Bear with me on this one. I'm trying to get my head straight on money and banking, by thinking weird thoughts.
Suppose, just suppose, that (for some unknown reason) the only asset that banks liked owning was bicycles. They refused to own any other sort of asset, except, of course, central bank currency and reserves at the central bank. Banks buy bikes, and rent them out for people to ride. Banks are bike intermediaries.
On the asset side of a bank's balance sheet you see the bikes it owns, plus a small amount of currency and reserves at the central bank. On the liability side you see demand deposits, just like regular banks in the real world. And the demand deposits serve as a medium of exchange. The demand deposits are redeemable on demand at par in central bank currency, just like regular banks in the real world. A bank's profits come from the fees it earns from renting out the bikes, minus any interest it pays on demand deposits, minus administrative costs, depreciation on the bikes, losses due to stolen or damaged bikes, etc.
Whenever a bank buys a new bike, it creates a deposit, and the money supply expands. Whenever a bike rider writes a cheque to the bank to pay his monthly bike rental fee, the money supply contracts.
This is a very strange monetary system. It's probably not at all optimal.
Suppose more cyclists decided that it was more convenient to rent a bike than to own their own bike. Banks would respond to the increased demand for bike rentals by buying more bikes and expanding the money supply. A fad for owning your own personalised bike would reduce the demand for bike rentals, reduce the size of the banking industry, and contract the money supply. Changes in the popularity of cycling vs driving would have similar effects on the money supply.
Changes in the supply of newly-produced bikes would affect the money supply too. An increase in the supply of new bikes would reduce the price of bikes. Depending on the elasticity of demand for bike rentals, this could either increase or decrease banks' balance sheets and the money supply.
Mass theft of rental bikes, a rash of mechanical faults in bikes, or a fall in the value of bikes, could cause banks to fail and destroy the monetary system.
If the supply of money is a by-product of the bike rental business, anything that affects the bike rental business will affect the supply of money. The reverse is true as well. Anything that affects the money business will affect the supply of bike rentals, and the demand for bikes by banks.
We don't need to assume that banks are the only firms in the bike rental business. Non banks could also buy bikes and rent them out. But the fact that banks produce money, and also really like owning bikes, would make banks really important players in the bike rental market. Being able to pay for bikes by creating deposits would tend to give them an advantage over other bike rental firms.
A central bank that aimed for monetary stability would need to keep a very close eye on the bike rental market. If there's an increased demand for bike rentals, banks would meet that demand by buying more bikes, and increasing the money supply. But this doesn't mean that people want to hold the increased stock of money that banks created when they bought more bikes. The demand for rental bikes and the demand to hold money are presumably unrelated. The seller of the bike wants to spend the money on something else. That excess supply of money would hot potato around the economy, causing an inflationary boom. Unless the central bank took offsetting action. And a fall in demand for bike rentals would reduce the money supply, as banks stopped buying bikes. And this excess demand for money would cause a deflationary recession, unless the central bank took offsetting action.
The reverse is true as well. Any actions taken by the central bank, whether or not they were designed to offset the effects on the money supply of changes in the bike rental market, would themselves affect the supply of rental bikes. If the central bank decided to increase the supply of money, perhaps to match an increase in the demand for money, this would cause a boom in the bike rental business. By increasing the supply of reserves, making it easier or cheaper for banks to buy a bike and create a deposit, the central bank is indirectly increasing the supply of rental bikes.
If the demand for bike rentals collapsed, or if no trustworthy bike renters could be found, central banks would find it hard to increase the money supply. Central banks could create more reserves, but banks would just leave them on deposit at the central bank. They wouldn't go out and buy bikes and create deposits. They can't find anyone trustworthy willing to pay to borrow those bikes.
The bike rental business would be a macroeconomic phenomenon, not a microeconomic phenomenon. And people would tend to confuse the demand for rental bikes with the demand for money. And this confusion would be perfectly understandable, because an increased demand for rental bikes would cause both an increased quantity of rental bikes supplied and an increased quantity of money supplied.
What I have just imagined is a very weird world. But the real world looks just as weird. Banks don't like having bikes on their balance sheets. Instead, they like having certain types of IOUs on their balance sheets. We can understand why banks prefer certain types of IOUs to bikes. It makes sense for the individual bank. But does it make sense for the monetary system as a whole? Bikes, the yellow metal, IOUs; these are all very weird ways to run a monetary system. What has the optimum quantity of the medium of exchange and medium of account got to do with optimum amounts of bikes, yellow metal, or those certain types of IOUs?
Worthwhile Canadian Initiative 
Nothing, indeed. The question is, why should there be any cost to holding money at all? Why is it that the banks earn seignorage rather than the owners of the money themselves? Money would be a truly neutral veil if only it earned the risk free rate of interest. Want more money? Borrow against good collateral at the risk free rate. The money earns the risk free rate while you hold it. Net cost of holding money? Zero. The current system is a ridiculous relic of an ancient era when secure storage, transport, and payment necessarily involved stagecoaches and six-shooters. You’d think we could get beyond that.
Or alternatively, imagine that our assets could be instantly monetized whenever we needed it. Money would only exist for the mere instant when a transaction takes place between the moment when it is created for the purchaser and when it is subsequently deleted by the seller. The quantity of money would be zero (like the quantity of reserves in the Canadian banking system). And the value would be determined via convertibility to capital assets rather quantity and velocity. We don’t even need money! (Or at least we need zero amount of it)
Nick, My answer to your final question is: “nothing”.
The extent to which central banks are independent of governments / treasuries is very variable in the real world. One could merge the two, with the new merged entity having no assets at all. One of its main aims every year would be to spend an amount of money into the economy (net of tax) that brought full employment without too much inflation. It could issue no bonds (i.e. interest paying debt) at all.
A system of roughly this sort was advocated by Milton Freidman here:
Click to access friedman.pdf
Ralph: “It could issue no bonds (i.e. interest paying debt) at all.”
Sure. Then govt would earn the seignorage profit. Or electronic money could pay interest. That way the owners would earn the interest (seignorage) and the money would become neutral. Seems more efficient to me. There ought to be better justification for the imposition of taxes than somebody’s temporary need for liquidity.
I think the answer is “history”, Nick. Although history might have happened somewhat differently than it did, there are an infinite number of anachronistic ways in which it couldn’t have happened. For example, Terminator or no Terminator, I couldn’t have met my grandmother before my mother was born and become my mother’s father.
I think the point of this post would be even sharper if we clarified that it’s not even IOUs in general, but disproportionately IOUs for housing. In other words, the bike analogy is, I think, closer to reality than it seems to be, when you consider the extent to which banks really are kind of in the house-rental business.
Nick, this post has promise.
“And the demand deposits serve as a medium of exchange. The demand deposits are redeemable on demand at par in central bank currency, just like regular banks in the real world.”
IMO, that sounds like medium of exchange = demand deposits plus currency. If so, good. I believe that assumes the bank is a member of the central bank.
“A bank’s profits come from the fees it earns from renting out the bikes, minus any interest it pays on demand deposits, minus administrative costs, depreciation on the bikes, losses due to stolen or damaged bikes, etc.”
Good.
“Mass theft of rental bikes, a rash of mechanical faults in bikes, or a fall in the value of bikes, could cause banks to fail and destroy the monetary system.”
Right idea.
“We don’t need to assume that banks are the only firms in the bike rental business. Non banks could also buy bikes and rent them out.”
I think there is a difference there (banks and non-banks). JKH might have to help me/us out here.
“What I have just imagined is a very weird world. But the real world looks just as weird. Banks don’t like having bikes on their balance sheets. Instead, they like having certain types of IOUs on their balance sheets. We can understand why banks prefer certain types of IOUs to bikes”
Can we make a slight adjustment here? Banks don’t rent bikes. Most of them want to make loans/create debt (let’s skip investment banks that could speculate in financial assets here). Let’s have people buy the bikes (a durable good) and get loans from the bank with the bike serving as collateral for the loan. I believe that would make the post more realistic. Sound good?
This is the best thing I’ve read in awhile. I do think that it prompts the question: what are the actual IOUs that are favored as collateral for money issued, since in this world it isn’t actually bicycles?
K: I think that money would still be a neutral veil in the long run in this world. The Quantity Theory would still hold true, eventually. Start in equilibrium, double central bank currency and reserves, and the nominal price of bikes, the nominal stock of bank deposits, and everything else would double, leaving the real size of the bike rental market unchanged. But the transmission mechanism for getting to that new equilibrium, with prices slow to adjust, would involve a lot of changes in the market for bikes and bike rentals.
Neo-Biksellians, like Bike Woodford, would model a cashless economy where the central bank simply set bike rental rates, as a percentage of the price of a bike, and controlled Aggregate Demand and the price level that way.
@Nick – You’re really writing this for the opportunity for truly awful puns, aren’t you?
Nick:
Banks will find that once people are holding an optimal amount of money, people will start renting any additonal bikes from non-bank bike rental companies.
“Banks buy bikes, and rent them out for people to ride. Banks are bike intermediaries.
On the asset side of a bank’s balance sheet you see the bikes it owns, plus a small amount of currency and reserves at the central bank. On the liability side you see demand deposits, just like regular banks in the real world. ”
No, I would say that on the liability side you see deposits, bonds, and equity, just like a bank in the real world.
Yes, deposits are money and they are bank liabilities, but that does not mean that all bank liabilities are deposits.
In the U.S. , we have about 7 Trillion in deposits, 5 Trillion in the financial sector bonds, and about 4.5 Trillion in financial sector equities. So less than half of bank liabilities are deposits.
How would you fit that into your bike model? The easiest thing to do is to acknowledge that
1) Banks only own a portion of all bikes (e.g. capital). Most bikes are owned by non-banks. And
2) Households have the option of “buying a bike from a bank” with their deposit, thereby extinguishing the deposit, just as banks have the option of buying a bike (creating a deposit).
3) And what wins out? Banks in fact do not have the option of buying as many bikes as they want to own. Rather, each time a household deposits money into the bank, the bank is allowed to buy one bike. Each time a household withdraws a deposit, the bank is forced to sell a bike. The bank does not turn away the household making a deposit, it always accepts the deposit. At the same time, the bank does not turn away the household making a withdrawal. And as banks pay no interest on deposits (it is free money), they stand in a position to accept all the free money households give them, but they cannot effectively force households to give them more.
In this way, we see that there is no relationship between demand for bike rentals and the stock of money, as most bikes are held by the non-bank sector. Rather, the bank sector is allowed to earn some economic rents by “liquifying” that portion of bikes that households prefer to exchange for deposits — but no more.
So those are the two things that your model is missing — households forcing banks to sell bikes to them, and households owning most of the bikes at all times.
Now I don’t like the bike model, because what really happens when households increase their deposit withdrawals is not that the bank is forced to sell a bond out of its existing stock, but rather that the bank is forced to issue a new bond in its place. Really you need to think in terms of balance sheet expansion and contraction. But if we net it all out — e.g. we are talking about “net” bike positions, then banks do not decide how many bikes they buy or sell. Clearly banks decide how many loans they grant, but the price of talking instead about bikes to acknowledge that this decision does not determine the size of the money supply.
All of the above is heavily dependent on institutions. One can imagine a scenario of financial repression, in which, whenever a household withdrew a deposit, the bank would make up the difference by borrowing from the central bank rather than by selling another bond (e.g borrowing from households). In that case, the above would no longer hold. Again, it all boils down to some form of institutional awareness.
“What has the optimum quantity of the medium of exchange and medium of account got to do with optimum amounts of bikes, yellow metal, or those certain types of IOUs?”
Absolutely nothing. Talking about the “optimal amount of currency” is a bit like talking about the optimal amount of happiness. It is not a lever that you can control, it is an endogenous variable that you can only watch. To make a difference, start talking about institutional structure (e.g. bank regulations) and interest rates.
I’m a little fuzzy on the analogies here. You say,
Whenever a bank buys a new bike, it creates a deposit, and the money supply expands. Whenever a bike rider writes a cheque to the bank to pay his monthly bike rental fee, the money supply contracts.
You envisioning a system in which the the bike producer accepts the demand deposit in exchange for the bike? Correct?
If the central bank decided to increase the supply of money, perhaps to match an increase in the demand for money, this would cause a boom in the bike rental business. By increasing the supply of reserves, making it easier or cheaper for banks to buy a bike and create a deposit, the central bank is indirectly increasing the supply of rental bikes.
I’m not sure I understand what processes you have in mind when you talk about, “increasing the supply of reserves”. But anyway, to the extent that the central bank can increase or decrease the cost to the bank of creating the deposit and acquiring needed additional reserves, they might succeed in impacting the bicycle rental business. But a lot also depends on the prices demanded by the bike manufacturers and the fees the market will bear for bike rentals. A bank can generally create as a big a liability for themselves – the demand deposit – as they want. But the question they always have to answer is whether the asset they are getting in return is worth the liability. If everybody who wants a bike already has one; or if the only people who still want bikes are known bike-crashers whose rental of bikes would only be worth the banks’ while if the renter is willing to pay a risk premium to insure the bike; or if the only people left without bikes are too poor to rent them at any reasonable price; then central bank actions to decrease the cost of reserves are unlikely to have much effect.
A central bank that aimed for monetary stability would need to keep a very close eye on the bike rental market. If there’s an increased demand for bike rentals, banks would meet that demand by buying more bikes, and increasing the money supply. But this doesn’t mean that people want to hold the increased stock of money that banks created when they bought more bikes. The demand for rental bikes and the demand to hold money are presumably unrelated. The seller of the bike wants to spend the money on something else. That excess supply of money would hot potato around the economy, causing an inflationary boom. Unless the central bank took offsetting action.
If the manufacturer didn’t want the money that was delivered to them in exchange for the bike, they wouldn’t have exchanged a bike for that money, no?
From the standpoint of demand for money, the question of whether any agent wants money so that they can exchange it right away, or so that they can hold it now and exchange it later is a subsidiary one. These are just two different factors that determine the demand for money. If for some reason a bike manufacturer absolutely doesn’t want any more money to hold, and if market conditions are such that there are no opportunities for exchanging the amount of money offered by the bank for a bike immediately for something the manufacturer judges to be of sufficient value, then the manufacturer will just not sell a bike for that amount of money. Nobody can force market participants into acquiring a “hot potato”. (That’s not to say that a market participate might not end up with a hot potato on his hands because of a miscalculation.)
Similarly, a central bank can’t force people to acquire more money than they want to acquire.
And a fall in demand for bike rentals would reduce the money supply, as banks stopped buying bikes. And this excess demand for money would cause a deflationary recession, unless the central bank took offsetting action.
Why need it cause a deflationary recession? It looks like you are envisioning a situation in which the contraction in the amount of money in circulation corresponds to a drop in the demand for some product – and thus a drop in the production and distribution of that product. Isn’t this the way we would like the monetary system to work? The rise and fall in the supply of money adjusts automatically to the rise and fall in the production and distribution of the things money is exchanged for.
“By increasing the supply of [central bank] reserves, making it easier or cheaper for banks to buy a bike and create a deposit, the central bank is indirectly increasing the supply of rental bikes.”
How about what happens when the central bank lowers the overnight rate (let’s assume the bank’s spread widens making it more profitable) but the amount of central bank reserves doesn’t change?
@Sandwichman: “I think the answer is “history” ”
Could be right.
Or maybe it is all about religion and the legal system.
If so Lenny Bruce( the law ) might have been on to something.
rsj said: “No, I would say that on the liability side you see deposits, bonds, and equity, just like a bank in the real world.”
It appears Nick’s post does not include any capital for losses.
Would what is supposed to be risk-free CD’s be considered bonds?
Dan Kervick said: “You envisioning a system in which the the bike producer accepts the demand deposit in exchange for the bike? Correct?”
That is what it sounds like to me.
And, “If everybody who wants a bike already has one; or if the only people who still want bikes are known bike-crashers whose rental of bikes would only be worth the banks’ while if the renter is willing to pay a risk premium to insure the bike; or if the only people left without bikes are too poor to rent them at any reasonable price; then central bank actions to decrease the cost of reserves are unlikely to have much effect.”
What if the central bank believes the demand for bikes is unlimited when in reality it is not?
Nick’s part said: “But this doesn’t mean that people want to hold the increased stock of money that banks created when they bought more bikes. The demand for rental bikes and the demand to hold money are presumably unrelated. The seller of the bike wants to spend the money on something else.”
What if the seller of bikes is a corporation whose sole function is to “make as much money as possible” to jack up the stock price?
And, “Why need it cause a deflationary recession? It looks like you are envisioning a situation in which the contraction in the amount of money in circulation corresponds to a drop in the demand for some product – and thus a drop in the production and distribution of that product. Isn’t this the way we would like the monetary system to work? The rise and fall in the supply of money adjusts automatically to the rise and fall in the production and distribution of the things money is exchanged for.”
because if bike production and distribution goes down that can lead to increased unemployment and then debt defaults and then decreased bike production ; rinse, lather, and repeat
What if the central bank believes the demand for bikes is unlimited when in reality it is not?
Well, it seems to me that what happens then is mostly a lot of nothing, i.e., either the CB keeps pushing on a string to no avail or they stop pushing on the string and have to put up with monetarists saying “Push harder on the string!”
I have argued in the past that the only thing the CB can directly target are interest rates, and that the market for loans is significantly determined by the effective demand for credit. The latter depends on what potential borrowers can offer, and that is driven mainly by factors external to central bank decisions. Once the CB has done what it can do about interest rates, they’re pretty much done. I know the monetarists disagree with me.
“because if bike production and distribution goes down that can lead to increased unemployment and then debt defaults and then decreased bike production ; rinse, lather, and repeat.”
I agree that a drop in production and employment caused by a demand shock can cause secondary drops in demand and a recessionary cycle. But Nick seems to be worried about a specific kind of recession caused by a drop in the money supply.
Great post. A few of thoughts for what they’re worth:
(1) The loan market is the “rental” market for the medium of exchange. Of course, banks are in the business of issuing instruments (deposits) that function as a medium of exchange. So there is a natural connection between bank’ liabilities (deposits) and banks’ asset portfolios (loans), isn’t there? Banks are inherently manufacturers of the medium of exchange — but they aren’t inherently bike manufacturers.
(2) You say “the demand for rental bikes and the demand to hold money are presumably unrelated.” I’m curious whether you think the same is true for the demand for loanable funds. Isn’t the demand for loanable funds a manifestation of the demand for the medium of exchange (since the loan market is just the rental market for the medium of exchange)?
(3) In fact, the “demand for money” and the “demand for loanable funds” are typically drawn in the same space, with the interest rate on the y-axis. And they have the same shape: both curves are downward sloping. Are they distinct concepts in your view?
(4) Holders of deposits want them to be very unlikely to default. They therefore want banks to own low-volatility, relatively diversified portfolios. Assuming no deposit insurance, depositors presumably wouldn’t want their bank to put all its assets in tech stocks, or bikes.
(5) Generally speaking, portfolios of credit assets are less volatile than portfolios of equities or bikes. (Directly conducting a bike rental business is just “equity” ownership of that business.) Bank portfolios consist almost entirely of senior claims on other economic agents — not residual ownership. No doubt credit investing is risky, but it’s safer than bikes. A 5% impairment of a bank’s asset portfolio is catastrophic.
(6) There is a well-known banking case, M&M Leasing v. Seattle First National Bank (9th Cir. 1977), that addresses the question whether national banks may own fleets of cars and rent them out. The court concluded that this activity would be impermissible, because the bank would own the residual risk of the cars. On the other hand, the court said banks may own cars and lease them to customers so long as the residual risk of the cars is borne by the customers. In that case the lease is functionally equivalent to a loan secured by the car. (Coincidentally, I just taught this case in my banking law class today. Rather weird to stumble across your post after class.)
(7) In the United States, legal restrictions on bank portfolios go back to the earliest days of the republic. Banks have always been limited almost exclusively to loans and other credit assets. This restriction has a long and distinguished pedigree and, it seems to me, a sound economic logic. With the advent of deposit insurance in 1933, this rationale became even stronger.
(8) So I suppose I would question whether the real world looks “just as weird” as the bike world. In theory, the monetary authority sets a ceiling on the outstanding quantity of the medium of exchange (by controlling its own outstanding liabilities and setting reserve requirements for banks). The universe of credit assets is far larger than the targeted money supply. So banks typically have plenty of places to invest the proceeds of the deposits they issue. Fractional reserve banks manufacture and distribute the medium of exchange; they distribute it to borrowers, who by definition want to rent it. This seems efficient, at least compared to the realistic alternatives.
(9) You write “If the demand for bike rentals collapsed, or if no trustworthy bike renters could be found, central banks would find it hard to increase the money supply.” Substitute “loanable funds” and “borrowers” for “bike rentals” and “bike renters,” and isn’t that the liquidity trap?
Sorry for the long comment … it’s a great post.
Ralph: central banks need some bonds, in case they need to reduce the money supply. It gives them something to sell.
Sandwichman: “I think the answer is “history”, Nick.”
I have a hunch that “accident of history” might be part of the answer. But don’t really know. I’m still feeling my way around these questions.
TMF: “Can we make a slight adjustment here? Banks don’t rent bikes. Most of them want to make loans/create debt (let’s skip investment banks that could speculate in financial assets here). Let’s have people buy the bikes (a durable good) and get loans from the bank with the bike serving as collateral for the loan. I believe that would make the post more realistic. Sound good?”
Nope. I don’t want this post to be “realistic”. I want it to be unrealistic, so we can see how much of current “reality”, has to be that way, and how much doesn’t. Plus, this “unrealistic” scenario may let us see what is essential about the current reality, and understand it better.
Will: thanks!
“I do think that it prompts the question: what are the actual IOUs that are favored as collateral for money issued, since in this world it isn’t actually bicycles?”
Yes. Short-term relatively safe and liquid nominal assets?
david: I just couldn’t resist the puns!
Mike: “Banks will find that once people are holding an optimal amount of money, people will start renting any additonal bikes from non-bank bike rental companies.”
Why? How would the Law of Reflux work in this case?
rsj: “No, I would say that on the liability side you see deposits, bonds, and equity, just like a bank in the real world.
Yes, deposits are money and they are bank liabilities, but that does not mean that all bank liabilities are deposits.”
Banks are a subset of Financial Intermediaries, which have liabilities that serve as media of exchange. OK, so the things we call “banks” are only half banks.
“So those are the two things that your model is missing — households forcing banks to sell bikes to them, and households owning most of the bikes at all times.”
I am happy with households owning many of the bikes directly, just as not all IOUs are held by banks. I don’t see how households can force banks to sell them bikes. If one banks loses deposits to a second bank, the first bank will presumably sell some of its bikes to the second bank too, and the second bank will now collect the fees on the bike rental agreement.
Banks will presumably compete with each other for shares of the total deposits and bike rental market.
“Banks buy bikes, and rent them out for people to ride.”
You can also think about it as banks buying up bike rental contracts from the private sector. I find that a little less weird and easier to understand.
I like to think in terms of the supply of liquidity, not of money, so the issues in this post don’t worry me as much. Good post.
Dan: “You envisioning a system in which the the bike producer accepts the demand deposit in exchange for the bike? Correct?”
Yes, that’s the simplest story. Or, the bank could write a cheque which the bike seller deposits at his own bank. Same result: total bank deposits expand.
“If everybody who wants a bike already has one; or if the only people who still want bikes are known bike-crashers whose rental of bikes would only be worth the banks’ while if the renter is willing to pay a risk premium to insure the bike; or if the only people left without bikes are too poor to rent them at any reasonable price; then central bank actions to decrease the cost of reserves are unlikely to have much effect.”
Yep. Sometimes that may happen. Like right now, in the US?
“Similarly, a central bank can’t force people to acquire more money than they want to acquire.”
A bank (central or otherwise) can’t force any individual to acquire money. It has to persuade us to acquire money, by offering us a good price for what we are selling, just as any buyer does. But a bank can force us, in aggregate, to hold more money than we wish to hold. Each individual accepts the money because he believes, correctly, that he can always pass it on to a second individual, if he doesn’t want to hold it. And the second accepts the money, even if he doesn’t want to hold it, because he in turn believes he can pass it on, etc. Money is not like other assets, like refrigerators. It’s the medium of exchange.
“Why need it cause a deflationary recession? It looks like you are envisioning a situation in which the contraction in the amount of money in circulation corresponds to a drop in the demand for some product – and thus a drop in the production and distribution of that product. Isn’t this the way we would like the monetary system to work? The rise and fall in the supply of money adjusts automatically to the rise and fall in the production and distribution of the things money is exchanged for.”
But if causation runs the other way, from an excess demand for money to a reduction in demand for goods and a reduction in production, that is not how we want the monetary system to work.
“I am happy with households owning many of the bikes directly, just as not all IOUs are held by banks. I don’t see how households can force banks to sell them bikes. ”
Because you are thinking in terms of bikes, rather than IOUs. IOUs are really net bikes — i.e. bonds held – bonds owed.
The crux of the matter is understanding the institutions.
Individual banks cannot fund themselves by perpetually rolling over overnight loans from other banks, nor can they fund themselves by tapping the discount window. With other institutional structures they could do this — that would be financial repression.
But in a non-repressive regulatory regime, if any individual bank attempts to continuously borrow from other banks or from the government, it will incur regulatory scrutiny and its credit rating with the other banks will decline.
This, together with competition among banks, forces the banking system as a whole to tap the non-financial/non-government markets for funding, and this creates the release valve by which households can reduce their aggregate deposit holdings without needing to bid up the price of goods. Simply put, it is the fact that households can withdraw their deposits and purchase a bond sold by another bank. It is the exact same mechanism that keeps prices low in a competitive market, even though households will buy the good from someone — nevertheless, by threatening to not buy from you, you are forced to sell households what they want.
It’s a slightly subtle point, but it’s not very subtle. People on this blog have been shouting this to you for the last few years. Through the normal course of withdrawing money and purchasing bonds, each individual bank will need to also sell bonds to obtain funding for itself, even though in aggregate, if all banks collaborated, they would not need to sell any bonds.
I myself have made this point a dozen times, and each time you engage with this proposal, the reaction is as if I was saying something completely out of the blue.
M Ricks: Thanks!
1. Yes, there are good reasons why individual banks prefer owning OIUs for money than bikes. The price of bikes might fluctuate too much. Short term safe IOUs for money don’t fluctuate in price much, so the individual bank is safer. It’s a good system for the individual bank. But is it a good monetary system for the economy as a whole?
2. “You say “the demand for rental bikes and the demand to hold money are presumably unrelated.” I’m curious whether you think the same is true for the demand for loanable funds.”
Yes, I do think that. We can imagine a world in which there is zero (stock) demand for loans in equilibrium (like a simple representative agent model), but in which there is a large (stock) demand for the medium of exchange. Or vice versa. That was one of the main (subliminal, metaphoric) points of this post.
“(3) In fact, the “demand for money” and the “demand for loanable funds” are typically drawn in the same space, with the interest rate on the y-axis. And they have the same shape: both curves are downward sloping. Are they distinct concepts in your view?”
Absolutely distinct. No relation between them. One has a flow and the other a stock on the horizontal axis.
4. Not obviously so. I could imagine being willing to have the value of my chequable demand deposit fluctuate, if I could get a higher interest rate.
When I rent a bike, I normally hold the bike. When I get a loan, and “rent” the money, I do not normally hold that money. I borrowed it to spend it. Because money is the medium of exchange. It’s different from bikes, and refrigerators.
JP: Thanks!
“You can also think about it as banks buying up bike rental contracts from the private sector. I find that a little less weird and easier to understand.”
Yep. I think that works well.
Thanks for the reply, this is really interesting.
On (4), your business model of “fluctuating” demandable instruments exists — it’s just an open-end mutual fund. But we don’t usually think of these instruments as part of the money supply (except for money market fund “shares,” which are precisely designed not to fluctuate — they have a fixed NAV). So there seems to be something important about stable-price assets that makes them money-like. (“Price” just means value relative to the medium of exchange …) So this really goes to the question of what makes deposits suitable as “money” in the first place. It seems like their stable-price characteristic is essential to their money-ness.
You say “We can imagine a world in which there is zero (stock) demand for loans in equilibrium (like a simple representative agent model), but in which there is a large (stock) demand for the medium of exchange. Or vice versa.” To be precise about terminology, I’m suggesting that demand for loanable funds, not demand for “loans,” is closely related to money demand.
It seems odd to me that money demand could be extremely high, but demand to rent money (i.e., borrow) could be non-existent, no matter how low the rental price. It would be like a neighborhood in which houses cost $10 million, but nobody will pay a penny to rent one of those houses for 5 years. I suppose it’s possible but it seems unlikely. Rental prices usually bear some relationship to the values of the underlying assets.
Your emphatic answer that the demand for money is “absolutely distinct” from the demand for loanable funds is fascinating to me. In other markets, we depict the y-axis as “price” (value in terms of money). But this obviously won’t work for money: the value of a unit of money in terms of units of money is always 1. So we need something else on the y-axis. We use the rental price of money (the interest rate), also known as the cost of a loan. But you inform me that the money-demand curve, drawn in this space, has nothing to do with the demand for loanable funds. One of us is definitely thinking about this the wrong way (probably me). I need to think about the stock/flow distinction, which does seem important.
Thanks again, very thought-provoking. (Don’t feel obligated to respond BTW …)
M. Ricks:
4. Suppose we can write cheques on our mutual fund accounts. The cheque could be for a specific number of dollars, but the balance in our account could fluctuate with the stock market. If each cheque clears quickly, relative to the daily fluctuations in stock prices, this is not much of a problem, as long as we don’t try to spend all our balance at once.
Imagine a simple world with no investment, but there are consumption loans. Suppose everyone had exactly the same demand curve for borrowing (or supply curve of saving). In equilibrium the rate of interest would adjust until nobody wanted to borrow, and nobody wanted to lend. The rate of interest adjusts until desired saving = 0. But people still need to hold stocks of money as a medium of exchange.
Thanks. On (4), I think you shifted the ground on me a bit. I started by saying holders of deposit instruments (the medium of exchange) want them to be very unlikely to default. You said not obviously so, because you could imagine having the value of your checkable deposit fluctuate. I said such price-fluctuating instruments wouldn’t be well-suited to serve as a medium of exchange; they would be like open-end mutual fund shares. You said no problem, you could just write checks for a lower amount than the balance in your “fluctuating deposit.”
But then the instrument you were holding (the mutual fund share) isn’t really the medium of exchange. When you buy something with one of these checks, the seller in the transaction doesn’t receive what you were holding, i.e., an interest in a mutual fund share. The seller gets something else.
In fact some open-end mutual funds do offer limited check-writing privileges just as you describe, but the “check” is actually drawn on an affiliated bank. The seller in the transaction gets a credit to his deposit account, not a fluctuating-value mutual fund share. My point is that the “money-ness” of the instruments that actually serve as a medium of exchange seems closely related to their stable-price attribute. In a sense this is tautological, as “price” just means value relative to the medium of exchange.
(Do you think equity mutual funds that offer check-writing privileges should be included in monetary aggregates? I think that would be a novel position.)
Nick’s post said: “Nope. I don’t want this post to be “realistic”. I want it to be unrealistic, so we can see how much of current “reality”, has to be that way, and how much doesn’t. Plus, this “unrealistic” scenario may let us see what is essential about the current reality, and understand it better.”
OK, but then I basically agree with M. Ricks where:
“(1) The loan market is the “rental” market for the medium of exchange. Of course, banks are in the business of issuing instruments (deposits) that function as a medium of exchange. So there is a natural connection between bank’ liabilities (deposits) and banks’ asset portfolios (loans), isn’t there? Banks are inherently manufacturers of the medium of exchange — but they aren’t inherently bike manufacturers.”
M. Ricks said: “The universe of credit assets is far larger than the targeted money supply.”
Can you expand on that one? Thanks!
Dan Kervick’s post said: ‘”because if bike production and distribution goes down that can lead to increased unemployment and then debt defaults and then decreased bike production ; rinse, lather, and repeat.”
I agree that a drop in production and employment caused by a demand shock can cause secondary drops in demand and a recessionary cycle. But Nick seems to be worried about a specific kind of recession caused by a drop in the money supply.’
IMO, it is the amount of medium of exchange stops growing/falls then drop in demand then drop in production then unemployment then debt defaults (amount of medium of exchange falls).
Rinse, lather, repeat.
M. Ricks said: “It seems odd to me that money demand could be extremely high, but demand to rent money (i.e., borrow) could be non-existent, no matter how low the rental price.”
Is there more than one type of “money”?
M. Ricks. Fair response. I’m not sure.
M. Ricks: I don’t think it is useful to think of the interest rate as the rental price of money.
1. In a world without money, there would still be interest rate(s). We would measure those interest rates in goods. Sure, in a world of monetary exchange, we measure all prices in money, and we measure interest rates in money, and we buy IOUs in money and when they come due we sell them back for money. But then we buy and sell everything for money, so that doesn’t really count for anything.
2. Sure, the rate of interest is one opportunity cost of holding money (if money itself pays no interest). But the rate of interest is also one opportunity cost of holding your wealth in paintings, houses, furniture, etc. You could equally well say that the opportunity cost of holding money is the foregone pleasure of having a painting on your wall.
Nick, I don’t agree with your claim that “central banks need some bonds, in case they need to reduce the money supply.” Strikes me that if a central bank is “bond-less” and wants to reduce the money supply, it can just announce it is willing to borrow at above the going rate of interest.
Ralph: OK. In a sense, what your central bank is doing is issuing its own bonds. And in many (most) cases what you are suggesting would work. There are (rare?) cases where it wouldn’t though. Take an extreme example, just to show what I mean. The central bank wants to halve the money supply, and then keep it constant forever. In that case, it could never pay the interest on its bonds. It would have to borrow more, just to pay the interest, and keep on doing that forever. Like Mr Ponzi.
But in most normal cases, where the central bank wants to reduce the money supply only temporarily, what you are saying would work fine.
The way I think about it is this: the biggest asset held by any central bank is not on its balance sheet. It’s the future revenue it gets from printing money in future. Your central bank is borrowing against that future revenue.
Nick: But if causation runs the other way, from an excess demand for money to a reduction in demand for goods and a reduction in production, that is not how we want the monetary system to work.
But you seemed worried about the case where “a fall in demand for bike rentals would reduce the money supply, as banks stopped buying bikes,” and argued that “this excess demand for money would cause a deflationary recession, unless the central bank took offsetting action.”
And that’s what I don’t understand. If the contraction of the money supply occurred in response to diminished activity in the bike rental market; that is, if a decline in the supply of the medium of exchange occurred in response to a diminished pace of exchanges, then why wouldn’t that be a healthy phenomenon? And how else could any “monetary authority” measure gauge the need for money supplies?
I agree that in the hypothetical scenario of the bike-renting banks, it would indeed be a confusion to interpret the demand for the use of bicycles as manifested by the willingness to exchange money for their temporary use as a reflection of the demand for money.
But given that banks are not in the bike-lending business but in the money-lending business, then it is hardly a confusion to interpret the demand for immediate monetary credit as manifested by the willingness to exchange promises of future monetary payments for that immediate credit as a reflection of the demand for money.
The exchange of monetary assets in the market organized around banks and the like only reflects differences in the time structure preferences for monetary income. One person would prefer shifting X dollars of his current income to X + Y dollars of income at some later time t. Another person would prefer shifting X + Y dollars of future monetary income at time t to X dollars of current income. So the two parties make an exchange. On both sides, the behavior is explained by the demand for monetary income, but differences in the preferred temporal shape of that income.
rsj: Let me try this:
Assume that banks like to hold a certain type of asset X as backing for their deposits D. (In my post, X = rental bikes).
Start in equilibrium, where actual D equals desired D by households, and D and X held by banks are at the levels desired by banks.
Suddenly households want to halve their stock of D. The question we must now ask is: what do they want to hold instead?
Suppose the answer is X. Households want to halve their stock of D and replace it with X. In that case, it seems to me the Law of Reflux would work. Households would bid up the price of any X that wasn’t held by banks, and in response to that price increase in X (i.e. lower yields on X) banks would respond by selling their stocks of X to households, and D would decline.
But suppose instead the answer is Z. Households want to halve their stock of D and replace it with some asset not held by banks. They bid up the price of Z, but there is no effect on banks. Banks do not contract the stock of D held by households. The Law of Reflux does not work. As the price of Z rises, substitution effects bid up the price of everything — assets and consumption goods. The excess supply of D causes general inflation. (Yes, the price of X will be bid up too, but since the prices of all goods will be bid up, the yield on X may not fall.)
We can also imagine cases where the answer is “everything”.
The simplest macro model has only two assets: money and “bonds”, and “bonds” are held by banks as backing for D. By default then, we are looking at cases where households want to hold less D and more X.
Two things are missing from my story here:
1. Households own the banks.
2. In the simplest case, where there are no banks and a fixed stock of outside money, an excess supply of money gets resolved by inflation which reduces the real money stock without changing the real stock of any other asset (except nominal bonds, of course).
Dan: bike rentals are a very small part of total economic activity. Suppose the bike rental market disappeared, then bank money would disappear too. But we would still want to hold and use money for everything else.
Similarly, the market in bank loans is a small part of total economic activity.
“But given that banks are not in the bike-lending business but in the money-lending business, then it is hardly a confusion to interpret the demand for immediate monetary credit as manifested by the willingness to exchange promises of future monetary payments for that immediate credit as a reflection of the demand for money.”
Yes it is a confusion, and it is exactly the same confusion as confusing the demand for rental bikes with the demand for money.
The “demand for money” means the stock of media of exchange that people want to hold on average over the month. For example, the demand for bank loans could expand, but nobody wants to hold that extra money. They want to spend it, and hold exactly the same stock as they did before. Or, the demand for all loans could disappear completely, and everyone pays off their loans, but people would still want to use money to buy and sell goods, and want to hold the same stock of money as before.
Dan: “One person would prefer shifting X dollars of his current income to X + Y dollars of income at some later time t. Another person would prefer shifting X + Y dollars of future monetary income at time t to X dollars of current income. So the two parties make an exchange. On both sides, the behavior is explained by the demand for monetary income, but differences in the preferred temporal shape of that income.”
Now suppose that everyone had exactly the same temporal preferences, so the market in loans disappears completely. But people still want to hold money, because it makes the shopping easier. The loan market dries up, but the (stock) demand for money does not dry up. But if banks create money by making loans, the (stock) supply of money dries up. So we get an excess (stock) demand for money, and the economy goes into deflationary recession.
In short, money is not credit; and the demand for money is not the same as the demand for credit.
Short version of this post: the demand for bank money is not the same as the demand for bank credit. Yet we have a monetary system in which the supply of bank money is determined by the supply and demand for bank credit.
You’re (consciously?) channeling Keynes in your response to my last comment. General Theory, chapter 17 (“The Essential Properties of Interest and Money”). He asks “whether it is only money which has a rate of interest.” He notes that “for every durable commodity we have a rate of interest in terms of itself: a wheat-rate of interest, a copper-rate of interest,” etc. Keynes calls these “own-rates” of interest. He asks: “Wherein lies the peculiarity of the money-rate of interest which gives it predominating practical importance?”
He has an answer, which I won’t bore your readers with here. But he does famously define the interest rate as the “reward for parting with [money] for a specified period of time.” I would say the rental price of my house is the reward for parting with my house for a specified period of time. From the renter’s perspective, my reward is her cost. Why isn’t the interest rate just the cost of renting purchasing power for a specified period of time?
You are certainly right that in a world without money there would still be own-rates of interest. But Keynes’s point, which seems convincing, is that in a world with money there does seem to be something special about the money-rate of interest. So special, in fact, that the term “interest” always refers to the money-rate of interest.
So I don’t yet see what’s wrong with thinking about the capital market as the rental market for purchasing power. I don’t think the existence of own-rates of interest calls this view into question, if that’s what you’re suggesting. Issuers into the capital market all have one thing in common: they want to transact, and they are renting purchasing power for a period of time, in order to transact now(ish).
On your point about the interest rate being just one opportunity cost of holding money, I couldn’t agree more. And yet when we draw the demand curve for money, we put the interest rate on the y-axis, not paintings or whatever. (Why is that, if the interest rate is just one opportunity cost of holding money? It must be a particularly important or telling one…)
I confess I’m still struggling to see how this money-demand curve (with the interest rate, not paintings, on the y-axis) could have nothing whatsoever to do with the demand for loanable funds. I honestly don’t know how else to think about what “interest rate” means. You’ve tried to explain this but I’m not yet getting it; sorry if I’m being obtuse. You keep saying the demand for money is not the same thing as the demand for loanable funds (credit). It seems to me that the demand for loanable funds is a particularly clear manifestation of the demand for money. Why rent purchasing power if you have no use for purchasing power? But you think these two concepts have “no relation” to each other. As I said, one of us is completely wrong, probably me.
Apologies for rambling — it’s your blog, not mine! But these are really fundamental questions. I teach banking law at Harvard, and I’m going to use your blog post in class today. The topic this week happens to be bank portfolio restrictions. Super post.
Nick, You make a good point about “the biggest asset held by any central bank is not on its balance sheet . . ”. Likewise, a very big asset “owned” by the Treasury / Central Bank machine, which does not appear on their combined balance sheet, is the right to extort money, i.e. tax, from the population.
Nick
Completely unrelated, but I came across your post on the upward sloping IS curve via Tyler’s post today and comments were closed there. I couldn’t help thinking – is it fair to characterize the Wicksellian view of IS as ‘lower interest rates cause higher GDP’, and the ‘Market Monetarist’ view of IS as ‘higher GDP causes higher interest rates’?
If that’s the case, aren’t the two reconcilable with an assumption of a ‘self-regulating’ natural rate of interest towards which the economy reverts (lower interest rates cause higher interest rates)? I hope I’m not being stupid here.
Ralph: agreed.
M Ricks: let me try a couple of shots in the dark:
There are two separate questions:
1. when we measure interest rates, in what good do we measure them? this is like the question, in what good do we measure prices? In a monetary economy, it is a lot easier if we measure interest rates, and prices, in money, and this is what we normally do. That’s the nominal interest rate, and nominal price. But we also talk about real interest rates, and real prices, when we measure interest rates, and prices, in terms of the CPI bundle of goods. But we could use wheat, or barley, or anything.
2. Does money pay interest? Currency does not. It pays 0% interest, measured in money. In Canada, currency normally pays minus 2% measured in terms of the CPI basket (the Bank of Canada targets 2% CPI inflation). Demand deposits some times pay interest, but usually don’t. Holding money pays you 0% nominal interest. Holding an IOU to money (a bond) pays you (say) 3% nominal interest. The demand to hold money is what we mean by “the demand for money”.
When I rent $100 purchasing power for 1 year it is possible, but unlikely, that I plan to hold an extra $100 in my pocket on average over the next year. Only in that rare case will a $100 demand for loanable funds happen to coincide with a $100 additional demand for money. Far more likely I plan to buy something else, like a cell-phone, and hold that in my pocket for 1 year. Or a bike, or a…., or just spend the $100 on a meal, or a…whatever.
Ritwick: glad you found my old post!
I was exaggerating a bit when I said an upward-sloping IS curve reconciles ISLM with Scott Sumner. Though it does help reconcile the two.
You are not being (obviously) stupid. Or if you are, that makes two of us!
The (Neo-)Wicksellian model sees no automatic self-regulating properties in the economy. It needs the central bank to actively set the right interest rate, and keep moving the interest rate it sets in response to changes in the natural rate, and inflation and output, to keep the economy from exploding into an inflationary boom or imploding into a deflationary recession.
Neo-Wicksellians think of the central bank as setting a nominal interest rate. Scott Sumner sees the central bank as setting (expected future) Nominal GDP. That makes a big difference to whether the equilibrium is stable, if the central bank gets it wrong. And whether the IS slopes up or down also makes a difference to stability. It would probably take me a whole post to lay it all out with any attempt at clarity.
The “demand for money” means the stock of media of exchange that people want to hold on average over the month.
Nick, I know this is a sort of staple linguistic usage among some of the monetarists. But I don’t understand the principle behind it, and it just seems to me to introduce unnecessary obscurity in the place of clarity.
Compare: People desire to acquire paintings. Some want to acquire them so that they can put them on their wall and look at them in admiration. Some want to to acquire them so people will say, “There goes that guy with all the paintings!” Some want to acquire them so that they can safely keep them and hold onto them for some more-or-less extended period, and then exchange them at some point in the future for something else, having bet on an appreciation, or at least preservation, in their value over time. Others want to acquire them so that they can exchange them more-or-less immediately for something else.
All four of these factors figure into the reasons for which people desire to acquire paintings. And their prevalence and strength manifest themselves in the marketplace by people’s willingness to exchange stuff they already possess for painting, and by the value of what they are willing to exchange. In other words, all four factors contribute to the market demand for paintings.
I don’t see why it should be any different in the case of dollars. The first of the four factors doesn’t play much of a role in desires for acquiring dollars. But the other three do. They all contribute to the demand for dollars. I don’t see why we should isolate just one of those factors as constituting the demand for dollars. We wouldn’t say in the case of paintings that only those bids for paintings that are driven by the desire to hold the paintings for extended periods of time are constitutive of the demand for paintings, would we?
Credit allows considerations of time preference and time constraints to enter into exchange contracts. Leave dollars out of it for now. Even if we are bartering, we might employ credit to complete an exchange. The agreement might be for you to deliver a certain quantity of feed to me now, and for me to deliver a certain number of animals to you by a certain date in the fall. You give me the feed on credit – in exchange for the promise of delivery of the animals later.
Some credit exchanges might even involve the same kind of entity. You might agree to give me X bushels of seed corn now, in exchange for my promise to give you (1.15)X bushels of seed corn following my harvest. These are the kinds of exchanges that take place in the market for money: one party gives a second party a certain amount of money at an earlier time, and in exchange receives a promise that the second party will deliver a greater amount of money at some later time. But no matter what others are willing to exchange for my money, and on no matter what time schedule for the delivery of my money to them, the desire for such exchanges contributes to the demand for money.
If I expect to have some mature animals to sell in the fall, I might entertain bids on all sorts of contracts for the delivery of those animals when they are available. I might make some deals to deliver a certain number of animals in the fall in exchange for some other good being delivered to me immediately. Or I might make a deal to deliver the animals in exchange for some other good being delivered to me at the same time I deliver the animals. But the amount and value of the bids presented to me for my consideration constitutes the demand for my animals. Similarly, the amount and value of the bids presented to me for my consideration, offering goods in exchange for my money, constitutes the demand for my money.
Dan: “Nick, I know this is a sort of staple linguistic usage among some of the monetarists.”
It’s standard linguistic usage among (most) economists.
For paintings, just like money, we can talk about the desired stock of paintings. If people acquire paintings/money, some people will plan to hold those paintings/money for a short time, and others will plan to hold them for a long time. So what’s the difference? (Other than the fungibility of money and the non-fungibility of paintings).
Right here: when you say: “These are the kinds of exchanges that take place in the market for money….”
There is a market for paintings. Paintings exchange for money in that market. What is the “market for money”? It is every single market in a monetary exchange economy. “The money market” is an oxymoron (or a redundancy?) in a monetary exchange economy. Every market is a market for money, plus one other good. When people talk very loosely about “the money market”, what they really mean is “the bond market” or “the market for short-term bonds” or “the market for IOUs”.
Suppose I decide to hold less furniture and more paintings. First I sell my furniture for money in the furniture/money market, then I sell my money for paintings in the paintings/money market.
Suppose the price of paintings is fixed too low relative to everything else (by law, say) given the actual stock of paintings and people’s desire to hold paintings. People want to hold a higher stock of paintings at that price than they currently hold. There is excess demand for paintings in the market for paintings. People cannot buy as many paintings as they want. End of story. Realising they cannot buy more paintings, because nobody wants to sell paintings, they buy something else instead, and the world goes on otherwise as normal.
Let me repeat the above substituting “money” for “paintings”.
Suppose the price of money is fixed too low relative to everything else (by law, say) given the actual stock of money and people’s desire to hold money. (Since we measure all prices in money, this means that the price of everything else is fixed too high). People want to hold a higher stock of money at that price than they currently hold. There is excess demand for money in every market. People cannot buy as much money as they want. (I.e. they cannot sell as much of everything else as they want). That is not the end of the story. Realising they cannot buy more money (sell more goods), because nobody wants to sell money (buy goods), they stop selling money (stop buying goods). There’s an excess demand for money and an excess supply of other goods in all markets. There’s a recession.
The demand for money is like the desired average stock of inventory of paintings for an art dealer. Except everyone holds an inventory of money, and only art dealers hold an inventory of paintings. And when we buy or sell anything else, our inventory of money falls and rises. We are all like art dealers, who only desire paintings for dealing, and who buy and sell everything else only for paintings.
Nick’s post said: “The loan market dries up, but the (stock) demand for money does not dry up. But if banks create money by making loans, the (stock) supply of money dries up.”
So what happens to the amount of medium of exchange if no entity goes into debt (wants a loan)?