Let the Tsy account run down to zero and see what happens. They either need to sell more bonds or tax more. It’s about getting the accounting right….And the accounting says that you need a credit before you can have a debit. And the Tsy’s account needs to be credited before it can be debited.
Good that you have finally come to see this technicality.
This somewhat contradicts another tenet of MMT – that money had to be deficit spent into the economy before it can be taxed.
I have always made the points that
1) This point is useless debating (like chicken and egg; actually govt has taken over the monopoly from the priave sector)
2) This point is useless in knowing – it helps nothing by knowing which one was first
Best,
One part of JKH’s MR.com piece says the Fed can (essentially) provide a bridge loan (in lieu of an overdraft of the Tsy account, which is not permitted), so I read that to mean selling bonds first is not required. I think what you are saying, is that somewhere in the operations of the machine, one can prove that the Fed never does this so bond sales must precede spending. ?
The beauty of JKH’s work is that he shows the institutional design for what it is. No progressive agenda, just what it is. He offers some design options, but the key piece is seeing the current institutional design for what it is. We still glean many of the same conclusions as before using the MMT framework, but we get the accounting right and the institutional description right. That’s the beauty here. To understand why the Tsy is an operational currency user, but the govt as a whole is a currency issuer, you just need to understand the institutional arrangements and how they actually work. There’s no need to worry about “spend first” or other ideas that rationalize why the USA can’t run out of money. You just need to know how the relationships work and you come to the exact same conclusion.
No, in hyperinflation there’s actually too much money. That devalues the value (hence the word “devalues”) of the money (e.g. USD) against e.g. commodities which can’t be created “out of thin air”. The tax system collapses because people still pay their taxes but the government can’t do too much with the money because the value (purchasing power) has dramatically declined for the government too.
Bankers lend the money to speculators who speculate on rising commodity prices instead lending to a government and that sends interest rates through the roof and that leads to the collapse of the credit markets/bond market(s). And at that point the government is actually forced to literally print money, bank notes because it can’t issue credit (=T-bonds) anymore.
In that regard both deflation and Hyper-inflation have the same impact. Destruction of credit/debt. Currently there’s still one market prone to inflation and that’s the T-bond market. (=falling interest rates)
Too much money in the current situation would be that the FED expands its balance sheet to say $ 10, 20, 40 or 100 trillion. And that’s why it’s more dangerous to be currency issuer than a currency user.
I see a few issues with this post. Some is probably my misunderstanding, and some probably my different views.
1) You didn’t seem to answer Chuck’s question regarding what’s the difference between the FED buying debt on the secondary market as opposed to directy from the US Treasury.
2) You seem to be saying that we are not experiencing hyperinflation right now, which is true, but is also obvious, so not particularly useful information.
3) In hyperinflation gov currency is not held or hoarded. It is spent as quickly as possible as people have no confidence in it holding it’s value. Nobody holds money any longer than absolutely necessary, the worse the the hyperinflation becomes the more this is the case, and becomes a feedback loop.
4) The FED last year bough about 2/3 of US Debt, so it’s seem misleading at best to say there is high demand when most of it is from the government itself (I consider the FED to be a de facto government agency).
The Treasury selling new TBills on the Primary market is a holdover from the Gold era. Congress hasn’t changed it because they want newly issued TBills to go through the market; this in order to act as a supposed government fiscal restraint.
Looking at it in terms of circuitist theory, newly issued T bills could attract existing money in the economy (which would be a drain of the money supply). Or, newly issued T-Bills could attract new money, for example when the Fed Creates money on their keyboard.
If the FED creates new money and buys on the secondary market, that new money enters the supply and removes “old” bonds. The new cash has now entered the market and is available to buy “new” TBills, effectively funding deficit spending. At the FED level, an old bond enters their double entry ledger; so their books are balanced. But, that doesn’t change the ultimate fact that the FED created new money from nothing. That new money will find its counterpart “eventually”, as it flows through the market and becomes a counter to the new TBill (issued on the primary market). This is how deficit spending is funded.
If the new TBill has no counter in new money creation, then it is a drain on the supply. For example, when China recycles dollars to buy new TBills, they are taking their dollars off of the market, propping up dollar value (reduced supply), and helping maintain the Yuan peg. In the China scenario, recycling dollars makes demand for TBills and not real economy American products/services. (Economists should rethink the notion that money is demand.)
In our present day scenario, we deficit spend to gradually undo the balance sheet problems at private banks (more oligarchy). We deficit spend to fund overseas military bases (too much diverts real economy resources), and we deficit spend to fund unemployment (who become more productive how?). IMO none of the spending paths are in best interest of America, as deficit funds do not flow properly through the four modes of production: Labor/land/capital/public commons.
With regards to hyperinflation; the implication is that it is Government printing which is the problem. In the case of Weimar hyperinflation, it was a confluence of events. Mostly, new credit was created at private banks in order to fund bear raids (shorting of the Mark by creating more). Eventually it was the German government, via bank nationalization and issuing Federer money, that solved the problem. The Mark was under downward pressure to buy hard currency (dollars and Pounds and Francs) to pay inter ally debts left over from WW1 and per the Versaille treaty. Currency speculators attacked smelling weakness. We American’s were complicit in that hyperinflation as we demanded inter ally debts be paid; then France and England turned around and squeezed Germany.
In some cases, Governments do run Amok and create too much money, causing money to loose its “value.” The country formerly known as Rhodesia is an example. But, is the government doing the bidding of the people in that scenario?
Cullen,
You said: It sells bonds when it cannot procure enough funds from tax payments. This is what we refer to as the “budget deficit”.
I thought the budget deficit is, in general speak, how new net financial assets the U.S. government spends/creates into existence beyond how much it has taxed out of existence.
Is that description, in general, accurate? And so, can you explain how selling bonds spends/creates new net financial assets into existence?
Is bank money, via “loans create deposits” created to purchase the bonds, which then the Treasury ‘uses’ to spend/create new net financial assets into existence?
I’m very confused about this.
When the Tsy spends the proceeds from bond payments they not only credit accounts, but also credited accounts with new bonds. So the PD’s buy the bonds at auction, obtain bonds in exchange for cash and someone else gets the cash when Tsy spends it back into the system. Abracadabra, net new financial assets.
Would it be accurate then to say that the bond (not the spending of the money from the PD who bought the bond) is the new net financial asset spent/created into existence?
p.s.i liked the use of abracadabra
… can’t think of a time of seen it in writing
And also… if the bond is the new net financial asset…
Is the way that asset gets liquidated.. open market operations?
Come to think of it…
It seems that the first swap, where a PD exchanges cash for a bond, this action does not change the amount of net financial assets in the private sector. But then when the Tsy spends that money… it is that act of spending that adds (creates) new net financial assets.
That has got to be it.
You can also actually come to think of it as the bank creating the cash “out of thin air” by buying the bond that creates the net NFAs that are later let into circulation by govt fiscal spending.
Investor,
That’s a interesting point. Can someone take out a loan (loan creating the deposit) and then use that money to purchase a government bond? Is that allowed? And, would that ever be profitable? If it is profitable, it seems like pure arbitrage… no?
Banks can give themselves a loan as well (as far as I know), creating a deposit, buying the govt bond.
You state in your brief summary that “…even in a worse case scenario the banks can always sell to the Fed or the Fed can buy directly…”.
Buy directly from whom?
I have another question. What’s the difference (pragmatically speaking) between the Fed buying up government bonds in an inflationary scenario and the Fed expanding its balance sheet? Isn’t the result of both an expanded balance sheet?
Depends. For instance, in the current environment demand for govt debt is strong so I wouldn’t call this debt monetization. If, however, we were in a hyperinflation and demand for debt collapsed the Fed would indeed need to step in and purchase the bonds. Big difference in my opinion.