Yes. The amazing thing about this topic is how confidently otherwise intelligent people routinely spout utter bullshit about it. It's particularly ironic (but completely unsurprising) to see people flaunting their misconceptioms in response to an extremely clear explanation of how it actually works from one of the most authoritative sources possible.
Please, if you think you know something on this topic and haven't read this pdf, just try and forget what you think you know and read it.
(Of course, now somebody will explain to me why the Bank of England doesn't actually know how money creation works, or is deliberately lying about it for some reason, and on it goes.)
> to an extremely clear explanation of how it actually works from one of the most authoritative sources possible.
If I'm not mistaken late into the 2000s authoritative sources like the Bank of England were still not completely on board with the idea that commercial banks themselves are in the process of actual money creation, a process explained described/explained by Schumpeter (I guess amongst many others) ever since back in the 1940s.
So a little bit of "not following the authoritative sources" on this one is understandable, as those authoritative sources themselves are pretty new to the idea.
I think that fractional reserve banking assumes it starts with a person saving in a bank, but the article asssumes this is the reverse
"""This article explains how, rather than banks lending out deposits that are placed with them, the act of lending creates deposits — the reverse of the sequence typically described in textbooks.(3)"""
and
"""
While the money multiplier theory can be a useful way of introducing money and banking in economic textbooks, it is not an accurate description of how money is created in reality.
"""
As such there is a new understanding of money developing - QE, lending and even MMT are involved.
So I think this article says quite clearly that the Econ 101 idea of fractional reserve lending as a money multiplier based on deposits made is just out of date
I do see this as a good argument to remove money creation from banks because
"""
Banks first decide how much to lend depending on the profitable lending opportunities available to them
"""
which is determined by time horizons and risk appetite much more than central bank interest rates - and why most lending needs collateral.
Government lending (government as a VC) however has much longer time horizons
Yes, the article expands on simplifications used in economics textbooks as an illustration of how leverage can works. But none of this is "new understanding" of the fractional reserve system operates in practice, least of all to the Bank of England that oversees its operation.
> Banks first decide how much to lend depending on the profitable lending opportunities available to them which is determined by time horizons and risk appetite much more than central bank interest rates
Hardly "much more", the delta between the rate the bank offers and the central bank rate is the bank's profit, adjusted over a timescale as a net present value calculation and reduced by expected losses. More importantly, it's also the cost of the loan to the person or company deciding on whether borrow or not, so when banks bump their loan rates up in response to a raise in the central bank rate, fewer people wish to borrow and so the bank has fewer profitable opportunities to lend. Time horizons and risk appetite just give banks and companies reason to turn down probably profitable/beneficial loans that are outside their comfort zone. And how does a central bank respond to the banks collectively reducing their risk appetites? It lowers the interest rates....
> Government lending (government as a VC) however has much longer time horizons
Ultimately I'd rather have market as VC (government can participate as well) than government as the only VC...
1. It's nuts that all central banks globally have one lever to pull - interest rates.
2. yes the central bank sets the interest rate, but the banks choose how much to lend and to whom. I think that if interests rates are at zero, that implies (?) an infinite amount of investment opportunities - ability to grow productive capacity. And yet most loans are real estate or real estate backed (ie collateral). VC is a drop in the ocean of land based lending - and yet our entire economy is based on small amounts of land use for amazing factories and power stations
3. banks aren't doing the business of lending to those capable of productive capacity generation - and they won't cause they want safe near term returns.
VC speculative investment should not be the small corner case - it should be the norm. Industry investment should be not left to crazies willing to risk family house or wealthy with enough to spare.
I’m talking about commercial banks themselves being in the process of creating money, which in the UK was indeed not an established fact until the 2000s. I’m on my phone but I remember reading about that in an Economist issue a few years ago, I’ll try to find it once I get in front of a computer.
Later edit: Found an Economist review [1] of this article/book itself, which I think it’s not what I had in mind but close enough:
> This often-cited short paper lucidly explains how commercial banks create money and central banks influence that process. It dispels many common misconceptions about money. For instance, most introductory economic textbooks say that commercial banks lend out the money that savers deposit in them. In fact banks can lend money and create corresponding deposits even without savings flowing in–in other words, banks are quite literally creating “new money” when they make a loan and a corresponding deposit.
Most probably you were thinking about Central Bank money creation, or, if not, those introductory economy courses seem to have not had any effect on the educated masses, hence why The Economist still has to re-iterate to its educated readers how money creation works.
> I’m talking about commercial banks themselves being in the process of creating money, which in the UK was indeed not an established fact until the 2000s.
There is a concerted effort amongst some fringe economists to make this argument, which is superficially plausible to the sort of laymen easily convinced that reading a couple of blogs arguing that an entire field is wrong is a substitute for reading anything written by that field.
But of course the Bank of England knew that commercial banks were involved in the process of creating money all along. Propping up private bank money was literally their job.
> Propping up private bank money was literally their job.
I thought their job was limiting the private money creation that without them would be totally unrestricted which would result in uncontrolled inflation and runs on banks.
Private banks create monetary aggregates by lending it to private borrowers. The fact that money is created as credit means there are already natural limits to private bank money creation (there aren't unlimited numbers of bank customers who want to borrow money and pay back more later, and some of the people and companies that would like to borrow are not people the bank trusts to be able to make the repayment)
The role of the Bank of England is providing the banks with central bank reserves they can borrow as and when needed to back that privately created money up. So it gets to influence the demand for borrowing private bank money by setting the basic interest rate at which the banks can borrow reserves (mainly by intervening in secondary markets for them, but that's an implementation detail), which means it can make it more expensive to borrow reserves, which will lead to banks lending money at higher interest rates to fewer people, which will lead to less money creation
But if the private banks already have some money they can lend it again to borrowers. So they could be inflating money supply pretty much indefinitely. Sure, the number of private borrowers is limited but they are happy to sign all of their future life earnings away in exchange for some money now. And as supply of money grows inflation happens so borrowers need more and more money feeding creation of more money and further inflation.
The role of central bank as a limiter of credit action (through fractional reserve) is way more important than feeding new core money into the private banking system so it can be borrowed by private borrowers. Feeding new money is kinda optional and it's most important role is possibly enabling new banks to be created. In absence of it only companies that already have a lot of money could create a bank.
The other thing is that if economy development outpaces growth of money supply created by private banks whole system could get stuck in deflation. Which was not great last time it happened.
Honestly, I have no idea why you're arguing popular misconceptions about central banks under a paper written by a central bank explaining how central banking works...
Banks have absolutely no incentive to provide money to everyone that wants to borrow all their future life earnings now, not because of hard limits on their funds but because they want their lending to be repaid at a profit (when it won't be, you get 2008). So the quantity of money at a given interest rate is ultimately set by the demand of creditworthy borrowers at that interest rate, which is certainly not unlimited.
Central banks were created to stop banks with solvent loan portfolios collapsing due to demands on their reserves, by ensuring banks could always borrow the reserves to back up the numbers on their spreadsheet. The "reserve requirement" (technically replaced by a capital requirement) isn't something central banks tinker with, and it's not "kinda optional" for them to provide enough reserves for the system's day to day needs. Instead, they influence credit action by adjusting the price of borrowing those reserves, and thus the demand for credit in the wider economy.
> Banks have absolutely no incentive to provide money to everyone that wants to borrow all their future life earnings now, not because of hard limits on their funds but because they want their lending to be repaid at a profit (when it won't be, you get 2008).
2008 happened because clearly the banks do have that incentive. The only thing that stops them are the regulations.
> The "reserve requirement" (technically replaced by a capital requirement) isn't something central banks tinker with,
In the last two decades that requirement was adjusted at least 5 times in my country so tinkering with it is definitely a tool that some central banks use. In the nineties it was even set to 30% to quench hyperinflation.
> Central banks were created to stop banks with solvent loan portfolios collapsing due to demands on their reserves, by ensuring banks could always borrow the reserves to back up the numbers on their spreadsheet.
First formal central bank, The Bank of England was created to finance the war. Central banks gradually acquired their modern roles as they developed.
The role you so much focus on is called being 'lender of last resort' to private banks. Private banks use it only if they can't get money cheaper anywhere.
> they influence credit action by adjusting the price of borrowing those reserves, and thus the demand for credit in the wider economy.
That the thing they do most often but not their most powerful tool.
> 2008 happened because clearly the banks do have that incentive. The only thing that stops them are the regulations.
2008 happened because banks overestimated the resilience of the value of housing collateral that backed their loans to an economic downturn, meaning they got back less than they lent out, not because they had any incentive to inflate money supply pretty much indefinitely, which they obviously didn't do (I mean, if they really had no practical constraints on money creation, they could have solved 2008 for themselves by loaning unlimited amounts to pump house prices back up again...). Inflation wasn't even high in 2008.
> The role you so much focus on is called being 'lender of last resort' to private banks. Private banks use it only if they can't get money cheaper anywhere.
And how do they get cheaper money elsewhere? They borrow it on the interbank lending market, the one which the central bank actively intervenes in to set the base interest rate by buying and selling bonds in sufficient quantities to drive the rate up or down (Why does the interbank lending market even exist? Because the 'lender of last resort' makes lending spare reserves to other banks a low risk activity equivalent to exchanging them for government bonds) Seriously, I suggest you read TFA which explains all this rather than continuing to insist that it is wrong ...
> That the thing they do most often but not their most powerful tool.
It's their most powerful tool in normal circumstances, and how monetary policy works. Preventing private credit creation isn't a tool, it's a nuclear weapon, and one most likely to be introduced and enforced by a government department which isn't a central bank...
> Does that mean that currently US banks don't need to hold any fractional reserve?
They still need to maintain some reserves to permit cash withdrawals and certain transfers, and are still incentivised to lend out at a higher rate than the base interest rate the central bank controls (and reserves still count towards the bank capital requirements which are the actual limiting factor on a particular bank's ability to lend). So the central bank still can act to encourage or discourage money being created by lending exactly as before, the banks just don't have an arbitrary reserve target to hit.
It's not a policy which hasn't been adopted much earlier in other parts of the world, or a cause of the current inflation. The UK had no reserve requirement at all during the mostly low inflation period since 2009, and small symmetric reserve targets chosen by the bank where they incurred a penalty for having too many as well as too few reserves between 1980 and 2009 (a period where inflation came down from pre-1980 record highs to an unprecedented long, stable period of low inflation)
>2008 happened because clearly the banks do have that incentive. The only thing that stops them are the regulations.
It’s truly absurd to argue that banks have no incentive to limit lending to make sure they get paid back. The 2008 crisis proves no such thing. Banks simply made a lot of bad bets.
They had all the incentives to make all those bad bets. And to counteract those incentives further regulations needed to be enacted.
What's good for the company in the long run is not necessarily what really happens because companies consist mostly of short sighted decision makers driven by short term incentives. They will happily collectively burn their respective markets to the ground if it brings them short term profit.
What helps large corporations to survive this is their sheer inertia. Banks are kind of special because they, despite being huge, can topple very quickly.
Funny that, because I distinctly remember participating in an early 2000s competition the Bank of England organised for schoolkids to teach us about monetary policy. Of course, that didn't exactly guarantee participants a graduate-level understanding even if their team got through the first round, but it's not exactly the action of an entity that's trying to hide what they do!
Maybe people not understanding the mechanics of complicated things isn't actually the result of a conspiracy to stop them from understanding complicated things. Perhaps they'd also be somewhat more likely to understand how complicated things work if they didn't start from the premise that the people who know how they work are inherently untrustworthy...
The role that commercial banks in money creation has been known since much before the 2000s. It's all laid out in Keynes' General Theory which was published in 1936, and it wasn't even a novel idea at the time.
Like I said previously, the Bank of England wasn’t acknowledging that as a fact publicly until quite recently, which most probably had an effect on their actions/decisions.
For example the same Keynes had some other better known takes which have also not been acknowledged as facts to this day.
So you're saying that the Bank of England was involved in a conspiracy that wouldn't acknowledge something that has been public knowledge for over a century?
There's even patterns to it. I've learned to avoid posting to (or even reading) comment sections touching certain topics, and to postpone submissions of certain stories until Monday, when the quality of discussion gets back to normal.
It also happens frequently on economic topics, especially regarding inflation and monetary policy, which seem to be fertile topics for conspiracy theories.
It isn't just finance and economics, although I totally agree with you that they're especially bad. The forum has become too wide in scope and started attracting the usual online idiot horde that dominates all other sites already.
You can see this even in tech. Most of the commenters have been handed incredibly powerful database / IDE / computation tools which makes them think they're very smart to be playing around with it. What they don't realize is high level tools are not meant for people who want to understand the tooling. They are meant for users of the tools whose task is something else entirely. So you get folks who don't understand how a simple race condition would occur or something else very obvious to folks who have done a bit of low level programming, but they will hold forth on mutexes and locks as though they invented them.
Interest rate and monetary policy articles on HN never produce productive discussion, for various reasons, which are quite obvious if you scan the comments of these types of articles so I won’t elaborate on what they are.
I quite often write and post a comment to correct a misconception and just end up deleting it because the person I am replying to is going to dismiss what I am posting anyways. Now I don’t waste my time.
Reading the source. It feels like it’s written for a specific audience, perhaps people who have taken a year of economics or more. So it’s difficult for an engineer to parse. One example is this paragraph:
> when households choose to save more money in bank accounts, those deposits come simply at the expense of deposits that would have otherwise gone to companies in payment for goods and services. Saving does not by itself increase the deposits or ‘funds available’ for banks to lend. Indeed, viewing banks simply as intermediaries ignores the fact that, in reality in the modern economy, commercial banks are the creators of deposit money.
That could be taken to mean that savings accounts have only a second order contribution on the amount of money banks loan out, it could also mean that savings accounts reduce the amount of money banks can loan out.
I believe that passage is looking at banking from a systems level, and that savings may not directly impact loans a bank makes, but at a systemic level has an end result of reducing loans. It’s not clear at what level they’re taking about here.
A few days ago there was a post on here about how technical documentation only makes sense if you already have a working mental model of the system you’re working with. I feel like this may be an example of this.
My interpretation of that paragraph is the following:
The money that you don't deposit but you pay a company eventually also ends up in a bank account, of the company, of it's employees, of it's suppliers.
From the banking system point of view it's the same money amount, just split differently between accounts.
It would be interesting if somehow credible (including heterodox) experts could be tagged so you could differentiate them from people with no or limited knowledge larping as experts. Then, anyone could choose to filter by experts or not, to their preference.
I love the format of discussing articles/links, but the amount of people confidently pretending they know what they are talking about that you see when you visit a topic you're an expert on is wild. It must be the same for all topics, I just don't notice when I'm not an expert myself.
Deciding on "expert" seems completely intractable. What if one fellow has spent a decade studying the subject and agrees with 90% of his fellow decade-studiers on 90% of things but the remaining 10% of topics he has views shared by approximately no one. Is he a crank or an expert? What if he's fairly persuasive to layman/politicians and thus is the head of the Bank of England?
Additionally, if a subject matter does not lend itself to experimentation or feedback mechanisms then you can easily have all the experts be consistently wrong for decades due to group think or conformal pressures.
Not to mention that people are often legitimate experts in a field and a leading expert on their own theory, and yet spout palpable nonsense about what supporters of the alternative theory believe to boost the credentials of their own theory.
Heterodox monetary theories are a particularly good example of this...
In this case I think the fellow should be allowed to speak on the topic he studied.
I agree generally in the end you'd have to trust the forum admins to publish some rules and follow them, but that's already the case in any forum. I'm sure some people would like it and some would hate it. It's at least an interesting alternative to the majority of people not knowing what they are talking about but confidently posting as if they do, even if it's just one of the kinds of forums in the mix in the world.
> It would be interesting if somehow credible (including heterodox) experts could be tagged so you could differentiate them from people with no or limited knowledge larping as experts.
These endorsements are highly sought by the larpers and confidently incorrect. Meanwhile the actual experts aren’t as interested in going through the process of getting the mark.
One pattern I’ve seen repeated across many forums over decades of internet use: The confidently incorrect have more time and energy to dedicate to online misinformation than the actual experts. I’ve watched many knowledgeable forum users eventually tire of participation and leave communities because they get overrun by younger, inexperienced people who are still in the phase of life where we think we know everything.
You can't frame rules to capture something like "expertise" which has an infinite scope. I don't know how to express it, but it feels wrong. If you could frame such rules then the subject under consideration might be so trivial, making the whole exercise of tagging pointless.
Or maybe you mean you want people to be able to share credentials like a degree and claim expertise without us having to do the extra work of judging the quality of those qualifications?
Alas, discussion of money and especially money creation online tends to be of uniformly low quality.
(To be fair, even many economists have a fairly nebulous understanding.)
What's somewhat ironic is that the some of the actual 'Austrian' economists like George Selgin have a pretty good handle on the topic, but 'Internet Austrians' invariable are some of the worst misunderstanders.
Agreed, read the article. Economics is a bit like physics in (and only in) that much of it isn't something you can solve through intuition alone, and quite often you'll get the wrong answer.
My question for you is does your intuition think this trend will continue, or is it temporary?
Your naive hypothesis is true. The ratio of cash to non-cash has been decreasing over time, as more digital payments are used. As everyone also knows, cash is a nominal value, it doesn't inflate. Anyone trying to intuit cash demand would be thinking about inflation rates and nominal gdp growth.
I think digital payments would explain a lot of why the correlation sharply reverses in 2020.
The problem is if you want to go further and say; since digital payments will increase during the pandemic, all demand for cash will decrease. I disagree that's an intuitive idea.
Covid changed the money supply and peoples behavior rapidly, saying you can't use intuition under higher uncertainty is a tautology - no field is immune to this. Also going past intuition, the decline in cash as a percentage of total money accelerated, due to the M2 increase being much larger.
Just because covid shocked a few things doesn't mean that economics insn't intuitive or that peoples intuition is wrong. It's like trying to reason about terrorist attack danger using data just after 9/11.
“I have money and people told me I was smart while growing up” is the gateway to being an armchair expert on economics on the internet. I’m not even sure the latter is required, but I’m assuming it’s also a driver for folks here.
My theory is that because a lot of people here come from CS/STEM, they think they can easily understand economics because they have a sufficient background in "hard" fields. You see the same with engineers/mathematicians/physicists who have a disdain for economics.
Or history, or logistics... Regardless so, the comment sectuons are still good enough to learn something new even in fields like supply chain, in which I consider myself to be rather knowledgeable.
Banks create the money supply (~97% of it). A bank "loan" is not a legally a loan at all but a new security that is created and purchased by the bank. Furthermore, the money for that purchase does not come from any other account or "fed reserves" it is literally instantiated in the account. This "credit creation" theory of the money supply is what the article is about and was empirically proven by the work of Richard Werner.
This is a very important level of understanding, and one that is obfuscated by the central banks.
Once understood, it is self-evident that banks are wholly responsible for asset bubbles. Banks, especially large banks, create loans mostly for existing asset purchases, and not for new productive investment. Of course, they do this in part because those assets work as collateral.
What happens when this behavior is aggregated in the whole system is essentially wealthy people jumping over each other to secure an amount of loans approaching infinity to acquire FINITE real estate and SEMI-FINITE stock. You can see how that ratio will repeatedly create asset price inflation which inevitably implodes along with the banks who issued the loans.
The only solution is for the regulator, in this case the central banks, to issue guidance for the banks to create credit for only new productive investments, whether that be new housing, factories, machinery, or firms, because those are not inflationary and increase the size of the GDP pie. If done, the economy would grow at a high clip with low inflation. The current system of credit creation for leveraged buyouts ad infinitum of a slow growing economic pie, has only one logical outcome....
"Once understood, it is self-evident that banks are wholly responsible for asset bubbles. "
No, it's not remotely.
They are no more responsible than the counterparts to the loan.
Every loan a bank makes comes with risks to the bank.
This idea you have about what is a 'productive investment' or not is fairy interventionist.
Who are you to say what is productive, and what is not? When someone buys a building to lease out flats, is that not productive?
If you believe that there is clearly such a thing as 'non productive assets', for example, pure real estate speculation, then it's the fault of those speculators for the speculating, not the banks.
The banks make the loan if the collateral and risk line up - that's what they do.
They are not in the business of deciding what is good for the economy overall, nor should they be.
Finally, that banks create the money is not 'obfuscated' moreover, the amount of leverage in the system is actually controlled by the central bank by setting reserve requirements.
I suggest there's a lot of misunderstanding in our comment.
"When someone buys a building to lease out flats, is that not productive?"
That is simply Joe's $10M building leasing out flats has now become Sally's $12M building leasing out flats. Unless Sally makes productive investments in renovations and so on, there is a net zero gain to GDP but there is asset price inflation as a result of the bank's credit creation. If the bank instead loaned $12M to Sally to build a new identical building, there would be twice as many flats available for renters, the local builders would have work for several months, and the building materials manufacturers would make sales. Joe may have to sell his old building for $9.5M instead, but he may have slightly better rents as the local economy added jobs. And the bank, in the end, should even earn a higher rate of interest from Sally.
"it's the fault of those speculators for the speculating, not the banks."
Both the speculators and banks respond to financial incentives and must work within the laws.
"This idea you have about what is a 'productive investment' or not is fairy interventionist."
Is it any any more interventionist than regulating the bank so that they can't make loans to borrowers who can't pay? Or that they cannot finance new coal plants? Or that they must report all transactions over $600?
It is inevitable that banks will eventually only loan for productive purposes. If we still have a working banking system in 50 years, that is how it will work.
But at least in the US mortgage origination is heavily influenced by the fact that Fannie Mae will buy the mortgages which means the banks don't really care whether new construction is being funded or old buildings are being bought at yet a higher price.
Just looking at the commercial banks alone won't get you the high level picture.
Landlords provide an economic service, in fact for most of them it's not just arbitrage.
"It is inevitable that banks will eventually only loan for productive purposes."
They will always lend to those who can pay them back with the least risk at the highest rate. That's it. Productivity is a second order thing and it will always be second order.
Productivity is indeed a second order thing in the way the banking system works now. But you still have not understood my point.
As long as IT IS LEGAL for banks to create credit for non-productive existing-asset purchases, there is a limit to infinity function applied to the economy that leads to consolidation of all assets with ex nihilo money. The banking system as it exists now leads to a future where one hypothetical private equity firm will have purchased all assets in the economy with bank created money (in practice, it will simply be government/central bank ownership).
I understood your comment, I'm sorry to say that I don't think you have yet absorbed what 'productivity' is or how 'fractional reserve' banking works.
Productivity isn't some adjective in your head that describes 'factories making stuff' vs. 'home buying'.
Fractional Reserve lending benefits the entire economy, lending is done for a variety of reasons and banks don't have the ability to (for the most part) define what is a 'better' kind of economic lending aka 'magic productivity'. They assume that demands on their assets will be borne out by supply and demand.
Yes, centralization of power is an issue, but it's not fractional reserve lending that is the driver of this.
> When someone buys a building to lease out flats, is that not productive?
Why would that be productive? Would the previous owner have left the building empty? If so, why, and isn't that an issue that should be addressed first?
Maybe you have other scenarios in mind where a change of ownership isn't the only thing that happens, but then the productivity is, at least to the first order, due to that other thing, not due to the change of ownership.
When a builder Alex builds an original building (typically using a construction loan), we both agree that’s productive (that construction loans are good things to exist).
Now that the building is standing, Billie wants to buy it. Maybe they want to live in it; maybe they want to rent it out. Is that purchase productive or unproductive? Since it’s the way Alex gets the money to pay off the construction loan (and thus be able to build another building), I think it’s as productive as the construction loan. (Further, no bank would make the original construction loan if there was no prospect for it to be paid off, so builders would have to hold buildings for their economic life if no one else could get loans to buy them.)
Now some more time passes and Charlie wishes to take a loan to buy the building from Billie. Is that productive? Well, it supports Billie’s ability to pay off their loan which supported Alex’s ability to pay off theirs, which is what supported the building existing at all, so…
Yeah, it's pretty clear that the first two loans are productive. The third one not so much, and I think that's the original point about asset inflation creeping in at some point.
Though it's obviously a matter of degree and context. If we're really talking about Billies and Charlies here, chances are that the last loan really is beneficial in terms of how it allows capital allocation to change. Perhaps Billie rented out units and just can't continue with that business anymore for some reason, but Charlie can.
If however we're talking about institutional investors or the very rich who will anyway employ somebody else to do the productive work, then the case for the loan is much weaker.
Ok. I read your previous as suggesting that even Billie’s loan was unproductive or Alex was doing something wrong. (Alex has built a spec property and intends to leave it empty until sale as they’re in the construction business.)
> They are not in the business of deciding what is good for the economy overall, nor should they be.
> This idea you have about what is a 'productive investment' or not is fairy interventionist.
So, banks do not give a damn about what's good for the economy (while having enormous power to drive economies), but you believe interventionsim is bad anyway?
Also, if you don't know what a productive asset is, you just need to read a little bit: it's not hard to know.
The delusion of efficient central planning is maybe one of the most poignant problems in civilization, because free markets are completely counter intuitive.
The 'reading' that needs to be done here is Adam Smith.
It is extremely difficult to know what a 'productive asset' is.
Luxury items: feels like a waste? But this is a way for rich people to spend stupid amounts of money on something that only they value in their heads, but which nobody else values, and money spreads into the hands of working people. And there are a lot of intangibles that come out of that. Rich guys spending billions on race car teams is where a lot of automotive innovation comes from.
Education: feels productive? Show me millions and millions of papers that nobody else reads, that nobody will ever read and where the 'value' in that is. It would be more 'productive' to send them into Nursing in many cases. But who is going to decide which 'Scientists' are worthy and which one's are not? Tough job!
> If you believe that there is clearly such a thing as 'non productive assets', for example, pure real estate speculation, then it's the fault of those speculators for the speculating, not the banks.
It doesnt matter whose responsibility is it if it screws up the entire economy. Breaking everything for everyone is not something that should be allowed regardless of justification.
> Who are you to say what is productive, and what is not? When someone buys a building to lease out flats, is that not productive?
In theory there is a powerful argument here, but in practice I am suspicious because when people attempt to start settling trades in, say, gold the police will soon get involved.
If we're appealing to principles of freedom of opinion, there has to be a really good justification for why we all have to agree with the bank's opinions on who is creditworthy? I think somewhere in the mess I'm being robbed, and everyone being forced to participate in the system is not allaying my suspicions.
It is the old argument against socialism - while there is often not a good argument against the individual parts; it is the communists building the wall to keep people in. Looks kinda suspicious and the people controlling the monetary system see little reason to compromise or allow people to choose the best personal options.
>> socialism - while there is often not a good argument against the individual parts; it is the communists building the wall
Communism and socialism are distinct. The confusion arises because Marx (& Engels) used the terms interchangeably but that was a long time ago and a lot has changed in the interim (as has just about every other discipline in the world) and now these terms refer to distinct ideas.
Within socialism, there’s a few major branches. The one of most interest to western countries would be democratic socialism.
Just to help cement the idea that there’s fundamental differences, here’s some ideas from socialism that might be surprising if you equate socialism with communism:
They’re against redistributive taxes because they’re heavy in administrative overhead and can reduce the desire to work. Instead they prefer the idea that income distributions are maintained fairly. E.g. for one example, the board can award the CEO any amount they desire limited only by the available money to the company - however, the lowest paid employee, regardless of role within the firm, must not earn less than 0.X times the remuneration of the CEO. Where X is set such that the lowest and highest paid workers dont diverge by excessive amounts.
Socialists believe in individual creativity - they strive to provide freedom to be creative to all individuals.
While ultimately they’re against the waste of capitalism (e.g. a sweeping simplification but advertising driven consumption - consumption not based on need but based on wants) socialists do seek to beat production of the capitalist approach where communism has no interest in this.
> regardless of role within the firm, must not earn less than 0.X times the remuneration of the CEO
So penalizing companies in low margin sectors and depriving them of top tier leadership talent while favoring companies in high margin industries seems reasonable to them? e.g. Amazon would have a much lower cap on CEO remuneration mainly because they are in retail and distribution (not implying that warehouse workers are not treated poorly) than Google for instance just because they employee less software engineers etc. proportionally?
You’re trying to add apples and bricks and getting fishing hooks as an answer.
The concept of a CEO is quite radically different when there’s collective ownership of the means of production.
Going back to the grandparents point - it’s often healthy and useful to disagree, often the root of progress but we can’t skip the necessary first step of learning first. An opinion built from a misunderstanding of what’s being discussed is not useful.
> You’re trying to add apples and bricks and getting fishing hooks as an answer.
I'm sorry but I really don't understand what are you trying to say.
> The concept of a CEO is quite radically
Perhaps. My point was that tying minimum pay to CEO salary/remuneration doesn't really make sense. Nothing else.
> An opinion built from a misunderstanding of what’s being discussed is not useful
You mean like my opinion? Why? I mean your comment is very vague and unspecific. You didn't say anything besides that I don't understand what I'm talking about with no explanation or arguments.. (which sort of illustrates my implicit point that idealistic socialists tend to ignore most of the hard issues and mainly focus on gaslighting their 'ideological opponents' instead of offering actual solutions)
> bank "loan" is not a legally a loan at all but a new security that is created and purchased by the bank
No. Bank loans are legally loans in all competent jurisdictions and definitely not securities. (One can securitise loans, e.g. leveraged loans, but that's separate from lending.)
> a very important level of understanding, and one that is obfuscated by the central banks
What? Central banks somewhat consolidate credit creation. Credit has always been the basis of money, even in the commodity era.
Well, the obligation to pay back the loan itself may not be a security, but the receivership of the money that is paid back is usually packaged into a security, so that the entity that give out the loan may not be the one that is eventually paid back for it. Banks package and sell the rights to get the money back.
The bit I don't get is where there can be An accurate (?) measure of the productive capacity of ... anything. So under MMT as Inunderstand it the idea is it's feasible to create money to level of productive capacity - via government purchase of that capacity to build roads and buy policing services etc.
Wartime economies get more "productive" because the government is able to divert resources to its needs (ie stop being a hairdresser and work in the arms factory) - thus increasing the productive capacity.
So, if the government wanted to (peacetime) redirect resources it would have to bid for services at a level that encouraged hairdressers to become munitions workers - and then create money to pay for that.
And they create money through (ok lots of pieces I don't get but I think it's QE ala Richard Werner).
This is typified by a thought exercise: if the government simply replace fiat money with a crypto fiat (ie all money is in bank of englands blockchain ledger ) then suddenly private banks cannot make loans becaus they cannot create money on that blockchain. But the bank of england could ... and this is equivalent of QE except much cleaner and more obvious.
But ... and we eventually get to my point ... if money creation is taken away from banks (regulation, 2008 crash, crypto) and in theory banks are close to the real world and able to judge if loaning money for a factory is a good idea - then how does one judge how much money should be created ?
I am dubious of "we measure inflation" because not just lag but the fairly common view of "prices go up while RPI stays flat"
And since money creation is tied to collateral, and collateral is basically land, land absorbs all money creation in end - which is where we see our land price issues - and essentially means money creation is rich get richer.
If we could break the link between collateral and increasing productive capacity there might be a flowering of equality.
Just in the first half of your comment, it reminds me of Diane Coyle's long standing argument that the GDP, and many productivity indexes, are just very sexist and not for any real reason other than our patriarchal government, and economics specifically.
This is the "if you paid for a housekeeper and nanny it's the same as paying your wife but it does not show up in statistics "
Yes - but it still does not get closer to measuring productivity. I mean we are all pretty much unproductive sitting in a damp field. It's infrastructure, organisation, trade and then low down the lost, skills that matter.
It would if the government printed money to pay women that were doing aforementioned unpaid work. Or paid in social security credits. It is not an unmeasurable thing, just ignored. And these skills and work do matter, it is just subsidized by money from other employment and often not very efficiently, even less fair.
The problem with land is that you can't produce it. Producers arbitrage the interest rate and the profit rate of manufacturing until the difference is almost zero. But you can't produce land so the cost of land goes up instead.
This is like MMT inspired nonsense, lower interest rates from central banks aren't causing lower growth.
We have lower growth, so we have lower interest rates. Increasing rates would decrease the price of assets, as we've just seen from the past 12 months. The purpose of that was to handle the NGDP overshoot from covid.
> The only solution is for the regulator, in this case the central banks, to issue guidance for the banks to create credit for only new productive investments
Just plain stupid. I don't even know what this means or what the policy would look like.
> new housing, factories, machinery, or firms, because those are not inflationary
Obviously wrong. What does inflation mean? An increase in the price level. If you increase demand all else being equal, what happens to the price of something?
By the way - that's the whole point of central banking affecting interest rates. To lower the cost of credit so aggregate demand increases (which is inflationary).
You need to give me a really good thesis on why increasing the cost of credit for """"semi-finite""" assets causes lower growth. The gall to end this with "only one logical outcome" smh.
Edit:
It's like you live in post 2008 fantasy land where aggregate demand is depressed forever. We do not live in this world.
Sorry, that's incorrect. Banks loan out money that is backed by collateral. This money is indeed created. But when the loan is paid back, the money is destroyed. The money tracks the value in the economy, so the inflation is zero.
What the fed does is create money that is backed solely by the fed promising to pay it back in the future. But the money is paid back by issuing more debt! Hence, inflation.
I do agree that the assertion holds true in most cases, but it break down during extreme environments - widespread bank collapses or when the productive ability of the economy is sharply reduced. Loans that cannot be repaid break the equation at some point.
As for government spending, it greatly depends on what it does with the money, and also on taxation. The government promises to pay back money based on future taxation, and if the taxation grow faster than the debt then there won't be any inflation.
Yes, it does imply this, because inflation is a supply & demand thing. More money representing the value of goods and services means each dollar is worth less == inflation.
Borrowing using assets as collateral means the money created matches the value of the collateral.
Theoretically, yes. Technically, however, these people are getting bailed out. That makes inflation. The government still have to pay for the debt, but it seems like it’s not happening anytime soon (the debt ceiling just keep getting higher and higher).
US government debt is around 31.5 Trillion $$ at the moment. If the US government were to pay its debt today, all that liquidity will be removed from the market and this will have enormous deflationary pressures.
In the same way for private individuals, if you can always keep renewing your debt and pricing your assets higher (a bubble would help), you’d create inflation.
> The only solution is for the regulator, in this case the central banks, to issue guidance for the banks to create credit for only new productive investments, whether that be new housing, factories, machinery, or firms, because those are not inflationary and increase the size of the GDP pie.
I struggle to see exactly what you are advocating for. If a family wants to buy a house, they will generally have to take out a loan to cover the upfront cost and pay off the loan over a long period of time. However, the loan is not a "productive investment" (no new assets are being created, only traded) and as such the central bank should regulate normal banks to not be allowed to issue loans for existing houses.
Without the ability to take out a loan for a house, I think we can all see how no normal family without 20-40 years of combined salary payments would be able to afford a house. Is this in line with what you are suggesting, or is it something else?
I'm not trying to be asinine, this is just my interpretation of your suggestion and I am trying to understand what you are suggesting.
They create all of it. God knows where the 97% comes from. We don't have silver coins any more, and even they were tokens for a promise.
It was empirically proven by Keynes by in the 1930s. Werner was way behind the curve.
It's not new knowledge. Reginald McKenna published a book on it in the 1920s.[0]
"What happens when this behavior is aggregated in the whole system is essentially wealthy people jumping over each other to secure an amount of loans approaching infinity "
Wrong. The entire system is a liquidity provision for existing stuff. It is systemically limited by physical collateral and risk limits within banks. If a bank takes on too much risk then it goes bust and the shareholder capital is wiped out.
There is no asset bubble. Artificially intervening in the market for money (which is what interest rate setting is) suppresses the price of assets, which means insufficient are produced.
We're seeing that now as the house builders go into hibernation - another round of stop start in the construction industry - which is one of the reasons why they tend to use subcontractors rather than hire and train employees.
Channelling the stability process via the banks, and thereby making them 'special' means we can't let the standard process of capitalism, bankruptcy and loss of capital, control the risk limits in banks.
Out of curiosity, can you wax lyrical a bit more on construction? I’m about to build a house but I’m considering holding off for 12 months to see what happens with materials prices. Particularly steel. Not too concerned about contractors as I am building myself.
Interest rates go up, money is harder to come by, people are building less since they can't get a loan, contractors find other work until construction picks back up.
I understand this much. But, the world is pretty weird at the moment. It seems more complex now than just moving interest rate levers around to manipulate the economy.
> Once understood, it is self-evident that banks are wholly responsible for asset bubbles. Banks, especially large banks, create loans mostly for existing asset purchases, and not for new productive investment. Of course, they do this in part because those assets work as collateral.
This is probably the best part of what you wrote. Loans are typically given to help purchase a finite resource, helping increase demand for limited supply, having the overall effect of inflation.
The person who gets screwed is the person who saved their money. Now, when they withdraw it to buy an asset, it costs more, because the bank allowed somebody else to buy it.
An IRL example I see is tonnes of young people driving around in brand new cars on finance. Not a single one of them can afford the car they apparently own. Now I, as somebody who saved to buy their car, will need to pay more because the bank increased the demand for these cars so much.
The problem really occurs when high numbers of people start defaulting on these loans, and the things they purchased were highly inflated at the time of purchase. Even if the bank goes to collect the asset, they will find it's not worth what was originally paid, but they are still missing a large amount of money not originally factored into their risk.
Needless to say, 2023/2024 is about to get really bad.
> The person who gets screwed is the person who saved their money. Now, when they withdraw it to buy an asset, it costs more, because the bank allowed somebody else to buy it.
I don’t know needs to hear this, but I should remind everyone that putting your long-term savings in cash is, and always has been, a guaranteed way to lose to inflation. Long-term savers should be using a mix of bonds, CDs, stocks, and money market accounts depending on time horizons and risk tolerance.
Loans create deposits. Balance sheets have to balance.
The loan creates the advance, and the advance is transferred to the credit of another person - which is either directly the payee, or the bank where the payee has the account.
Yeah but now you answered your own question. The deposits represent a claim to debt, they are what is used to settle the debt. So the bank can't just delete the deposits it creates. Debts would become unpayable electronically and then cash must be used to settle all transactions.
The obvious factor is liquidity between banking institutions and liquidity requirements by the central bank.
When withdrawing cash, the bank has to give you central bank money, not commercial bank issued money. If it doesn't have this CB money it has to borrow it from the central bank. So having central bank deposits saves you these costs. The other factor is that transfers between banks are also settled with CB money so you either borrow it from the central bank or get it from customers. Finally, the central bank wants to limit maximum money creation by mandating that a bank must keep a percentage of central bank money.
In short the bank could exclusively operate on money borrowed from the central bank but customer cash deposits are cheaper.
People don't withdraw cash. They transfer it to another bank. When they transfer it to another bank the destination bank becomes a depositor in the source bank.
That's how correspondence banking works.
Central banking is nothing more than an optimisation of that process. There is no need for central bank money to do bank transfers. The liquidity automatically arises.
They don’t which is why they have become hostile to customers in the last 10-20 years. Especially with negative interest rates, the customer is a net negative. Still, people with money are usually the people who generate most business for the bank (either via loans or by buying other products), but it helps to have your account there first.
Gold buggery, now crypto, is an old scam. In truth, sound money has never existed.
And it cannot exist. Value is a subjective measure on reality. Transporting that across time requires work. Expecting to get that transport for free is the perpetual-motion problem of finance.
Gold buggery and crypto may be scams, but the current system where the people with the most money get to issue themselves more money is definitely a scam.
The largest banks in the US collectively decide things like what the federal reserve interest rates will be. They literally decide how much money they can create.
I did and didn't learn anything new. The big banks make up the federal reserve. The federal reserve sets the interest rates. The interest rates determine the amount of money created. The banks decide how much money they can create.
People create money substitutes at will when their central government isn't providing them with a suitable money system. That should give you an intuitive sense that money is always a social relationship.
A bank acts like an internet router but one that transforms the risk relationship. It is not an intermediary that accepts deposits and lends them on. It is an intermediary that acts as a third party to a double sided debt contract where one side owes products and services to the creditor and another side owes money to the debtor.
I don't agree. Instead of being between two people, think of the transaction as being between a person and society. Since society offers a lot of different things, you can sell your pigs to society for a token which you can later use to redeem goods and services later. In this sense money is a debt - or an IOU - from society to the holder.
I don't understand what "sound" is supposed to mean in the first place.
From an engineering perspective a debt based system has a lot more guarantees about what will happen to the economy from a stability perspective. There is still the obvious non determinism of not paying debts on time but it sounds like it could be solved by introducing negative feedback so that the system self stabilized toward a desired state.
But this "sound" money thing, it is supposed to work just by sheer willpower alone.
The central bank only controls the nominal "time value of money". The money holders have a big impact on the real "time value of money". Why haven't you looked there?
Also, the system behaves erratically even when there is no central bank. In fact, the erratic behaviour becomes even more frequent. That would imply that private market participants are a bigger factor in erratic behaviour than central banks.
Except there isn't agreement on the definition of "money" (because it's more complex than it seems on the surface), and there isn't agreement on how it works.
Agreed, but the point of (good) education is not inject revealed truth but train people to think. The different "definitions" are in reality different social contracts so there is nothing to agree on (but one can question which types are more appropriate for what type of economy etc).
Maybe, but this isn't elementary school stuff. It's complicated. To really understand it requires some knowledge of law, business, finance, human psychology, math and history. There is a good reason it's so misunderstood.
Elementary school was mostly a figure of speech. It should really be an ongoing learning track, starting as early as possible. I think in gamified and simplified form one can definitely setup trading games and the like and "get them while they are young". Later adapted to the cognitive abilities of each level. Content is not strictly just about monetery systems: a number of other core concepts from economics and law / contracts in particular.
The bottom line is that modern society is absolutely obsessed with money and knows practically nothing about it. Not a good combination.
I would hazard a guess that it has been intentionally muddied, made confusing, opaque, and left til the last minute for a reason.....interest rates and credit scores.
I don't think its intentional (the overlords?) but it is an unjustifiable gap for any society that considers itself democratic.
Its not just the broad unwashed masses. Essentially outside a tiny number of central banking economists the rest of the world has still "flat Earth" level understanding of how monetary systems work, the role of private banks and, crucially, what alternatives there might be and pros/cons.
Efforts for financial literacy, to the extend they exist, are inadequate and incomplete, essentially manuals for smoothing financial product consumption, not background knowledge on how and why these systems are setup.
Its a fixable gap. Compared to the amount of complexity of material people are exposed to at different school levels, this stuff is really not that difficult.
People have a flat earth understanding of many topics because they are complex and take a lot of work to understand. It would be nice if everyone understood the basics of human biology and health care - but it's incredibly complex. I'm a health care flat earther.... I might always be.
Understanding the legal system would likely be more helpful to most than the financial system. But again, takes lot of work to get there.
Still, for many complex topics the standard curriculum would at least try: there is plenty of biology, chemistry, physics and math on offer. While for many people unfortunately this material flies over their head (a different topic), there is a reasonable fraction that gets comfortable and can contribute to the public debate when needed (think eg vaccines)
The contrast with money is that it affects literaly everybody and everyday, but far fewer people have a grasp of the basics. btw, this became very obvious with the whole cryptocurrency phenomenon which for the first time exposed this gap (it also prompted some central banks to start explaining)
The legal angle is very important too. People have maybe an intuitive understanding of rights and obligations linked to contracts, but understanding legal entities and the role of corporations is very limited.
Ultimately the objective is not to make everybody an expert on these specialized domains, but have enough informed citizens so that they participate in debates, influence policy decisions etc instead of being at the mercy of insiders
Not trying to be argumentative, but, what would the result be? Ie, let's say citizens were informed enough on the banking/financial system - what would be different?
I was literally reading this article just 4 hours before it was submitted to HN. It's a little concerning when my 'unique' search from 2014 isn't even unique.
Some questions I am trying to answer:
1. What are the minimal reserves held by BoE (as %) and what do they currently hold. The same question for all the banks they have a contract with.
2. What is breakdown of the reserves? I.e. Foreign currency, gold, etc.
Thanks, but that only answers part of my question. I also want to see what the ratio is for contracted banks, it's no good if they outsource poor reserve ratios.
That banks can create as much money as they want, without limit?
Or that, even if the bank has a hard limit to how much they can loan out to the amount they hold in reserves in the central bank, what counts as money being created into the wider economy is the act of making a new loan for a customer?
One is way more boring, technical and unsurprising than the other.
Banks can create deposits but creating deposits that aren't backed with reserves entails a risk of insolvency. Thus the amount of deposits banks can create is ultimately limited by the reserves that they hold. The demand for credit also affects the amount of deposits that are created in practice.
Money creation is limited by the productive investments that the banks can find and if it is overly pessimistic it will not lend even to productive investments.
Sound money advocates think that money for loans should be limited by the availability of "prudent savers".
This article is pretty bad. In theory, in aggregate, it's sort of right. But it doesn't track the flow of what happens after that loan is made, it doesn't discuss how credit spends just like money (ie, anyone can create "money" depending on how you define it), and it doesn't discuss capital requirements (either from a reg perspective or a practical risk management perspective) other than a passing note.
People/companies don't borrow money and then leave it in the bank which lent them the money. The money typically gets wired somewhere, and then the bank which "created money" has to actually have the money in their account with the central bank or have assets to give another bank so they'll give some of their central bank reserves. So, their description is true for a hot minute, then lacking. A bank can't just keep lending and lending.
As for limits on lending, both for a specific bank and in aggregate - they don't discuss capital requirements or anything about Basel III... It's mentioned in passing with a reference to another paper, as though it's a minor detail. It isn't. The limit on lending could be the limit they lay out... but in practice the capital requirements mean that the amount of lending is severely constrained, both in quantity and the types of loans and the types of customers. Just look at a balance sheet of an actual bank... (like https://www.sec.gov/ix?doc=/Archives/edgar/data/72971/000007... , page 62). Using the BOE paper you'd assume there are banks levered 1000-1. There are not.
This article tries to gently update the public’s 1930s worldview in teaching us that bank reserve don’t matter anymore and haven’t for a long while and to a lesser degree their respective central banks don’t either.
What matters is balance sheet capacity and how constrained it is or isn’t (Hint: it’s very constrained since 2008).
I wrote a blog post which is pretty good - even if I say so myself (it’s had great feedback) I agree with the fundamentals of this article but I fill in some missing pieces.
https://www.onedb.online/blog/what_is_money
Your article is really interesting. I was wondering what you think about the type of maths Krugman does here [1]?
“ This willingness of foreigners to hold American cash means, in effect, that the world has lent the U.S. a substantial amount of money — maybe on the order of $1 trillion — at zero interest. That’s not a big deal when interest rates are as low as they are now, but in the past it has been worth more — maybe as much as 0.25 percent of G.D.P.”
What is the mechanism where being global reserve currency leads to greater benefits than what Paul calculated?
I've never understood the complaint. Once upon a time the central bank would load up trains and armoured trucks with hard currency that they would send to the bank. The bank would then loan that out for stuff (houses, cars, whatever).
The banks realised pretty early on that they could loan out more than they had in physical cash because people didn't need the currency they needed the bank cheque.
At the end of the day the banks would reconcile with one another and the central banks.
Now, because there's even less need for hard currency in our economies, the central banks can (effectively) email to banks how much that they can have to use for loans.
Accounting rules limit the banks on how much they can loan out.
Nothing has changed from when the hard currency was shipped around countries, except that it's done with a telephone.
And crucially: what used to be called 'free banking' in 19th century America wasn't really 'free banking' like they practiced in more laissez faire places like Scotland or Canada.
> what used to be called 'free banking' in 19th century America wasn't really 'free banking' like they practiced in more laissez faire places like Scotland or Canada
Agreed. Point is, banks created money willy nilly from the start. This was problematic when growth was zero. It was catastrophic when growth was positive. Central banks were invented to regulate this process. Not the other way.
Yes, they used that specific panic as an excuse for starting the Fed. But that's a bit like saying WWI was caused by some specific crisis (like the assassination of Franz Ferdinand). Those happened all the time, but typically didn't lead to a war.
Btw, that panic was long over by the time the Fed was established.
The Fed is primarily a clearing house and lender of last resort, which is needed because having individuals like JP Morgan rescue the financial system when it goes haywire stopped being possible long ago.
> The Fed is primarily a clearing house and lender of last resort, which is needed because having individuals like JP Morgan rescue the financial system when it goes haywire stopped being possible long ago.
The Canadians never had such crises, and neither did they have a central bank. The American banking system was uniquely fragile, thanks to its widespread bans on branch banking: if most or all of your banks only have a single office in a single city, they are bound to be more fragile.
> Count how many times the word "panic" occurs before the Fed was formed (1913) versus after:
Language evolves?
Recessions were deeper after the Fed was established.
In the long run and averaged among many individuals that may be true but in the short run and on an individual basis allowing banks to regulate themselves is a sure fire way for some people to unfairly lose a lot of their accumulated worth.
In practice in eg the 19th century, the very lightly regulated banking of Scotland and Canada was much safer than the US's and England's more heavily regulated banking sector.
In the 20th century it got even worse, with (effectively government backed) deposit insurance becoming popular: it removed most of the incentives customers had for regulating the safety of their banks.
I’d be curious about the history of the creation of central banks - do you have a source for this? My intuition would be that central banks were developed as asset bubbles became so large that the state became involved - and are designed to maximize state control over assets.
> curious about the history of the creation of central banks - do you have a source for this
Your canonical source is Lombard Street [1].
Central banking is a new concept. For most of human history, the game was zero sum. Fixed-supply money worked, which made the complexity of central banking superfluous. It's only in recent centuries that central banks made sense.
However, I'm not sure about your fixed-supply money. Credit and debit have been around for ages, probably longer than money.
So in eg for a long time most of the economy didn't use money. It was around the time when the Fuggers came to power that money came to dominate the economy. But around that time (or not much later), bank notes and fractional reserve banking also already became really popular. And those are decidedly not in fixed supply.
Fugger was a German merchant, I'm not sure where you take the Anglosphere from?
(Btw, I'm not claiming that the Fuggers introduced any of this. Just that their rise coincides with this development in Europe.)
I am specifically replying to your claim of 'For most of human history, the game was zero sum. Fixed-supply money worked, [...]' which has no relation to the Anglosphere.
Lots of things precede central banking - that doesn't change how the system now is still essentially the same as when hard currency was shipped out and about
How much of "core money" lives as physical banknotes and how much is just electronic account balance of accounts that central bank has for private banks?
It's a elastic system, there's no fixed proportions like that and that is by design.
What you call core money are Federal Reserve Notes(FRN), the other component that you call account balance is called Federal Reserve Credit(FRC), together they make up what's called Monetary Base or base money. Think of monetary base as level one fundamental money. FRN is not better or more "real" than FRC in in any way or vice versa, they're both exactly the same and controlled by the Fed. How much cash is in circulation changes depending on what the public wants, if they demand more cash then Fed decreases credit balance of banks and sends hard cash to their branches to satisfy the demand. If the opposite happens where the banks are holding too much cash, they deposit it back to Fed and fed increases their balance and reduces cash in circulation.
In reality though, FRN is miniscule compared to FRC because the vast majority of transactions are settled digitally.
Hard to believe now, but in 1913 one of the primary reasons for the Fed's creation was to provide an "elastic currency", which is provide notes to member banks when their customers demand 'hard cash'. The primary concern back then was if a bank failed to furnish cash when it's depositors asked for it, there was a real risk of a bank run driven by panic, even if the bank was in good shape.
I think it's like 0.2-ish iirc. I remember during covid stimulus fed's assets were about 8Tn, and cash was about 1.5 Tn or something. Also worth noting vast portions of that cash is held outside America and not really 'circulated' like cash in your pocket is.
For UK it's 4% physical banknotes/coins [1]. I would assume it's about the same for all developed countries. Which is why sometimes "ATMs ran out of money" if there is a mismatch between physical demand and supply.
When someone repays their loan, the money does not go into the bank's accounts, it is destroyed. Only interest are effectively added to the bank credit.
The bank does not get infinitely rich, because interests are used to pay bank employees, taxes, shareholders. It can get infinitely rich the same way any other company that makes profits can.
What's stopping anyone from opening a bank is laws and regulations, you have to get a banking license from the government.
It was the comment section of Graeber’s Guardian article written in response to this paper which first lead me to MMT.
EDIT: since I see the discovery of MMT as the most important information I have ever learned, I think it’s a good idea to mention it in this comment section also, in case someone else discovers it as a result and thus learns the most important information they will ever learn.
All economy is fragile because it is based on perception.
The value of a company, of an asset, security, labor, whatever is completely subjective. Yes frequently valuations sound realistic because some people have gambled a ton of money on them (e.g. Tesla is Trillion Dollar Company, or a software engineer’s labor is worth 10 times the labor of a teacher ), but at the end of the day it’s a completely arbitrary valuation.
I am really surprised that the system works at all.
If it were arbitrary, it would be random. It's not, you can build a consistent model that applies to a microbusiness as well as it applies to Tesla.
Should a farm that can feed 100 people be worth more than one that can feed 20? That's not an arbitrary distinction. How about a farm that will feed 100 people in 10 years vs one that feeds 100 now but is eroding?
> I am really surprised that the system works at all.
I hold the same view, but IMO it's proof that we don't actually understand how the system works. Whatever mental model of it we're holding is probably a mere caricature of the actual thing. If it wasn't, we'd understand exactly how and why it works, and we'd be able to take advantage of it.
Just like the philosophers/nerds who are not successful with women and often punt that women are a mystery.
> All economy is fragile because it is based on perception.
There are several ways of looking at it. Rather than "based on perception" I prefer "based on relationships"
The economy is a network of relationships concerning material goods and services (is one definition). One of the mistakes a lot of dry economists make is to forget that. As Maggie (in)famously said: "There is no such thing as society there, are just individuals". Wrong
This was illustrated in the Asian Financial Crises in the late '90s. Thailand and Malaysia were badly effected. Thailand, that has hardly been colonised, did better than Malaysia (according to Joe Stiglitz) because the ancient communal ties still existed and people, hence the economy, could fall back on them when the money system collapsed. In Malaysia those ties had been broken by colonisation and there was more misery.
Gotta remember, money literally has never been objectively anything. It feels like it should be because there are numbers involved, but it's always been wildly subjective.
Many societies used money with inherent value, maybe even most societies. Are gold and silver objectively nothing?
Sure, high trust societies used IOUs successfully, but modern society is not high trust so we have traditionally used things with inherent value for trade, and any IOU-based money needs to be ultra-regulated to function well (which always ends in failure).
Notice how people escaping a collapsing country can always spend their saved gold, but as soon as society has a crisis, hyperinflation destroys fiat currency. That’s because the first case is a currency with inherent value, and the other is an IOU from a society that doesn’t exist anymore.
Thinking all money has never been objectively anything is a mistake.
Gold, especially, only has one valuable property, which is why I believe it was used as a representative "money" - it's an inert metal.
It doesn't degrade, ever, not for thousands of years.
The amount of gold in a blob can change/be diluted, but touchstones revolutionised that - people could easily discover what the percentage of real gold was in a given blob.
We haven't used gold for currency in centuries, because it's not actually very good as a currency. Why not? Because the amount of gold in circulation doesn't reflect the size of the economy. It's value sharply drops when a new discovery is made. And it's value tracks upwards between discoveries because the economies are growing (when economies collapse the local value of the gold drops too)
Your understanding of the classic gold standard is flawed.
First, in the long run the value of gold had been rather stable. That's because the supply of gold isn't random: it takes effort to mine gold. If the gold price is higher than usual for a longer time, people will put more effort into mining. (More formally: even more marginal mines will be worked.) Similar for periods of low gold prices.
Of course, these adjustments happen over long time periods only.
Second, in anything but a very naive gold standard the amount of gold in circulation doesn't need to reflect the size of the economy. At least not directly. Gold was mostly a unit of account, but people seldom exchanged physical gold coins. Which fractional reserve banking, there's no simple relation between the amount of money in an economy and the amount of gold, even if every note is redeemable in gold on demand.
During its free banking era, Scotland was famous for having both a very strong and stable banking sector (one of the reasons it manage to mostly catch up with the England during the Industrial Revolution), and also scarcely any gold in the country.
In Scotland during that time, typical banks held about 2% of their as assets as gold reserves, the rest was invested in the real economy. However, they also had much thicker equity cushions than was is common today: about 30% instead of 8%. That's equivalent to saying that Scottish banks were much less leveraged.
If people constantly use money substitutes (including fractional banking) then how exactly is gold supposed to be a good basis for money? It sounds like there is a fundamental problem with gold and the fact that it doesn't depreciate that has rubbed off onto the pure fiat model and this is why central banks try to create 2% inflation to make money decay.
The value of fiat is subjective until the IRS knocks at your door and then suddenly it has the very objective value of keeping you out of prison. Taxes drive money.
And those network effects can be marketed in the form of a liquidity premium which is paid to the saboteurs of the money system in the form of an artificially high interest rate and now we have invented capitalism...
Taxation is what defines fiscal policy, which is only half the story. You need to take both fiscal and monetary policy into account to get a full picture of the economy as a whole.
If Alice and Bob agree that Alice owes Bob $100, Bob will have show this $100 in his books and pay taxes on it. Does this mean Alice and Bob have created $100?
Are there other players, who can create money that is "more real" than the $100 Alice and Bob created?
When you (you as a figure of speech, only vetted banks can do this) borrow money from the Fed, they do not have to take it out of their own bank account. They're the only participants who are allowed to loan out an unlimited amount of money, while Bob can't lend more than he has, and also can't loan $100 more to someone else until Alice pays him back. So that's money creation in the simple videogame sense.
Banks can make loans that go more or less straight to the Fed's balance sheet without a lot of due diligence, so they have a delegated but slightly more limited money creation power (hypothetically the federal reserve can say no but that's not their policy). From then on every layer gets one additional creditor who is going to look at what's being done and who is on the hook for it being paid back, and it slowly starts working less like money creation and more like loans as most people experience them.
Loaning out money that's deposited in banks, in contrast to loaning money borrowed from the Fed, can have effects that are a lot like money creation and which influence the money supply. Nested debts contribute to the velocity of money by allowing quantities of money to be spent over and over again in quick succession. That makes money more available and has an effect similar to increasing the total supply - and that's why some people call it money creation.
Any bank could loan out an infinite amount of money. If you come to a bank and ask for a loan (and they approve), the bank will create the loan (an asset on their balance sheet) and a matching deposit (a liability on their balance sheet). They could do this an unlimited number of times, as long as the deposits they create are recognized and accepted by the economy. Generally banks will have some reserves (currency, accounts they can draw on, etc) to pay out people when they withdraw from their deposits, at their discretion
The problem comes if people try to withdraw or move their deposits, and the bank does not have reserves to cover the withdrawal. In these cases, the bank would need to sell some of it's assets -or- borrow money from somewhere else. If they can't liquidate assets fast enough, they have a liquidity crunch and will need to borrow or sell the company. If the value of the assets is below the amount being withdrawn, they're insolvent.
The Fed can act as a lender of last resort in liquidity times, but if a bank is insolvent they likely won't be able to post collateral to get loans from the Fed.
>the bank will create the loan (an asset on their balance sheet) and a matching deposit (a liability on their balance sheet). They could do this an unlimited number of times, as long as the deposits they create are recognized and accepted by the economy.
What you're describing is a matter of accounting, I could do it too but it wouldn't create any money. The person who gets the loan usually intends to spend it in the near future, and if that was all it took for there to be a run on the banks then the system would always be in a run. The bank's ability to make good on deposits is limited by the amount of money it has in the accounts it has with other banks and the Fed, and if you trace it back, the federal reserve is the only player who is not limited by that fact.
Well, obviously if you were starting a bank, you'd want a pile of reserves and cash so that this doesn't happen to you on day one. Alternately, you'd offer to hold deposits for some people in exchange for interest, and then lend it out. Your reserve ratio might be close to 100% in this case
But over time, as your reputation got better and you gathered more deposits, you would lend out far more than your reserves. You could do this as much as your risk tolerance allowed.
Historically, the Fed had a minimum reserve requirement of around 10%, meaning banks needed to have only 10% of the value of their deposits in safe, liquid assets. As of 2020, that requirement was dropped to... 0%
If Bob had 100 dollars, he can go about loaning 110$ to other people, making a bet that all of the other persons aren't going to turn up at the same instant and ask for withdrawals. Then 10$ has been created by Bob without central bank involvement.
That wouldn't have any effect on the money supply unless the other persons actually spend the money they borrowed, so most people wouldn't call that money creation.
P.S. usually borrowers spend the money they borrow as quickly as possible because they don't want to make interest payments just so it can sit in an account. That's not where the slack in bank deposits come from. Instead, it comes from people who are saving money.
The real question is: who gets to create an account with the central bank? The answer isn’t everybody, so everybody cannot create money.
If Alice owes $100 to Bobs bank, they can use that to credit a central bank account, so $100 can be created (assuming they are within reserve limits). If Alice only owes Bob, the result is zero sum because neither party can directly interact with the central bank. The result in zero sum.
The central-bank case is only zero sum when the loan is relayed and the $100 is removed from the reserve account. If the loan is never repayed, the economy just added $100 permanently.
But Charles will say he can’t take a debt as payment unless it is done through a party his bank trusts, which will be other banks with fed accounts. Sure it’s possible to setup a parallel banking system, but who would use it?
But Charles will say he can’t take a debt as
payment unless it is done through a party his bank trusts
Charles just takes an "Bob owes you $50" as payment. Just like Bob took an "Alice owes you $100" from Alice.
When you hire a web designer to build a new website for you, they will get to work and boom you owe them $X. The webdesigner will not say "I don't accept your IOU. They will say, "Ok, I'll get to work. You will owe me $X".
And what happens if when designer is done, Bob gives him half of his "Alice owe's you $50"? He would protest, because he wants money, not another IOU. Unless the designers bank accepts IOUs from the bank of Alice, real money is required.
It's the same result: "give me payment in a method my bank trusts. I don't trust Alice, or Bob." This trustworthy bank will need deposits, reserves, or a way to purchase reserves from the Fed when noone else will.
You bring up the question of how the debt is settled. That is out of the scope of my question. The debt could be settled in any number of ways. For example, Alice could give Bob the $100 she owes him. Or not be settled at all. Or Alice could give Bob something else. However Alice and Bob come to the agreement, that Alice does not owe Bob anything anymore, that settles the debt.
My question is if Alice and Bob created money, when they agreed that Alice owes Bob money. If so this money would be destroyed the moment the debt is settled.
The statements "money is debt" and "debt is money" come to mind. So if you create debt, do you create money?
How the debt is settled is essential to the question of whether Alice and Bob can create money. You are theoretically correct when you argue that they can create money via debt, but practically this is not correct.
The difference is that in reality, Alice and Bob will need to complete this transaction through some commercial bank vicariously recognized by the Central bank. These banks are recognized because they are (perceived to be?) solvent and follow the rules; and this recognition confers them the right to generate deposits based on loans. Alice and Bob can do the transaction without these banks, but no other party will recognize the money/deposits that this transaction creates.
Sure, debt creates money. But the only debt that creates USD is debt that is recognized by US authorities. If it's not recognized, your transaction simply created BobBucks(TM). No-one seems to want BobBucks (besides Alice for some reason).
And is zero-sum, thus not creating money. If a bank wants to pay everyone's invoices early they can, but unless that bank is using a fed account to pay the account, the result will be zero sum and no money will be created.
The critical point is whether the final lender in the chain is lending $$ that previously existed, or whether they are lending $$ that popped up in a Fed Account 20 minutes ago. The first case is zero-sum, the second case is only zero sum _when(if?) the loan is repayed_.
> critical point is whether the final lender in the chain is lending $$ that previously existed, or whether they are lending $$ that popped up in a Fed Account 20 minutes ago
The Fed isn't fundamental. Money precedes central banking by millennia.
Banks, even today, can create money willy nilly intraday. (They'll get boned EOD.) But whether the money is created at the Fed is an accounting matter, not monetary.
Because they can't meet their reserve requirement. If JPMorgan creates a trillion dollars of deposits at 11AM, that's fine. If, at the end of the day, they have insufficient reserves, they'll go into receivership. This system is far superior to the old method, which more resembled FTX.
Exactly! Their money-creating behavior is practically restricted by fundamental Reserve Requirements set by the Fed. Thus the Fed _is_ fundamental for money creation. If you want to create new money you either need reserves (set by the Fed), or a Fed account to buy reserves temporarily.
The alternative is loaning out deposits, which is zero-sum, and thus is not 'money creation.' [Edit: I am not suggesting this actually happens - I am suggesting it is the only alternative to lending within Fed limits, which is also the only way to lend without creating money]
You are well into a philosophical discussion here but I want to point out reserve requirements are not the practical limit on lending (even when they are non-zero).
The practical limits are actually the a&l requirements of their regulatory frameworks and the expectations of their equity holders. Right now equity holders get antsy if the deposit to loan ratio gets much above .8. Their regulators will almost certainly take them over if their total a:l ratio stays above 1.
But! There are accredited US banks right now with deposit to loan ratios approaching 2 and there is no federal reserve requirement. Those banks are likely in dire financial straights but for the moment they have created money with no federal reserve requirements.
Thanks for the response! I'm definitely no expert but I think I understand the basics.
Other commenters made a similar point about reserve requirements, I definitely overstated their importance. Interesting to hear about the A&L limits and kind of terrifying to know the limits are self-imposed.
The a&l regulatory requirements are certainly not self imposed. And the limits aren’t just in the form of rough ratios. They outline the classes of debt the balance sheets can have, how they are accounted for, operational procedures around the balance sheet and extensive reporting requirements.
But after 2008 equity holders became more risk adverse so demand tighter deposit to loan ratios. This caused the banks to disengage with lending, one of the things that caused the fed to remove the reserve requirement.
> money-creating behavior is practically restricted by fundamental Reserve Requirements set by the Fed
Fundamentally, for money, the Fed and central banks, generally, are irrelevant. Even practically, for modern U.S. dollars, the reserve requirement is close to irrelevant. It's an archaic tool; most countries simply set it to zero.
> loaning out deposits
Did you read the link you're commenting on? Deposits aren't loaned out. Loans create deposits. This is the difference between middle-school monetarism and real banking.
Yes. That's why it makes more sense to talk about money-ness as a gradual property.
Eg Amazon store credit has some properties of money, but is not universally accepted. Similar for cigarettes in prison (or whatever is popular for that purpose at the moment).
No. The arbitrary definition of 'money' being used by the article is the UK's M4 (https://www.bankofengland.co.uk/statistics/details/further-d...). This basically includes bank accounts along with any cash held outside banks. Since Alice isn't a bank, her IOUs don't count.
Correct. It doesn't have to be a bank or central bank: lot's of other entities create money through lending -- Insurance companies, pawn shops, hedge funds all do it. This it's what's known commonly referred to as shadow banking (https://en.wikipedia.org/wiki/Shadow_banking_system)
Unless the pawn shop has an account at the Fed the result is zero sum: no money is ‘created’ because deposits balance credits. The other commenter explains it well - you need the privilege of a central bank account to create money.
That’s incorrect, as the original article explains. Absent QE (which the BOE explains in this article can result in both creation and destruction of money, depending on complex factors) money is usually created and destroyed without any influence from central authorities. You can read the article for a detailed explanation, but the simple version is that the financial sector as a whole can increase/decrease the entire balance sheet together. Yes assets and liabilities are balanced, but if the size of the balance sheet grows then there are more liabilities and this more money.
I’m not saying the central authorities create money, I’m saying they confer the privilege to do so. Minor players do not truly create money in this way because they are isolated from the financial sector as a whole. They don’t get to play at the big boys table, so their contribution is zero sum.
This starts to get outside the scope of the article, but it is not true that USD central authorities fully or even majority control the money supply through regulation. You can create a bank in UAE or many other jurisdictions with a local license that allows you to deal in USD completely outside the jurisdiction of US enforcement agencies. Even within the jurisdiction of US agencies, there is a shadow banking ecosystem that is poorly understood and highly illegible. Most USD are not created within the jurisdiction of US authorities because USD is an international trade and reserve currency.
Banks are regulated and restricted, but any entity who lends money would technically create money. Individuals are rarely at a size where they can do this, but hedge funds and insurance companies are.
I'm not sure this is correct -- a bank could create money by issuing loans without needing a central bank account. The central bank account provides banks with extra liquidity, but it isn't required to create deposits.
When a bank creates a loan, it also creates a deposit. Say they make a loan to Bob:
Loan -- an asset on the Bank's balance sheet. It's a promise that Bob will repay the bank a certain amount of money, plus interest
Deposit -- a liability on the Bank's balance sheet. It's an account that Bob can draw from at any time, and the Bank will promise to honor or pay the deposit.
These are created at the same time, and balance one another. The bank does not need the Fed to create the deposit at all, the create it out of thin air when issuing the loan.
When Bob comes to withdraw from his account, the Bank will pay him from it's *reserves*. The banks reserves are a pool of assets, cash, and deposits from other customers that the bank holds in excess of it's loans.
The bank can get these from a lot of places, not necessarily the Fed -- they could offer checking accounts to other customers, and hold some fraction of the assets in a customer's checking account in reserve. Banks offer all sorts of incentives to get people to deposit long term(CD, Savings Accounts, e.g.) for this reason, but they almost never have 100% of the money they are loaning out held in reserves. Their entire bet is that only a fraction of the deposits will be withdrawn at any time.
The Fed doesn't need to be present for any of this, but they provide an important stabilizing function, ensure liquidity, and can adjust the money supply through open market operations or banking regulations.
I should have been more specific and said "without deposits or reserves", but I was assuming that without a central bank account, reserves are moot. This is because the central bank determines reserve requirements for the commercial banks, so a bank without a fed account is limited in how much money they can 'create.'
If the Bank doesn't have enough reserves to give Bob money, they simply cannot give him a loan. If a commercial bank doesn't have enough reserves, they can take a temporary loan from the Fed at the cash rate and create money to their hearts (or their risk-teams) content.
If you don't have reserves, borrowing from the Fed would be your very last resort. The first thing you would do is borrow short term from other banks to meet your reserve needs. This is what the Overnight Federal Funds rate represents -- the rate that banks use to lend to other banks short term.
AS a very last resort, you might borrow from the Fed, but banks will generally avoid this as much as possible -- The Fed often charges a higher than market rate to discourage banks from using the discount window too much. Furthermore, borrowing from the Fed's discount window comes with a lot of stigma, since it implies you are unable to get funding from the market.
The most recent Odd Lots podcast episode has a great primer on the Fed Discount Window
> I was assuming that without a central bank account, reserves are moot
There are tons of ways to get reserves without a central bank account -- issuing stock, selling CDs or Bonds, selling liquid assets, etc.
> This is because the central bank determines reserve requirements for the commercial banks, so a bank without a fed account is limited in how much money they can 'create.'
FWIW, in the US the reserve requirements are currently 0%. They're also 0 in a lot of other countries (e.g., Canada)
The Fed also sets eg interest on excess reserves. They are also involved in quite a bit of regulation these days.
They also don't really 'set' the interest rate (apart from the interest on excess reserves). They have an interest rate target, and then buy or sell bonds to reach that target. Instead of setting an interest rate target, they could also have other intermediate targets.
Eg they could directly target the TIPS spread, and buy/sell offsetting pairs of U.S. Treasury bonds and Treasury Inflation-Protected Securities (TIPS) to make the spread go where they wanted it. That might be a better way to implement their inflation target, too.
I this case the bank is making a bet that the loanees, and depositors won't all ask withdrawals at the same point of time. If they all did, it becomes a run on the bank and it will fail.
I suppose Bob could become an illegal shadow bank. The two things that define a bank are 1) regulation and supervision by the Government and 2) access to exchange settlement accounts at the central bank, but you can ignore #1 if you want (it makes it a lot more likely for customers to lose their money either by collapse or fraud) and what banks do through #2 can be done to various extents through other channels.
But they’re still not quite at the same level as money creating banks as you probably can’t, say, pay taxes or receive money from the Government directly with such an entity, and probably have to launder or at least convert assets to be able to credit a normal bank account.
Sorta, but the real question is would you want to? An IOU is an asset representing a future cash flow. From you. It’s an asset for someone else, but a liability for you.
Similarly a bank can create money but it’s a liability for them. They won’t do it for free; you have to pay them to do it with something of value.
Well, it has some properties of money in some degree.
Eg the brewery that supplies beer to your bar will typically not accept your a bar tab as payment. But they will accept USD.
(However, 'factoring' of accounts receivables is a thing. And it's conceivable that under some circumstances third parties might start accepting bar tabs of reputable patrons.
Everybody can multiply money, but only the government (and it’s entities) can make it.
Turns out, multiplying money is way more impactful than making it, much to the confusion of people who learned economics when M2 money supply was an economically relevant metric.
(This phenomenon manifested as the Fed targeting interest rates rather than money supply, starting in the late 20th Century).
Whatever else dollars might be, they are not money. They are debt instruments. They say so right on themselves, in big all-caps letters, right at the top: "FEDERAL RESERVE NOTE". A "note" is a security that obligates repayment of debt.
Money is the thing that would be used to repay the debt.
Whether we like it or not, willing or not, regardless of our judgement of their ability to repay the debt, literally almost individual, institution, and nation in the world is a creditor to the fed.
How money is created depends on how you define 'money':
1) If money is (cash) + (accounts at the central bank), then only the central bank can create it.
2) If money is (cash held outside banks) + (accounts at private banks), then it's created when private banks make loans.
However I think the main point is that saying 'private banks create money' makes it sound like they're getting something for nothing, when in fact this isn't true. Because (2) doesn't include cash held by a bank as 'money' you can't say a bank is losing money when it loans out some cash. But it's still losing an asset worth the same amount.
Personally I think (2) is a silly definition for most discussions.
Bank robber: Give me the money!
Bank teller: We don't have any money, only cash. But it will become money if I give it to you.
This is commercial bank money. (The first is central bank money.) The latter is the only definition appropriate for most discussions. Nobody says Apple has no money because they don't have an account at the Federal Reserve.
People say that Warren Buffet has lots of money, but I wouldn't call shares money either.
EDIT: I'd say that in common parlence bank accounts are called 'money' because they're a claim on money (1) that the bank has to pay out on demand. But if you're calculating an aggregate then you shouldn't include both the claim and the thing it's a claim on.
All loans, even between private individuals increase the number of assets in the system, because as well as the cash that changes hands, a new IOU has been created, and that in itself is an asset. If the IOU is in the form of a number in a bank account, and you include private bank account balances as money (as most people do), then bam! money created.
The magic status of banks just comes from counting private bank IOUs in the form of balances as money and not counting IOUs issued by private individuals as money.