I just read Life in the Middle Ages, a reflective set of memoir-essays by James Atlas, who we discussed a few years ago in the context of his book about biography-writing, The Shadow in the Garden. Life in the Middle Ages came out twenty years ago, when Atlas was in his mid-fifties. The book is suffused with gentle regrets about his life and an awareness of how little time was left to him—which all makes me kinda sad since I’m almost 60!—actually will be 60 once this post appears—and, yeah, I think all the time about the dwindling number of years ahead of us. Atlas himself lived only to 70; to his credit, he completed his excellent Shadow in the Garden book in his late sixties. So he lived a full life, professionally speaking, even though in Life in the Middle Ages he expresses many regrets about his career setbacks.
I think the real problem with Atlas’s career has nothing to do with him; he just happened to enter a field—general-interest writing about literature—that was declining. Less people interested in literature, the gradual collapse of the economic model for newspapers and magazines, ease of mechanical reproduction so less need for live bodies doing the writing . . . put that all together, and Atlas in his career basically joined a decades-long game of musical chairs. When the chairs keep being removed, it’s natural to blame yourself, but it’s more just that he was caught in the wrong game. Which makes me sad because I love reading about literature. I think it would be cool to have been James Atlas, although I guess not in the later years when his audience declined, along with the rest of the audience for the sort of thing he was writing.
The other funny thing about Atlas’s book is that he talks about being broke—not poor, as he and his family seem to have all the possessions they might want, except for a fancy car (I don’t get why he wants a BMW or Jaguar, I guess it’s some sort of boomer thing?), but broke enough that they never have quite enough money to pay the bills, they’re always on the edge with credit card debt, he calls himself “lower upper middle class”—but then he keeps talking about the summer home they own in Vermont.
I can understand people being rich enough to have two homes, and I can understand people being broke enough to struggle to pay their bills every month—but it’s hard for me to picture both of these at once!
But then I had two thoughts which made it all clear to me:
1. The United States of America is kinda like James Atlas: We’re the richest country in the history of forever, we can have pretty much whatever we want (except that not everybody gets a BMW or Jaguar), and our national debt keeps going up, up, up. So, yeah, it’s possible to live the good life and have that second home, even though you can’t really pay your bills.
2. I’ve always been fortunate enough to have enough money to pay for whatever I want—I mean, not always, there’s a reason I ask for research grants to pay salaries for postdocs etc., but at a personal level, I can afford unlimited celery and Jamaican beef patties, pay to get my flat tires changed as needed, fly to faraway places, etc.—but I’m in an Atlasian mixture of debt and riches when it comes to time:
I have tons of free time, as is evidenced by (a) that I’m spending a half hour writing this blog post and (b) that I spent a couple hours earlier this week reading Atlas’s book, for no other reason than I felt like it. And this wasn’t even the only pleasure book I read over the Thanksgiving weekend. But I’m also in a continual time debt, a veritable treadmill of time commitments. I’m in the middle of writing 5 books and a few dozen research articles, and I keep taking on new projects. No way I can do all of these! But, as with Atlas and his finances, somehow I keep going.
So, from that point of view, my comfortable finances are an anomaly. Debt financing is the usual way of the world.
“Less people interested in literature”
Andrew:
“Less” wants to be “Fewer”.
The fewer of these linguistic indiscretions that appear in your posts, the less often I’ll have to comment on them. Unless, of course, you could care less, in which case my corrections will be fewer in the future. You can count on that.
John
Ditto regarding “number” vs. “amount.” Such distinctions were important in my school days partly because hardly anything had been invented or discovered until later.
The less/fewer thing is totally made up, and can even be traced to a specific guy (Robert Baker). There is a Wikipedia article with the details.
Well if we are being pedantic then clearly the phrase should be “couldn’t care less”. “could care less” is just incoherent. Here’s an angry Englishman explaining it, appropriately for this blog, with a graph: https://www.youtube.com/watch?v=om7O0MFkmpw
Lukas,
“I could care less” is like “T’inquiète”: it’s an expression that’s missing a word or two by convention, but the meaning is clear.
I assumed it was intentional, to stress how little Andrew would care if he didn’t.
(I agree that even if the meaning is clear for all intensive purposes it’s wrong – as a matter of principal. It’s not rocket surgery but I guess it’s a mute point.)
It’s just like overbooking of airlines tickets to maximize efficiency (seat utilization). We fill our refrigerators with more food and if we were to live the fullest life we will have a few spoiled avocados. Time, money — all are at our service (for happiness) — to be maximized by overbooking 😎
> and our national debt keeps going up, up, up.
It’s really critically important that we push back against the idea that national debt is similar to household debt.
The government isn’t borrowing money from banks or whatever to afford some profligate lifestyle, it CREATES money. Every time the US govt spends $1000 the money supply goes up by $1000
Now, if you just only create something, then the quantity of that thing will increase without bound. So, the government has a means to destroy money as well. That’s called taxes.
Furthermore, the government has a means of moving money creation around in time. If the government wants to create $1100 a year from now, it can spend $1000 now to buy bureaucrat salary and sell a zero coupon bond for $1000 at 10% to a financier. In effect this is a promise to tax the bureaucrat and give even more to the plutocrat at a later date to make sure the plutocrats don’t rebel against the government empowering the poors.
The purpose of bonds isn’t “to get the money for the govt to afford to do things” and that’s not the purpose of taxes either. The purpose of bonds is to shift the effect of creating money in time and space (time because the principal and interest is paid later, space because rather than paying $1000 to a supplier today, they’re instead paying $1100 to a banker next year).
Now taxes… That straight up destroys money. So the key to keeping the oligarchy happy is to tax the poor never the oligarchs, and bribe plutocrats with interest and QE and such. The purpose of bonds is so the poor don’t get ahead by selling things to the government by effectively committing to transfer money from the poors to the rich at a later date.
I prefer the simpler approach. Government debt and household debt are fundamentally different because the latter must be repaid but not the former. The government, unlike you and me, can keep borrowing indefinitely as long as their are willing lenders. And there are, provided that the government is not viewed as a credit risk. The only real consequence of the government debt is that it must be serviced – and at high interest rates that can be substantial. But even then, most of that debt is held by US people, so we are paying ourselves to sustain the debt (of course, the distribution of who pays and who receives is important as always).
“The government, unlike you and me, can keep borrowing indefinitely as long as their are willing lenders. ” Somehow, Trump seems to have managed it, even before he became the government (https://www.washingtonpost.com/politics/2016/live-updates/general-election/real-time-fact-checking-and-analysis-of-the-first-presidential-debate/fact-check-has-trump-declared-bankruptcy-four-or-six-times/).
Dale, the problem with the simpler approach is it’s wrong. The government simply does NOT borrow. This isn’t like back in the day when to buy things you had to have coins, so the government might write an IOU to a Lord who happened to have a large stash of coins and then it has coins to give to the ditch diggers.
The government originates money, by declaration that *these electronic records are official money* for the most part and to a tiny extent by printing bills.
The choice to issue federal debt is related to some other purpose than “getting money to spend”.
Worse yet the only way to “pay off the debt” is to tax people, removing money from the money supply and if we DID pay off the debt it would be a disaster causing the economy to lock up.
It’s not just that we don’t “have to” pay off the debt, it’s that it’s absolutely essential that we NOT pay off the debt
> The government simply does NOT borrow.
I wonder why your government pays me interests twice a year and will (hopefully) reimburse me the principal when the day comes.
Carlos, it’s a good question. Perhaps I should have said there is no requirement to borrow. If money is metals, there’s a finite quantity which changes slowly as we mine, so there’s a physical requirement to actually borrow coins.
But money isn’t like that today, even though we treat it like that.
Issuing bonds is a choice we make. We could just as easily simply spend the money into existence. But this can cause inflation, so then we’d need to tax money out of existence, and that’s politically painful.
Bonds are in essence a political negotiation l. Unlike a household the government CAN spend without borrowing or selling a product first, when it doesn’t its a product of a political process, a kind of buying and selling favor.
Also when the govt sells a bond to a person or fund it reduces money supply, but when it sells a bond to a bank it doesn’t, since banks buy bonds with reserve which isn’t money. So in that sense they aren’t even necessarily “borrowing money” with bonds.
So, it’s complicated, but mostly it’s a way to make policy more palatable.
I own several government bonds. I’m pretty sure that I lent them my money – and I receive interest and expect to be repaid my principal (though the current administration makes me wonder if I can count on that). I believe the Fed open market operations are a principle way for the government to finance the debt. At least that was the way I learned it (and taught it!) back in the days that you could get me to do any macroeconomics. But if that is all wrong, please enlighten me.
Not sure about that. Afaik there has been only ONE published attempt in history to actually study the dynamics of this process, and they conclude no.
https://www.sciencedirect.com/science/article/pii/S1057521914001070
Why don’t economists ever study this stuff so everyone can stop writing endless paragraphs?
Dale, from your perspective it appears you lent them “money” but money is something they can manufacture in unlimited quantities. So, no. Not in any way that makes sense such as compared to you borrowing money to buy a house for example. When you do that you are literally constrained by lack of money, you’ve identified a house for sale, but you need money to buy it. but the govt is NOT constrained by lack of money.
The government IS constrained by lack of resources, and by political will to protest or riot or civil war though. For example if it wants to occupy an office building in LA but there are no free buildings, it can’t do that without kicking someone out or building a new one. If it is too brutal with its resource acquisition it may trigger protests or worse.
So, the government may want to build a building say, and this means it wants others to not use the steel, concrete, wood, and labor it needs. It can do this by a number of methods, by outbidding using it’s spending/printing power (prices go up and we get some inflation), by taxing so people who build buildings don’t have enough money to start building, by temporarily reducing the availability of money by locking it away in an agreement in which you don’t spend your money until later (selling bonds), by increasing the cost of commercial borrowing (raising interest rates) or by eminent domain where it forces people to give up resources in exchange for money at a fixed price it determines (possibly other mechanisms as well)
So all of these mechanisms are expressions of the violent power of the state to compel but also the preference to negotiate for cooperation. Rather than financial resource constraints as you have when you try to buy an existing house but don’t have cash to pay.
That’s the sense in which I mean that the government doesn’t borrow money.
It’s quite convenient though that by selling bonds the government not only temporarily reduces the availability of money (by locking it away in an agreement in which you don’t spend your money until later) but also acquires the ability to CREATE the same exact quantity of money – to pay for the building – keeping the money supply constant.
> Rather than financial resource constraints as you have when you try to buy an existing house but don’t have cash to pay.
Google doesn’t have any financial resource constraints – they have $96bn in cash, cash equivalents, and short-term marketable securities. They choose to do a massive sale of bonds in 2020 – for reasons – but in a sense Google doesn’t borrow money.
Daniel
I really can’t understand what you are saying. I believe the government is borrowing money when I buy their bond, and more significantly, when they sell a bond to financial institutions. The points you raise seem to me to say that the government has other choices – that I don’t have regarding how much they spend and what they spend it on. That is certainly true. We’ve had cases where governments created money by running physical printing presses to create currency, something I am not able to do. That predictably had bad consequences in terms of hyperinflation. But that is not something the US government has done. The availability of multiple choices does not nullify issuance of government bonds as “borrowing.” You also keep bringing in issues about taxes and government ability to compel which don’t seem germane to me. I’m just missing too many pieces of your argument to understand it.
Not really, because the treasuries can be used as collateral to get a loan. In fact this appears to be the by far most important role they play, check that Jeff Snider discussion below.
One of the concepts is that modern financial crises are caused by too little available collateral. Ie, not enough government debt. This generally fits with what Daniel is saying.
> This generally fits with what Daniel is saying.
For what it’s worth, it was Daniel who identified “selling bonds” with “temporarily reducing the availability of money by locking it away in an agreement in which you don’t spend your money until later”.
You are correct, it came from Daniel.
Well, its technically true that “temporarily the money is locked up”, but afaict the second order effects are the opposite (ie, more money can be generated from that original amount). And to a quite extreme extent.
Carlos, the government creates money every time it buys something. At its core, It does this by having guns (and thus political power to demand that people treat its money as money). It doesn’t temporarily take $1000 from you and then spend the $1000, that wouldn’t create any money on net. Yet the money supply goes up. How does it go up?
There are two issues, one is what is the incantation it uses to legitimize its money creation (ie. it follows some rules it is always free to change but it pretends are somehow “legit”) and two is the reality of the government’s power.
The incantation they use to legitimize the fact that deficit spending creates money is that they in essence sell bonds to themselves. The Fed buys govt bonds with an infinite supply of “reserve”. All that incantation is just so they don’t have to admit that every single dollar spent by the government creates a dollar out of thin air. Where’s the evidence for this view? here in these timeseries:
When COVID hit, M2 money supply went through the roof almost overnight in march 2020 or so:
https://fred.stlouisfed.org/series/M2SL
However, the M2 / Debt ratio was unchanged. The fred website changed so I can’t link to you custom graphs anymore, but if you graph the above divided by the total public debt series below, you’ll see it has no discontinuity through the 2020 event.
https://fred.stlouisfed.org/series/GFDEBTN
So every dollar created was also a dollar of bonds sold.
If the government is *borrowing* money, so that you’re buying bonds with $1000 of your cash destroying $1000 temporarily, and they’re spending the $1000 you gave them, which creates an equal amount of new money, then money supply would be unchanged. If so, then *how could M2 suddenly jump up*?
The answer is, the government is really selling those bonds to itself through a money laundering process to legitimize what it’s up to, and it confuses people enough that even economists such as Dale perhaps don’t know what’s going on.
Here’s the asset level of bonds held by The Fed itself:
https://fred.stlouisfed.org/series/WSHONBNL
If you graph M2 divided by The Fed assets, it actually went DOWN. So The Fed bought all those newly issued bonds *and more*.
Now I’m not an expert in the laws here, but I don’t think the treasury is allowed to sell to The Fed directly, so the way this works I think is that Treasury sells bonds to Banks, who buy the bonds with Reserve (which is not money because it can only be used to buy bonds and other financial assets and to pay other banks, not buy services and goods). Then, immediately after that, the banks say “gee we need more reserve” so they go to The Fed’s window and sell the bond to The Fed, who creates more reserve in exchange for the bond out of *completely thin air*. The Fed has an infinite supply of reserve. And so, the US government has an infinite supply of money because whenever it needs to create money it can do this money laundering jiggery pokery. Which is just a pointless bunch of bureaucracy to say it creates money out of thin air.
This means, money supply went up, because the government spent $1000 but no one like Dale gave it any money.
All of that is just wanking to hide the fact that the government creates money every time it decides to spend, and it does so at its root, with guns (from which all political power flows).
Now. Sometimes the Treasury sells bonds to Carlos or Dale who do in fact give them actual Money. This is what I meant by temporarily you make a contract with the government that you won’t buy some stuff so they can have the stuff instead. And yes, when that happens, they spend new money into existence, and take your old money out of existence temporarily, and so it appears that they “borrowed money”.
But they have an infinite supply of money, so they are not constrained by having to borrow Carlos or Dale’s money before spending it. Unlike anyone else, such as if Dale want’s to buy a house and needs a mortgage first. So if the purpose of selling a bond to Carlos or Dale isn’t to get money so they can spend it (because they *could* spend it even before getting it) then there must be a different purpose. And that purpose is to reduce demand from Carlos or Dale temporarily so the government doesn’t have to compete with you for real goods and services driving up prices.
Another way to reduce competition with the private sector is to raise interest rates so that people won’t demand bank loans and then go out and buy stuff the government wants (like concrete, steel, or computers or labor hours). Bank loans *also* create money, out of thin air. Banks may write you a new balance in your checking account (creating money) provided they get something of value from you to convince The Fed that they’re solvent. The thing of value is your IOU for future payments. Provided they can convince the markets that your IOU is worthwhile, they can continue to operate according to The Fed rules. Or they can sell your IOU to someone else, or sell securitized tranches of multiple mortgages, or whatever. But in the end, provided they have enough assets on the books, they can continue to create money out of thin air through personal bank loans/mortgages/business loans etc.
So we have the following rules that can replicate what actually happens when we look through the fluff:
1) The government creates a new dollar every time it spends a dollar, laundering it through banks and The Fed and creating bonds largely as an accounting device.
2) The government can destroy dollars permanently by taxing them
3) The government can destroy dollars temporarily by selling bonds to private bank account holders like Dale and Carlos in order to reduce your present demand for goods and services, in which case the bonds play a slightly different role than if they wind up held by The Fed or Banks.
4) Banks create money by giving personal/mortgage/business loans
5) Banks can only do (4) so long as they have enough equity, which means bonds, IOUs, cash, real estate assets they could sell, etc etc.
6) The Fed has an infinite supply of reserve and can choose to buy bonds and other IOUs at any price and in any quantity they like, controlling the ability of banks to lend by manufacturing reserve out of thin air.
7) There are several real world constraints on all of this such as: govts can’t buy stuff that doesn’t exist, and govts will have a hard time if they screw with the financial system so much that they trigger armed insurrection and have to actually make good on the threat of violence that underlies the government’s power.
If you run an agent based simulation with the above ruleset you’ll find that it can reproduce the dynamics of the M2, Fed asset level, total public debt, etc etc.
If you believe that the government must first sell bonds to “get the money” to buy goods and services, then you *can not* explain why M2 jumps overnight, and so your model must be wrong/incomplete.
Daniel
I don’t have the stomach for the detailed measurements of the money supply, but I have a hard time reconciling what you say with the Fed actions during COVID (see https://www.brookings.edu/articles/fed-response-to-covid19/ for a description of those changes – which sound like large government “borrowing”). At heart, regardless of how we measure “the money supply” (none of the traditional measures which I particularly like), the issue is how/whether the government can expand or contract the ability of economic actors to spend. I do believe the government (I won’t distinguish between the Treasury and the Fed here, though I think that detail is important) has a unique ability to expand the economy’s ability to spend. I think we saw that in use during the Great Recession and COVID. I won’t go so far as to say that every dollar Treasury spends increases the money supply by $1, but it is the ability to create money that leads some economists to believe the Fed should use limited (or no) discretion to increase the money supply (Friedman’s famous declaration that inflation is “always and everywhere a monetary phenomenon”).
I do wish you would drop the inflammatory language such as “laundering” and “incantation” – you are getting close to calling Social Security a “Ponzi scheme.” The fact is that the entire financial system is built on the fact that it isn’t feasible for people to collect their “money” (however measured) all at the same time. If we all tried to cash in our IOUs simultaneously, we would rapidly find ourselves unable to be made whole. I don’t view that in negative terms at all – it is the main virtue of financial institutions. And the Fed’s ability to create money is a similarly positive tool.
I think the main constraints on these institutions is credibility and trust. Your analysis strikes me like a conspiracy story of how the powerful use financial institutions to enrich themselves at the expense of the rest of us. While I have no doubt there are instances and abuses like that, I just don’t buy the institutional conspiracy you seem to suggest. I am probably not understanding your points, but I don’t think the details of how the money supply is measured are really that crucial. In any case, I have as much interest in exploring those measurement details as trying to understand the tax code. Sorry.
Dale, to bring this back to the original context, which was Andrew describing the US debt as similar to household debt.
Every dollar the govt spends which isn’t matched by a dollar of taxation is a dollar created, either now, or at some point in the future, so the government has decided it should be paired with a bond for accounting purposes. These bonds don’t represent some limited physical objects (like gold coins) whose use the government borrowed that need to be returned, they represent the literal creation of money.
The government CAN NOT pay it down, that would be disastrous because it would annihilate all money.
This is WILDLY different from household finance.
The public debt isn’t an inconvenient situation that needs to be “remedied” it’s an essential component of the currency. We could just declare that bonds would no longer be created, and the scheme would work similarly, but we’d be forced to tax more quickly to keep the money supply from ballooning.
I use the words “laundering” etc because I believe that this process is intentionally made opaque, that The Fed intentionally misleads in its wording on online documents describing the monetary system, and that economists misunderstand what’s going on.
The “modern monetary theory” people who are considered heterodox economists are often ridiculed by the rest of economics like Semmelweiss was for insisting doctors should wash their hands.
The situation is pretty dire and worthy of inflammatory language because this whole scheme is part of the transition to fascism. The notion that the debt needs to be repayed is a kind of propaganda which always comes from conservative members of government (such as Republicans and also many Democrats) and is explicitly a part of the austerity and govt destruction that the DOGE people have going on now.
Pushing back against that requires some inflammatory argumentation I think.
To add to Daniel’s charts regarding spring 2020, which show ~$3 trillion in new debt and the same in M2, with the vast majority in money market funds and saving/checking accounts. Then ~$1.5 trillion in treasury assets held by the fed.
Total reserves held by banks, also rose ~$1.5 trillion:
https://fred.stlouisfed.org/series/TOTRESNS/
To complete the accounting, we need to know where is the extra $1.5 trillion and how exactly did the total $3 trillion end up in those MMF/checking/savings accounts?
> Every time the US govt spends $1000 the money supply goes up by $1000
The simplest example is that if the US government spends one dollar that the US treasury kept in a box the money supply does clearly go up by one dollar (because the currency held by the US treasury is excluded from the money supply calculation).
Of course it’s just a choice of the US government to borrow – and collect taxes – instead of printing more dollar bills to keep the box replenished. Maybe Trump will retake the idea of minting a trillion platinum coin, wouldn’t that be fun?
You can usually tell that someone has no idea how macroeconomics works if they mention the national debt. The deficit is sometimes relevant, but the debt almost never. Considering the basics have been understood for about a century, it must take some effort to not understand it.
‘The deficit is sometimes relevant, but the debt almost never.’ Allow me to specify the cases where either of these are relevant. The level of debt is highly relevant to a country’s interest payments. The higher the debt, the higher the cost of servicing it to keep it at the same level (no surprise there). If the funds needed to service the debt are high relative to the government’s budget, this is political dynamite. It greatly constrains what an indebted government can do.
The deficit is relevant in the context of intergenerational equity. If the deficit is positive, it means that today’s government spends money today and tomorrow’s generation pays back the debt tomorrow. This is an intergenerational transfer of wealth.
In summary, debt is relevant when interest is high relative to the government’s budget. The deficit is relevant in the context of intergenerational equity.
> If the funds needed to service the debt are high
This recent paper about the tipping point “at which interest payments exceed defense spending […] after which the centripetal forces of the aggregate debt burden tend to pull apart the geopolitical grip of a great power” is interesting (right or wrong, I found it interesting).
https://www.hoover.org/research/fergusons-law-debt-service-military-spending-and-fiscal-limits-power
That is not a bad concise description of the relevance of debt and deficits – but relevance does not equate with “problem.” High interest payments indeed constrain government finances and (without considerable political effort) shift government spending from providing services to servicing the debt. But that means you want to know who holds that debt – who is receiving that interest, and then what do they do with those payments. The chain is somewhat complex. So, I agree with the characterization of ways in which debt and deficits are relevant, but that is only the first step in understanding its implications.
As for some tipping point between defense spending and interest payments, I’ve never heard such a theory but it sounds pretty shaky to me – along the lines of the RR critical level of Debt/GDP, or for that matter any of the myriad statistical artifacts we routinely discuss on this blog.
David Marcus says:
“The deficit is sometimes relevant, but the debt almost never.”
You are aware that even the great John Maynard Keynes has written many things about the dangers of high debt.
You don’t have to be an adherent of the Austrian School to be concerned about the national debt.
Daniel says:
“Every time the US govt spends $1000 the money supply goes up by $1000… So, the government has a means to destroy money as well. That’s called taxes.“
These statements seem to be confusing the fiscal policies of the federal government and Treasury with the monetary policies of the Federal Reserve.
You could say it is indirectly, since the gov sells treasuries to (gets loans from) the primary dealer banks, who then can sell those to the fed.
Evidence for it is the super-inflated prices of everything the government starts to spend money on. The most obvious are healthcare and higher education, which have skyrocketed since the 1950s. This is exactly as you expect from the Cantillon effect.
Back in the day, if a new gold mine was discovered, whoever got the gold first could spend it before prices adjusted. Then prices rise nearest the mine first before spreading afar. Now the “gold mine” is the source of new ledger entries, and the “locals” are the government and banks (the other super-inflated prices are financial assets like stocks and real estate). The “next town over” is people working in those industries, and finally at the bottom is the average wage-earner. Once the price of labor (wages) starts rising, the fed tries to stem the flow of gold from the mine. They explicitly say this is their policy.
This is another thing, like studying the actual dynamics of bank loans, that seems to be ignored by economists. It doesn’t even have its own wikipedia page.
> You could say it is indirectly, since the gov sells treasuries to (gets loans from) the primary dealer banks, who then can sell those to the fed.
You could say many things but the Federal reserve holds less than 15% of the US Treasury debt outstanding.
If you read carefully, I was careful and referred to “source of new ledger entries” which need not be the fed. What actually happens seems more like as discussed here:
https://www.macrovoices.com/podcast-transcripts/548-jeff-snider-eurodollar-system-overview
It’s not clear to me that it’s necessarily the _government_ spending money on that drives up the prices. If you are pointing the Cantillon effect and saying that things purchased closest to the creation of money should experience the largest increase in prices, then you would probably have to recognize that the government (to my understanding) is not the source of most money creation, but rather lending via banks.
Since money is created in banks via lending, one could look at the things that people are most willing to have a lending demand for. Higher education has obviously been a benefactor of massively increased demand and willingness of people to take on enormous amounts of debt. Higher ed debt is mostly public, so it seems sensible that the government’s willingness to support this lending has had an impact on cost.
I think housing is another important area to look as it is another instance of enormous increases in cost and similarly represents something that people are willing to take on a significant amount of debt to afford. From a very cursory search it seems there is an order of magnitude more mortgage debt than student debt held by Americans ($12.6 trillion vs $1.7 trillion). Does that mean that the money creation due to housing debt is significantly greater than money creation via student debt? I don’t know. For mortgage debt, it’s less clear to the me the direct effect of government spending causing the issue? Since my understanding is that the government largely does not play much of a role in supporting mortgage lending?
I’m not an economist, but it seems like
1) the proximity to money creation leading to outsized growth in prices seems compatible with what’s happening in the economy
2) most money creation happens in banks not the government
3) the massive price growth over the past few decades is not restricted to areas with heavy government spending
I also think it’s hard to make particular claims at the direction of causality in these areas as well.
Carlos, I’ve got a longer post with multiple timeseries explaining monetary dynamics, so it’s held in moderation hopefully it’ll become visible soon.
If the govt sells a bond to Carlos (or some other private entity with a checking account etc) then it destroys money temporarily.
If it sells the bond to a bank, then it doesn’t, because banks buy bonds with *reserve* which is only usable to buy financial assets and pay other banks, not general goods and services.
If a bank doesn’t want to hold a bond and needs reserve it can sell to The Fed who will create reserve out of nothing (subject to some negotiations etc).
The asset level of The Fed itself, plus all bonds held by Fed member banks as a fraction of total debt is about 32% right now.
So, over the last decades, when the Treasury spends $1 it creates $1, it also creates $1 of bond, which on average destroys $0.68 of public money (temporarily) and destroys $0.32 of reserve (non money). However this is averaged over the entire history of issued debt. In any given moment, such as during COVID 19 panic of 2020, $1 of new spending produced about $1 of new money with all of the bonds being bought by banks, and then immediately sold to The Fed to replenish reserve. All interest paid to The Fed is returned to the Treasury, so in essence the govt sold bonds to itself. This is just a legitimization technique, the bonds are irrelevant accounting tools when this occurs. my larger post explains the details.
It’s hard to know what – or who – are you arguing against. I’d say that all the commenters who resist being enlightened by your diatribe do understand that selling debt is just one of the fourteen things that the Treasury could do. Some things require approval from the Congress, some don’t. The “government” in a broader sense – including the Federal Reserve can do even more things.
You may insist that the Treasury isn’t really “borrowing” because it doesn’t need to. But neither do Google or Harvard need to sell bonds – just like many people don’t need to get a mortgage loan. A different issue is whether these things do or do not increase the (different kinds of) money supply. That depends on what form things take, into whose balance sheets they go: a mildly interesting technical subject that becomes boring quickly. The Treasury is not unique either in rolling its debt over, companies and individuals do the same. Of course increasing the debt is also a choice, while it has very rarely come down in nominal terms it’s not so rare in real terms and even less as a fraction of GDP. (Sure, the Treasury is special for other reasons. Nobody denies that.)
Anyway, everyone understands that the money supply may decrease or increase at the same time that outstanding debt goes up or down. In the first half of 2020 there was a net issuance of around three trillion dollars. The Federal Reserve bought less than that. The money supply increased more than that. These things are related, depending on the details of what happened, but they are not magically linked.
Since June 2020 the money supply has not increased (or has decreased, depending on the choice of money supply) while the debt outstanding has increased by seven trillion. In the last twenty years the money supply has increased fifteen trillion while the outstanding debt has increased by thirty. I don’t quite understand why this long-term imbalance becomes irrelevant to you because it’s “averaged over the entire history of issued debt”. What you call “in any given moment” is quite the contrary: in some specific periods.
Or course things could have been different and of course they could still change. Everyone understands that the US Treasury could cancel its debt at anytime printing money. More precisely they could mint a quadrillion dollar platinum coin (an interesting technical detail about platinum coins is that you can write on it a number as large as desired as long as the coin is large enough). Maybe Trump will get to it once he has fixed the trade imbalance. It will also work tremendously well.
> Since June 2020
For the record, I was writing from memory and I think I mixed things up. The “flat/decreasing money supply while debt goes up” applies since December 2021, if I remember correctly, and since June 2020 the money supply also goes up but not as much.
Carlos I think the topic is overdone for the blog here so far. I wrote up a series of examples about how money comes about on mastodon. If you care to discuss further perhaps there is the right place. I’ll write one last summary here of my position, and then invite you or anyone else to discuss over there.
here’s the thread with the accounting examples
https://mastodon.sdf.org/@dlakelan/114297633446329313
I’m mainly pushing back against three things:
1) That the government must borrow money from citizens by selling bonds before it can spend more than it taxes (instead, every bond sold to a bank creates money rather than “borrowing it”).
2) That the government debt represents “money it borrowed and needs to pay back”. (Instead from an economic perspective the government debt represents money it created, plus bribes it promises to rich people to keep their consumption rate down in the short term. Alternatively you can call it a promise to transfer money from poorer people to richer people in the future.)
3) That it is even possible to pay off the debt. (it isn’t because it would tank the money supply by a vast amount and lock up the economy).
all three of those are common misunderstandings that are utilized to the benefit of rich people and right wing politicians
I think this is far from some technical bullshit. It’s core to a way that right wing politicians utilize power. If government debt is like household debt then increasing it without bound is basically borrowing physical stuff from people and never giving it back. That gets an automatic kneejerk reaction from people who have had their lawnmower or power tools borrowed and never got it back. But government debt is NOTHING LIKE household debt and the narrative of “we need to pay it down” is actually a hidden political demand to do violence to poor people. Democrats like Bill Clinton were proud of themselves for “running a surplus” but that surplus was a transfer from poor people to rich people and created the dot com bubble, and triggered a series of catastrophies that led to the financial bullshit situation we’re in today. (https://fred.stlouisfed.org/series/WFRBST01134 shows how top 1% got massive increases in income between 1992-2000)
The government is not like Google in the sense that Google issues bonds but doesn’t have to right now because it already has some cash on hand. The government NEVER has to issue bonds to its citizens to get cash because it ALWAYS has infinite cash on hand. And money can ONLY come into existence under current rules by the govt selling a bond to a bank, or by a bank loaning it into existence against its loaning power granted by the government (which requires mainly that it have positive equity) so when the govt creates money it always has bonds associated with it (but not every bond equals new money right away, only those sold to banks get accounted right away as new money).
To inflate the currency the government doesn’t need to strike some platinum coin. In fact, there’s no reason for the coin to be platinum, it could declare a piece of polyvinylchloride to be a 100 trillion dollar “coin”.
But it doesn’t even need to do that. It just starts selling bonds all day every day. Even if some citizens buy them, eventually only banks will buy them, and then sell them to The Fed for more reserve so they can buy more.
The govt can make as much money as it wants by doing this at any time. Of course, it’s a disaster waiting to happen, but… it’s precisely the trajectory we’re on. https://fred.stlouisfed.org/series/GFDEBTN so it seems relevant to bring it up when we start discussing debt.
Having one name for two different concepts is a common way to create intentional confusion. “Government debt” and “statistical significance” are kind of similar in that sense. Just like statistical significance doesn’t mean what most avg scientists think it does, also “Government debt” doesn’t mean what most people think it does.
> In fact, there’s no reason for the coin to be platinum, it could declare a piece of polyvinylchloride to be a 100 trillion dollar “coin”.
Not according to the current law: https://www.law.cornell.edu/uscode/text/31/5112
But sure, the law could also be different. You’re right about everything. In particular about the topic being overdone. Let’s see how long until it’s brought up again…
>We’re the richest country in the history of forever. … , and our national debt keeps going up, up, up
This improperly conflates society, which is rich, with govt, which is in debt. Household debt is in fact going down as a pct of GDP https://fred.stlouisfed.org/series/HDTGPDUSQ163N and most govt debt is owed to Americans https://www.pgpf.org/article/the-federal-government-has-borrowed-trillions-but-who-owns-all-that-debt/ so it seems unlikely that the net debt owed by society has risen.
Isn’t the simpler explanation that Atlas isn’t broke, but isn’t self-aware enough to realize this?
Spending money on the second home while barely being able to pay recurring bills seems to indicate a sort of mental partitioning; the former is an untouchable pot of money, and the the second pot is nearly empty. One could, however, combine the pots (or eliminate the first one). I have the same impression when university administrations claim, sincerely, that they have no money; the pot for administrative salaries or non-instructional expansion is oddly separate in their minds.
That’s kind of a sad story. My parents never had a Beamer or a Jag, but they paid their bills without fail. Of course we were lucky: my dad’s career took off early when he crawled into the gunner’s nest of a B-29 around 1950 at age 21. He stuck with that for 20 years. You can imagine the wealth we accumulated, which thankfully made it easy to pay the bills. Sadly though by the mid-1960s he had been sidelined to “instructor” and he never got to bomb Cambodia or Hanoi and get combat pay. But some people on our block were shot down so the families got money from that. Just luck.
Beemer or not, we were blessed with some of the most amazing auto technology to ever put rubber on a highway. After Pop retired from the USAF in the early 70s it was one thrill after another. Despite his deep commitment as Captain of the Not-Handy team, during that time he miraculously found time to perform a full rebuild of the engine in one of our legendary muscle machines, the 1966 Mercury Commuter station wagon. When we finally allowed that amazing machine to go to a caring scrap collector, we acquired another of the US Auto Industry’s most impressive pieces of engineering: a Dodge Aspen. Later we owned *still another* shining example of Yankee Ingenuity, an AMC Matador! Just very lucky I guess.
Eventually dad tired of the Post-USAF playboy life style and got serious again. By the early 80s had finished a CPA and was back to producing massive wealth, doing sewer project accounting for EPA grants. Pop had all the luck, spending his weeks galavanting around America’s leading hotspots like Post Falls ID and Walla Walla WA. Still, despit their insidious wealth, just for fun, my parents never stopped using whatever tricks they could to increase their massive wealth. A year or so before he passed away and he was making his arrangements, Dad bought $30K worth of US Treasury bonds with his Discover card so he could get the cash back.
Nearly a decade after dad passed away I helped mom buy her first-ever new car – the latest topest Japanese tech, a gleeming silver Mazda 3! She was 84. The plan was to cash in some of the bonds dad bought with his Discover card and pay cash for the car, but with her 800-odd credit score she qualified for a no-interest loan. I harangued her into taking it. It was all I could do for the next six years to keep her from paying that damned car off early.
Despite many family struggles, I doubt my parents ever had any regrets about finances. (No, wait, that’s wrong. There was a big fight over their Microsoft shares in the late 1990s. It was the only time Dad was right about stocks: they should have sold them!) To be born in the US in the 20th Century was to win the lottery. They understood that.
Owning illiquid assets yet not being able to meet running expenses is actually fairly common, for both people and companies.
Both selling and borrowing against these assets involve large transaction costs, and it may be optimal to just tighten the belt if possible if it is expected that things will get better.
Assorted thoughts:
Atlas’s widow I believe is a Psychiatrist in New York City.
This reminds me of a post I can’t dig up by a University of Chicago economics professor, perhaps Casey Mulligan but my apologies if he is not the person, on how he couldn’t make it on $450,000 a year. I believe you commented on this in this blog, but I can’t find it. If I remember correctly what each of them wrote, a lot of it seemed to have to do with what is considered “necessary” given where they live, their types of jobs, and the expectations of the groups they associate with. In NYC I would imagine that has to do with living in the “right” neighborhoods, going to the “right” schools etc etc. I have family in NYC so I am not unsympathetic,
Later in life Atlas was running his own imprimatur. I do not know how well it did commercially.
The Bonfire of the Vanities includes a description of why Sherman McCoy is barely getting by on $950K per year or something. That seems laughably low for a “Master of the Universe” stock trader, even adjusting for forty year of inflation — I think someone in his position would be closer to the $30M/year level nowadays, and it’s not like prices are up 30x — but still it was a heckuva lot of money back in 1987. Anyway there’s the penthouse apartment, the love nest for assignations, private school for the kids, maid, vacations, yada yada…hey, it adds up!
Famously, the richer you are, the more money you think you need to be ‘happy’ or ‘satisfied’. There have been many polls over the years, here’s a recent one: https://www.axios.com/2023/12/01/money-needed-to-be-happy-wealth
I can’t vouch for this specific survey but the findings are generally in line with every other similar survey I’ve ever seen so at least qualitatively this seems to be a real thing.
One of the interesting things, maybe not surprising but interesting, is that the amount of _additional_ money someone feels that they need in order to be financially satisfied is roughly proportional to their income. It’s not like everybody thinks they need an additional $25K per year, to replace one of your household’s two cars every two years or whatever. Instead, if you make $60K/year then you think you need an additional $25K, but if you make $160K then you think you need an additional $50K, and if you make $260K then you think you need an additional $100K. Hey, I’ve got news for you: if you make $260K per year and you aren’t happy, more money is not going to solve your problem.
That said, I do understand the phenomenon of ratcheting up expectations (and expenses) and then suddenly something unexpected happens and you find that you’re worried about paying the bills. You decide to buy a house, prices are higher than you expect but you’ve got your heart set on a specific neighborhood so you end up paying more than you planned and the mortgage payments are a bit of a stretch, and then you’ve got some medical expenses or something, or some nutty thing happens with the stock market and your plan of selling stock a bit at a time to pay part of the mortgage suddenly seems like a bad idea but that’s where all your money is, but if you don’t do that then where are you getting the money? This hasn’t happened to me but it’s not like I think I’m immune.
Still, if someone asks me and my wife how much income we need to be happy, we’re choosing a number lower than what we make now. We’re already happy!
Phil:
$950K per year is a heckuva lot of money now too!
It’s a lot of money, sure, but not a heckuva lot.
Phil:
I asked the person sitting next to me in the Playroom and he agrees with me that it’s a heckuva lot.
It’s not a shit-ton of money, though; I’ll grant you that.
To put it in better perspective than “heckuva lot,” $1M is just above the 99% quantile of U.S. household income as reported on taxes after deductions, so think of it as a lower bound on actual incomes. To be more precise, the 99% quantile mark is $660K and 99.9% is $3.3M. See https://en.wikipedia.org/wiki/Income_in_the_United_States
I doubt $1M household income puts you anywhere near the top 1% in NYC, much less in Manhattan where Andrew and I live.
To Phil’s point, I actually think another $250K take home pay (let’s say $500K income) I could spend on housing would make me and Mitzi happier. In Manhattan, you would be lucky to find a rental apartment for $260K/year that was 2500 square feet, the size of house one could buy for that amount of money in less expensive cities. For example, a 2400 sq foot (200 sq m) 3 bedroom penthouse (i.e., top floor) apartment on West 108th and Riverside (pretty near Andrew’s apartment) from which you could walk to Columbia is $24K/month (https://streeteasy.com/building/324-west-108/ph)—you can browse the rest of the site and see that I’m not cherry picking here. And that neighborhood is much cheaper than downtown where I live and work.
> Isn’t the simpler explanation that Atlas isn’t broke, but isn’t self-aware enough to realize this?
Why is everyone hearing hoofbeats and guessing “zebra”? The simplest explanation for someone not being able to pay their bills is that they’ve racked up more debt than they can afford to pay off. Having children, buying houses, buying boats, buying cars, and getting sick (in the U.S.) are bog standard ways to do this.
I like your explanation also. I was hoping for more of this, rather than dissertations on macroeconomics.
‘ Why is everyone hearing hoofbeats and guessing “zebra”? ‘
We’re in the new geologic epoch: the Anthropomeme. For Americans it means zebras make all the hoofbeats.