Money demand and supply in a red/green world

Because it's Sunday.

"Green" money has positive value; "red" money has negative value. (An overdraft in your chequing account is red money.) We live in a red/green world with both types of money, so we ought to start thinking about money supply and demand in a red/green world. The red/green world is the real world.

For simplicity, consolidate commercial banks and central bank into One Big Bank.

In the green world, the bank issues green bits of paper worth plus $1 each. If you buy $10 worth of apples, the buyer must give the seller 10 green bits of paper.

It doesn't make any difference if people keep their green bits of paper in a cardboard box at the bank, instead of in their pockets. It doesn't make any difference if the bits of paper and cardboard boxes disappear, if the bank keeps a record in a ledger.

In the red world, the bank issues red bits of paper worth minus $1 each. If you sell $10 worth of apples, the seller must give the buyer 10 red bits of paper.

It doesn't make any difference if people keep their red bits of paper in a cardboard box at the bank, instead of in their pockets. It doesn't make any difference if the bits of paper and cardboard boxes disappear, if the bank keeps a record in a ledger.

There is a symmetry between the green world and the red world. Both worlds have a Cash-In-Advance constraint. In the green world the buyer of goods must have green cash to give to the seller; in the red world the seller of goods must have red cash to give to the buyer. But there is also an asymmetry: in the red world the bank must impose collateral constraints to put a limit on how many red notes an individual can hold, otherwise an individual would want to accumulate an unlimited number of red notes, and die holding more red notes than his wealth.

The red/green world is a composite world where both red and green notes are used as money.

The bank ensures that one red note is always worth exactly minus one green note, in exactly the same way that banks ensure that one $20 note is always worth exactly two $10 notes. The bank will freely swap one green note plus one red note for nothing, and vice versa. Just as banks will freely swap one $20 note for two $10 notes, and vice versa.

Define the "gross" money stock as the number of green notes plus the number of red notes. Define the "net" money stock as the number of green notes minus the number of red notes.

The bank has four monetary policy instruments:

1. Open Market Operations. If the bank buys $100 worth of bonds (or buys $100 worth of anything), this increases the net money stock by $100.
2. Collateral requirements. If the bank tightens collateral requirements, and if those are binding constraints, this will reduce the gross money stock but not the net money stock.
3. The green interest rate (which may be positive or negative) it pays on holdings of green notes. Increasing the green interest rate increases the demand to hold green notes.
4. The red interest rate (which may be positive or negative) it charges on holdings of red notes. Increasing the red interest rate reduces the demand to hold red notes.

The red interest rate cannot be below the green interest rate, or the bank would create arbitrage opportunities. The red-green interest rate spread cannot fall below zero. If you like, we could replace the fourth monetary policy instrument (the red interest rate) with the red-green spread.

The demand for gross money will be an increasing function of the volume of transactions and will be proportional to the price level. It will be a decreasing function of the degree of synchronisation of payments and receipts, and it is costly to synchronise your payments and receipts.  Standard Quantity-Theoretic stuff.

The demand for net money will be an increasing function of both red and green interest rates. The demand for gross money will be a decreasing function of the spread between red and green interest rates.

In a representative agent model, where payments and receipts of money are perfectly synchronised by assumption (if I buy $100 worth of goods from others, then others buy exactly $100 worth of goods from me at exactly the same time) the demand for gross money will always be zero. (In Michael Woodford's "cashless" representative agent model, the demand for gross money is always zero, and the central bank ensures the supply of net money is always zero, and sets the spread between red and green interest rates at zero.)

The bank controls the net money supply via open market operations, and controls the net money demand via red and green interest rates. An open market purchase increases the net money supply, and an increase in red and green interest rates raises the net money demand. So the bank can use open market operations and interest rates at the same time in an offsetting way without affecting the excess demand for net money.

If collateral constraints are non-binding, the bank controls the net money stock, but individuals determine the gross money stock. Any individual can go to the bank and ask to be given one green note and one red note. The only way the bank can reduce the gross money stock is by tightening collateral requirements, or by increasing the spread between red and green interest rates. With zero spread, and non-binding collateral constraints, individuals are satiated in gross money, but may or may not be satiated in net money.

I think that's right.

Comments

[Matt Youngh]

The ventral banker has one more option for its operations. It can leave rate and term unknown until the ratio of red to green becomes out of balance, at which time the ratio is re-balanced by subtracting green from accounts to pay off reds. In this scenario, the rate and term is discovered ex-post. Member banks only know their true rates and terms after the rebalance.

Something has to be discovered by the central banker, and the simplest variables to discover are rate and term. What is an imbalance? You decide, what is your economic model? There is no absolute determination that the short term, except in agricultural economies.

But ultimately one has to pick an economic model, otherwise the definition of imbalance is discretionary, and risk is not uniformly observed. Why would central banks know he true interest rate apriori? Member banks still have to operate from their estimation of the natural rate, which is unknown until discovered.

[Marko]

Do I get a tax deduction if I donate my red notes to charity ? 'Cuz that's all I got , is red notes. Lots.

[Jussi]

"If collateral constraints are non-binding, the bank controls the net money stock, but individuals determine the gross money stock. Any individual can go to the bank and ask to be given one green note and one red note."

I think these green/red posts are really interesting. And quite relevant as it seems to me that our economies are going towards something like this. Credit cards and likes IMO can be already viewed as red money and financial markets often times allows negative balances which are collateralized.

Anyway I had a question. Can we say anything on monetary transmission given green/red money world and if collateral constraints are not binding? Can the central bank target NGDP/inflation and should it use interest rates or manipulate net money stock?

[Nick Edmonds]

From an individual's perspective, it is useful to identify two important measures here. Ones is simply their net position, which represents their wealth. The second is the sum of their green money and their unused borrowing capacity. Here, we might think of this latter measure as representing their liquidity.

If there are no collateral constraints, the demand for money will be driven by the demand for wealth. Of course, there may be fluctuations because payments are not neatly co-ordinated, but this is an more an ex-post outcome, than it is a reflection of ex-ante demand.

As we start to tighten collateral requirements, people start needing to hold green money, not simply due to their demand for wealth, but also to allow them to make all the payments they want. If there is no scope to borrow, people have to hold a stock of green money to cover any possible temporary shortfall between payments and receipts. Technically, the reduced borrowing capacity reduces the possible future time-paths of the individual's net money position, which changes the individual's optimal decision in the face of uncertainty over the future.

In an extended model, we would need to consider the difference between committed and uncommitted borrowing capacity as we would between sight and term green money. But this type of framework is very useful for thinking about some of the things that were happening at the peak of the financial crisis.

[Nick Rowe]

Matt: I didn't follow that.

Marko: No! You get a tax deduction if you *take away* a charity's red notes.

Jussi: thanks. I don't think this post says anything about whether the Bank should target inflation, or NGDP, or whatever. But if it wanted to loosen monetary policy it could either cut the red and green interest rates, or increase the net money supply.

Nick E: I'm still thinking this through, but I think "unused borrowing capacity" is the wrong word here. We should instead say "unused capacity to hold red money". Some financial liabilities are money, but most are not.

Similarly, an individual's net money is only a small part of his wealth.

In the red world, for example, liquidity is the stock of red money. Because you cannot sell goods unless you have red money. But if the potential buyer of goods has a binding collateral constraint, and so cannot accept more red money, then he cannot buy goods. So you don't get an exchange unless the seller has red money and the buyer can accept more red money.

[JKH]

How does the liquidity characteristic of the bank's net position in non-money figure into the analysis?

[Nick Rowe]

JKH: if net money is positive, and so is a net liability of the bank, the bank has assets on the other side of its balance sheet. Those assets are not money, but may be more liquid or less liquid. Does it matter? Probably, because the more liquid the assets the bank holds, the less liquid the assets the non-bank public holds, which would presumably increase the demand for gross money.

But I'm still struggling to get my head around red and green money, so can't think about that complication.

[Jussi]

Nick: "4. The red interest rate"

I think there cannot be uniform interest on red money. The interest needs always reflect the creditworthiness (or collateral pool attached) of the agent holding the red money. Any consumption by receiving red money means lower credit.

[Jussi]

"But if it wanted to loosen monetary policy it could either cut the red and green interest rates, or increase the net money supply."

I see. I wasn't sure if you were thinking that the transmission channels are intact after having red money involved.

Does the increase in net money supply help in similar fashion that in green money world? I'm asking because that would mean holding more assets in the central bank's balance sheet - that reduces red money collateral pool and thus the amount of (potential) red money - as first iteration I would say it is a wash? Yet I see even the current system will utilize collateral - asset buying will probably mean, considering collateral channel only, less loans (and less deposit money) even in the current system.

[JKH]

"The demand for net money will be an increasing function of both red and green interest rates."

How might this relate to basic liquidity preference theory in the case of a single interest rate?

[Roger Sparks]

".........But there is also an asymmetry: in the red world the bank must impose collateral constraints..........."

I think this green-red model is incomplete due to the weak development of the strong link between red currency and collateral.

Lacking a strong tie between red currency and collateral, you write "The bank will freely swap one green note plus one red note for nothing, and vice versa." This implies that an actor can freely walk to the bank and obtain both red and green money, which is noticeably out-of-step with the real world.

In the real economy, red and green money would only be issued by the bank if the customer first provided collateral which the bank could tie to the red type of money. This collateral link to red money is preserved until the red money is finally combined with green and turned into "nothing". The destruction of both green and red money types would leave un-linked collateral still existing in the real economy.

[Nick Rowe]

JKH: Simple liquidity preference (money demand) theory assumes that "money" is like currency and pays 0% interest. So "the" interest rate in that theory is the rate of interest paid on non-money assets. The higher that interest rate, the greater the opportunity cost of holding money, and the lower the quantity of money demanded.

If we generalise that theory, to allow money to pay interest, then the opportunity cost of holding money is the spread between the non-money rate of interest minus the rate of interest paid on money.

So what I said was compatible with standard money demand theory. Just I'm talking about a different interest rate, which is why you get the opposite effect.

[Nick Rowe]

Jussi: "I think there cannot be uniform interest on red money. The interest needs always reflect the creditworthiness (or collateral pool attached) of the agent holding the red money."

Sounds right to me. But I'm going to ignore it, for simplicity. ;-)

"Does the increase in net money supply help in similar fashion that in green money world?"

In the green money world, net money and gross money are the same thing, so it's simpler. In the red/green world, an increase in net money will only increase gross money if collateral constraints are binding (I think). The green money world is like the limiting case of a red/green world in which haircuts approach 100%.

Roger: "In the real economy, red and green money would only be issued by the bank if the customer first provided collateral which the bank could tie to the red type of money."

That is true in my red/green world too. You are limited in how much red money you are allowed to hold, regardless of where you get it from.

"This collateral link to red money is preserved until the red money is finally combined with green and turned into "nothing"."

You lost me.

[Roger Sparks]

Nick,

"You are limited in how much red money you are allowed to hold......."

There may be some confusion about WHO holds the red money. In the real world, there is no limit upon the maximum amount of red money that could be held. The customer will not hold red money; instead, the bank or seller will hold the red money and the customer will retain possession of the collateral.

The limit on amount of red money that each customer is allowed to create (in cooperation with a bank or seller) is related to the amount of collateral owned by each customer.

[Majromax]

I think one missing step here is why both red and green money should circulate in this world.

At 7:15AM, you note that red money is a constraint on the sale of goods. But that's only true in a red-money-only world; in a green-money-only world a merchant would accept green money for the goods and we'd be done. In our mixed world, the merchant has both options.

I think we address this problem via financial intermediaries. The spontaneous creation of one red and one green money is only a problem because of default risk, but for a merchant who intends to immediately give that red money away with a purchase there is no default risk. The risk properly lies with the consumer -- but that's just where merchant banks come in.

In our red-and-green world, a merchant who accepts cash and credit cards is indifferent to two options:

*) They accept green money in exchange for goods, or
*) The customer swipes a piece of plastic, and a merchant bank creates red and green money in equal amounts. The consumer receives the red money and the goods, and the merchant receives the green money (less processing fees).

In both cases, the merchant's books look just like they're a green-money-only shop. The credit card payment fees cover the aggregate risk of default (since merchants are insulated from an individual customer's default risk), and if we assume that collateral constraints are effective this would be near-0% and only a convenience fee that we can neglect.

The red-green interest rate spread determines when the *customer* chooses to use red or green money for the transaction; the merchant does not make that choice. Collateral requirements determine which customers have the option to use red money and which must use only green money.

These are not independent monetary policies. In fact, your four variables reduce to three:

* Seigniorage Revenue: The Central Bank determines how much seigniorage revenue it wishes to generate, in terms of incoming revenue (from collateral, net interest on money). If we don't bring in a fiscal authority to receive a surplus / make up a shortfall, positive seigniorage revenue contracts the money supply and negative revenue expands it.
* Open Market Operations and the Green Interest Rate: Together, these determine the time-structure of the money supply. With a fixed target of seigniorage, an open market operation results in the CB purchasing an asset who's returns must be disbursed as interest on green money. If the central bank holds only nominal assets, then this interest is "interest on money" in the Neo-Fischerian sense which expands the money supply. If the central bank holds only real assets, then things are a lot more confusing because a green dollar would pay (ultimately) interest in milk or housing rent or Twinkies.
* Collateral and the Red/Green Interest Rate Spread: If the Central Bank took anything as collateral, conducted no open market operations, and targeted zero seigniorage revenue, then the red/green spread would be governed precisely by the collective default risk on that collateral. Conversely, if the Central Bank set a fixed red/green spread under the same zero-seigniorage condition, then the quality of collateral would be tightly-constrained by that interest rate. If the central bank accepts real assets as collateral, this also brings the inflation rate into the picture.

The Bank of Canada -- dealing in reserves -- seems to work by targeting low seigniorage revenue and a fixed red/green spread (bank rate minus deposit rate), using open market operations and the green interest rate as its primary policy lever.

The aggregate banking system still targets low seigniorage revenue (private banks can't collect seigniorage, the central bank's target is set by policy), fixes the green money rate (cash pays zero interest, chequeing accounts pay very low interest rates), and adjusts the red/green spread (0% cash less financing cost on unsecured loans) and rate of return on the banks' capital in tandem.

[Nick Rowe]

Majro: "I think one missing step here is why both red and green money should circulate in this world."

Suppose the bank sets net money at zero (which it can easily do, via OMO). Then unless we have both red and green money, there will be no money at all.

[Nick Edmonds]

"In the red world, for example, liquidity is the stock of red money."

I don't think you should worry about the red world. The real world, as you say, is a world of red and green money. Sometimes thinking about a world of green money is OK, as a simplification of the real world. But a world of red money alone is not a good approximation of the real world.

[Nick Rowe]

Nick E: True, but thinking about the red world helps me think more clearly about the red/green world.

[Roger Sparks]

Nick,

Some of our movies of the old west contain a good example of the creation (delivery) of red money: The character comes into a bar, orders a drink, and says "Put it on my tab.".

If the barkeeper agrees, the customer gets his drink and the bar adds a line to the customers tab. The bar is keeping the customers red money on deposit at the bar. The customer is delivering red money to the bar each time a drink is ordered.

Occasionally, the customer is refused. The barkeeper claims the tab is too long (the customer owns too much red money). You seldom hear this expression of a red money limitation in a modern western.

[Majromax]

@Nick:

(First, I realize now that I misread your comment from earlier this morning, where you were talking about red circulation in a red-only world; I mistook it as the red-green world.)

> Suppose the bank sets net money at zero (which it can easily do, via OMO). Then unless we have both red and green money, there will be no money at all.

I don't think that's enough. Sure, we must -have- red and green money in this zero-net world, but it would still be possible for only one colour to circulate as money.

Ordinarily, we think of green money as circulating even though we live in a mixed world. This is because an individual seller only deals in green money (Sparks's bar tab above notwithstanding) because individual sellers cannot determine the creditworthiness of an individual buyer. Look at the non-acceptance of cheques as payment, as well.

Modern plastic financial instruments allow individuals to use red money while merchants deal with only green money.

Still, I think the bicoloured world is important, because a red/green world presents a very different liquidity problem for consumers than a green-only world.

[Market Fiscalist]

'So the bank can use open market operations and interest rates at the same time in an offsetting way without affecting the excess demand for net money.'

I'm not fully getting this statement.

Suppose that people hold their non-money wealth in the form of assets that can be rented out so that they have an interest rate equivalent to rent / value. The interest rates on red and green money will have to be aligned with the interest on these non-money assets or arbitrage opportunities will exists that will cause non-money asset price to change, which will in turn cause the demand for money to change.

So if the bank wants to not affect the excess demand for net money won't it have have to set the red/green rate at a level that hold asset prices steady, and then use OMO to keep the money supply at just the right level to meet people's demand for it at those interest rates?

[Min]

"Open Market Operations. If the bank buys $100 worth of bonds (or buys $100 worth of anything), this increases the net money stock by $100."

Isn't this different from typical, if not universal practice today? Don't CBs keep their books balanced, so that the net money stock is unaffected by their open market operations? That's why some people worry about a CB's balance sheet, right?

[Min]

I mean, net money stock as defined in your metaphor. :)

[Nick Rowe]

Roger: If you give me a drink, and I give you my IOU for $5, that IOU is not money. It's just an IOU. But if you then use my IOU to buy apples, and the apple seller uses it to buy fertiliser, and it circulates around the town, then my IOU is money, and I have created money.

[why not use black and red? like we already do (sure, you're money specific but?)]

I don't get the point of this exercise. So instead of debit/credit we say red/green? So what? Instead of a ledger you use red notes to represent one side and green for the other? Again, what's the point? What is revealed by this relabelling of existing constructs? Or is this so clever its over my head?

[why not use black and red? like we already do (sure, you're money specific but?)]

or positive and negative

[Nick Rowe]

MF: currency pays 0% nominal. Bonds (usually) pay positive nominal rates. Does this create an arbitrage opportunity? No. (But it does create an incentive for banks to create money).

Min: take the green money world. The bank has money on the liability side of its balance sheet and non-money assets ("bonds") on the asset side. An Open Market Purchase of $100 worth of bonds increases both sides of the balance sheet by $100. Green money (which equals net money if there is no red) increases by $100. Standard story.

why not use....is totally missing it (and is not alone). ***Not all financial assets and liabilities are money***. When he starts talking about "debit" and "credit", you know he is missing the point.

[why not use black and red? like we already do (sure, you're money specific but?)]

Lol. So it's really clever. What does color coding positive and negative money balances reveal? Which part didn't I get?

Congratulations on being the first to color code positive and negative.

[Min]

Nick Rowe: "Min: take the green money world. The bank has money on the liability side of its balance sheet and non-money assets ("bonds") on the asset side. An Open Market Purchase of $100 worth of bonds increases both sides of the balance sheet by $100. Green money (which equals net money if there is no red) increases by $100. Standard story."

But what about the balance sheet before buying the bonds? If the bank already has the green money it will use to buy the bonds, then buying the bonds does not alter the net amount of money. It only does so when the bank creates green money in order to buy the bonds without at the same time creating red money to keep the net amount of money the same. Is is standard for the CB to create money for its own use without balancing its books?

[why not use black and red? like we already do (sure, you're money specific but?)]

debit

NOUN

an entry recording an amount owed, listed on the left-hand side or column of an account. The opposite of credit.

VERB

(be debited)

(of a bank or other financial organization) remove (an amount of money) from a customer's account, typically as payment for services or goods.

How is your red and green money different?

Are you suggesting that if we used debits and credits instead of red and green your post would make no sense?

"Credit" money has positive value; "debit" money has negative value. (An overdraft in your chequing account is a debit.) We live in a debit/credit world with both types of money, so we ought to start thinking about money supply and demand in a debit/credit world. The debit/credit world is the real world.

or

"Positive" money has positive value; "negative" money has negative value. (An overdraft in your chequing account is negative money.) We live in a negative/positive world with both types of money, so we ought to start thinking about money supply and demand in a negative/positive world. The negative/positive world is the real world.

[Min]

Sorry, Nick, I shifted back to your metaphor.

Nick Rowe: "An Open Market Purchase of $100 worth of bonds increases both sides of the balance sheet by $100. Green money (which equals net money if there is no red) increases by $100."

Are you saying that the (green) money created by the CB to buy the bonds remains with the bank? Doesn't it got to the seller of the bonds?

[Min]

OC, since we have a unified bank, the green money is presumably in the bank, but isn't it in the seller's account, not the bank's account? If there is nothing in the bank's account to represent its liability, then the bank has on its face gotten richer by the amount of the purchased bonds. The bank's books after the purchase are out of balance with the books before the purchase. Is that SOP?

[Nick Rowe]

why not...Google "medium of exchange".

And stop commenting here. You are trolling, and wasting my time.

Min: The bank prints green money, and when it uses it to buy bonds, of course it goes to the seller of the bonds. This is very standard stuff.

[Market Fiscalist]

"currency pays 0% nominal. Bonds (usually) pay positive nominal rates. Does this create an arbitrage opportunity? No. (But it does create an incentive for banks to create money)."

Well, if people want to hold 0% money then banks can make a profit by creating this money and buying interest-bearing assets with it.

But I understood your statement from the post to mean that the bank in your model can satisfy the demand for money either by adjusting the money supply via OMO, or by adjusting the interest rates it pays/charges on greed/red money or by a combination of the 2. My point was that for a given policy goal (say a stable price level) then there is only 1 interest rate (the one that equilibriates the rate paid on money, with that earned on non-money assets) consistent with that policy. Any other interest rates will lead to a change in the price level. And there will only be one level of money supply that matches money demand at that interest rate.

[Market Fiscalist]

Oh, I think I see where my thinking was wrong.

If you pay people interest on currency, then you will be able to vary either the money supply or the interest paid on currency to achieve similar results in terms of the price level. I was thinking of interest rates as something you get for lending out the currency you hold, but that is not true in your model.

[Nick Rowe]

MF: you've got it.

[Min]

OK, I have studied a bit, and I think that I understand a bit better. With a unified bank, there are no reserves, so when the bank buys a bond, it puts newly created (green) money in the seller's bank account, which is the bank's liability and balances the asset of the bond.

So in your metaphor some, but not all bank liabilities are green money and some, but not all bank assets are red money. So far the only real life bank asset that you have identified as red money is overdrafts. I suppose that there are others. What are they? Are all bank loans red money? Thanks.

[Jussi]

Having thinking about this more:

I can see there is a case where a transaction happens in red/green but not in green only world. But can there be a case where red/green makes the transaction without need for additional (buyer's) collateral? If not doesn't that mean a bank has to be involved to make the credit analysis (a merchant cannot do it and only bank can issue paired red/green notes). Can red money be genuine medium of exchange?

This seems to be, at the operational level, the same as in green-only world where a bank is also needed for credit analysis. The only difference now is that bank books a loan, not red money assets (customer's liability) on its books. Both needs the same collateral arrangement. But the credit rationing/analysis and the process itself, apart of the instruments is the same. Then does red money actually help the economy as medium of exchange? Or is the difference only that red/green world has an additional uniform liability type (central banks's asset) that is functionally the same as a bank loan but has different characteristics. Loans are priced against credit but red money's price is set by the central bank.

[Nick Rowe]

Min: First paragraph: you have got it. (But remember that regular commercial banks don't hold 100% reserves either, so they are not very different.)

Second paragraph: you have nearly got it. We can imagine a red/green bank with zero net money, where the only asset is red money and the only liability is green money. If net money > 0, then the bank must hold some non-money assets, which are non-money IOUs signed by someone other than the bank (e.g. govt bonds, mortgages). If net money < 0, the bank must have some non-money liabilities, which are IOUs signed by the bank (banks issue bonds).

Jussi: You lost me.

Strictly, we don't need banks at all. Red money is garbage, which people don't like holding, but can't be thrown away. All garbage is the same, so it works well as money. "If I give you 10 apples, will you take 1 kg of my garbage in exchange?"

[JKH]

Something that is workable at the margin should be examined in the limit.

How is money created in the pure red money world?

Or - what is the pure red money version of "loans create deposits" ?

(MMT headstands permissible if fully collaterized with clear thoughts)

:)

[Jussi]

Nick: "Jussi: You lost me."

JKH: "How is money created in the pure red money world?"

I second that.

"Strictly, we don't need banks at all. Red money is garbage, which people don't like holding, but can't be thrown away. All garbage is the same, so it works well as money. "If I give you 10 apples, will you take 1 kg of my garbage in exchange?""

Yes, that works but only if you have some amount of _standardized_ (in value sense) packages of garbage. But there is no such thing lying around. And it is not good enough that the central bank issues red notes as the notes need the collateral to support them - otherwise they are not standard and uniformly priced. And collateral needs to be repriced everytime the red note changes hands.

As you said otherwise people will create garbage and "die holding more red notes than his wealth." How do you limit garbage to concentrate to people who doesn't have wealth to match the garbage without credit rationing (and thus banks)?

[Nick Rowe]

JKH: "How is money created in the pure red money world?"

Good question.

My initial thought was that it was bits of paper that were convertible on supply into garbage! (You hold a red note, you must accept my garbage if I ask you to.) But they could equally be an obligation to pay gold to a bank.

I have some assets. I give you my assets, plus some red notes of equal value. If those red notes circulate as money, I am a red bank.

In the green world, you have some assets, and you give me your assets, and I give you some green notes of equal value.

Does that sound right?

[Nick Rowe]

JKH: maybe this is better/clearer:

Green world: Adam signs an IOU promising to pay the bank 100 green notes, and the bank gives Adam 100 green notes. Adam now uses the 100 green notes to buy a house.

Red world: The bank signs an IOU promising to accept 100 red notes from Adam, and the bank gives Adam 100 red notes. Adam now uses the 100 red notes to sell a house.

[Nick Rowe]

God it's weird.

An even weirder thought (which must be/I hope is wrong): start with a green world. What prevents it suddenly becoming a red world? Are there two sunspot equilibria? Could green money suddenly change colour to red? So positively-valued money suddenly becomes negatively-valued, simply because everyone expects it to become negatively-valued?

[Nick Rowe]

But there's an asymmetry here between the red and green worlds. The reason why banks exist is that everyone trusts the Bank's signature, but nobody trusts Adam's signature (except the bank, which has done research on Adam).

[Roger Sparks]

Nick,

You wrote ".......... in the red world the seller of goods must have red cash to give to the buyer.............."

In the real world, sometimes a home owner with debt-owing-on-the-home will sell the home and debt to a new buyer. Is this equivalent to the red world transaction referenced in the quote?

[Jussi]

"I have some assets. I give you my assets, plus some red notes of equal value. If those red notes circulate as money, I am a red bank."

IMO would do but is too kind to me as I will sell those red notes to someone without assets and dying soon. I'm sure someone poor and old is happy to get some assets and die with hefty stack of red notes.

[Market Fiscalist]

"How is money created in the pure red money world?"

Its easy to see how IOU's (that are held by the lender) can become green money. Isn't it also possible that invoices issued by the lender can become red money?

I agree to sell you 100 apples and give you an invoice to be paid in the future. You can either return the 100 apples plus interest in the future, or you sell something you own to someone who agree to accept the invoice in exchange. Isn't the invoice red money ? Whenever a debt is created isn't there always a potential for both green money (the IOU) and red money (the debt contract) to be created and to circulate ?

[Market Fiscalist]

And couldn't a CB create red money by OMO. It finds people who are prepared to accept a share in a govt-owned asset plus some red notes with negative interest which start to circulate and we're in a red world.

[Jussi]

MF: How do you ensure that people won't die without assets and with some (then defaulted) red notes?

[JKH]

I do think it is useful to consider a pure red money world to check logic - not only for the mixed world, but even for the usual green money world.

Suppose I am employed in the red money world, and I get paid at the end of the month.

On those dates, I get paid by transferring red money to my employer. Basically, I get my compensation by saddling him with my liabilities. So my money wealth increases each time that happens.

Suppose I come to one of those pay dates and I've run out of red money. So I have no means of getting paid.

I need red money.

So I go to the bank. 

In return for my issuing my own red money liability to the bank, the bank issues me a term deposit which is repayable by the bank at maturity through the transfer of red money from me to the bank.

This expands the banking system balance sheet, which consists largely of red money assets and term liabilities that are issued and settled in red money.

No green money is involved.

And I get paid by transferring newly created red money to my employer.

A few interesting things about this I think:

First, it is the opposite asset-liability orientation of conventional green money banking, where banks are thought to borrow short and lend long. This happens frequently through accounting entries where loans and green money deposits are established simultaneously at the point of the loan advance.

Here, banks borrow long and "lend" short in the form of red money.

Returning to the example, the red money banking system responds to the need for liquidity by expanding the supply of what are essentially transferable, highly liquid bank assets / customer liabilities in the form of red money.

Finally, I notice that this is a very weird variation on the 1930's "Chicago Plan" design for banking. The very essence of such plans which continue to be promoted today is that commercial banks be precluded from issuing green money and that they should instead issue term liabilities to protect their own liquidity.

Note:

I can imagine refined arguments and debates about whether red money is a form of "liability". I don't see that as an important discussion. The important point is that for purposes of simple, comparative balance sheet representation,  red money will show as a bank asset and a customer liability - opposite to the case of green money.

[Market Fiscalist]

Jussi,

"How do you ensure that people won't die without assets and with some (then defaulted) red notes?'

People who issue red notes would need to have good credit-worthiness checkers, just like bank do today.

[Jussi]

"People who issue red notes would need to have good credit-worthiness checkers, just like bank do today."

Indeed. But are the red notes then truly medium of exchange if the credit-worthiness needs to be checked at the time the payment is made?

[Nick Rowe]

JKH: "I do think it is useful to consider a pure red money world to check logic - not only for the mixed world, but even for the usual green money world."

I totally agree. It helps us see symmetries, and also possible asymmetries.

The role of banks, in the red world, is to certify that their customers are able to hold red notes. The bank guarantees their customers. I think that's right.

[Jussi]

"The role of banks, in the red world, is to certify that their customers are able to hold red notes. The bank guarantees their customers. I think that's right."

That is the same as in green world - but what kind of transaction can be then done only in red/green world but not in green-only world?

[Majromax]

@Nick Rowe:

> The role of banks, in the red world, is to certify that their customers are able to hold red notes. The bank guarantees their customers. I think that's right.

That's the role of banks today in the mixed world.

I use red money for the vast majority of my daily transactions: I buy things on a credit card and settle the statement monthly. I can do this because my bank certifies via my credit card that I'm good for it, to the point where the bank is willing to insulate merchants from any lingering default risk.

My use of this red money would be curtailed if the bank substantially increased collateral requirements (in effect reducing my credit limit) or if it raised the red/green interest rate spread, such that I had to pay interest on transactions between the purchase date and statement date. Both obviously correspond to monetary policy instruments you note in this post.

[JKH]

The role of banks, in the red world, is to certify that their customers are able to hold red notes. The bank guarantees their customers. I think that's right."

But the critical issue is that this type of credit risk becomes embedded in the velocity of red money.

Each transfer of red money assigns the credit risk anew to the payee that received the liability.

It's not just the credit risk on the red money as originally created - it's the re-assignment of that credit risk to a new counterparty (for the bank) every time that money is used in a new transaction.

You've captured this in one sense with across the board collateralization limits for individual agents.

But this aspect may warrant more attention.

[Nick Rowe]

JKH: "But this aspect may warrant more attention."

I think that's right.

When there is One Big Bank, it's relatively simple. The bank decides every individual's red money limit. When there are multiple banks, it is decentralised. I get rid of my overdraft at BMO by selling my car to you, so you now have an overdraft at TD. Two banks were involved.

(The red world is one where nobody has a positive balance in his chequing account.)

[Nick Rowe]

JKH: "But this aspect may warrant more attention."

I think that's right.

When there is One Big Bank, it's relatively simple. The bank decides every individual's red money limit. When there are multiple banks, it is decentralised. I get rid of my overdraft at BMO by selling my car to you, so you now have an overdraft at TD. Two banks were involved.

(The red world is one where nobody has a positive balance in his chequing account.)

[Min]

Nick Rowe: "We can imagine a red/green bank with zero net money, where the only asset is red money and the only liability is green money. If net money > 0, then the bank must hold some non-money assets, which are non-money IOUs signed by someone other than the bank (e.g. govt bonds, mortgages). If net money < 0, the bank must have some non-money liabilities, which are IOUs signed by the bank (banks issue bonds)."

Yes, we can imagine this and that, but you claim that we actually live in a red/green world. Bank overdrafts are the only real world red money that you have identified. What else?

[Min]

I think that the example from a recent comment ( http://worthwhile.typepad.com/worthwhile_canadian_initi/2015/12/what-is-a-cashless-economy-anyway.html?cid=6a00d83451688169e201b7c7f9c4a8970b#comment-6a00d83451688169e201b7c7f9c4a8970b ) is an example of local red money. There is a frontier town with very little green money. There is no bank, but the general store acts like a bank with red money. Nearly everyone owes money to the general store. What they owe is like a bank overdraft and is thus a kind of red money. From time to time, green money comes to town, and it is nearly always used to pay the general store what people owe it, reducing their red money. The store then uses the green money to order more goods, and the green money leaves town. The towns people pay each other as a rule with IOUs, which the general store accepts and decreases the red money in the account of the person to whom money is owed and increases the red money in the account of the person who owes money.

This still doe not get us to anonymous red money notes, does it?

[Jussi]

Given all above I think we can say that targeting quantity of net money (OMOs) is not very effective given mixed world if the amount of red money dwarfs the amount of green money. Then amount of red money (credit limits) will mostly determine how easily the transactions can be made. This also explains nicely why business cycles are so tightly correlated with credit conditions. Can we say anything about red/green world and the alpha-beta bank distinction?

[Nick Rowe]

Min: "Bank overdrafts are the only real world red money that you have identified. What else?"

You mean bank overdrafts are not enough?

The very centre of the Canadian monetary system is a red/green system. Commercial bank accounts at the Bank of Canada are red/green accounts. The red interest rate is the Bank rate; the green interest rate is the deposit rate. Clearing houses operate on a red/green basis, I think.

Maybe JKH has a better answer.

For ALL OF YOU: not all financial liabilities are red money (just as not all financial assets are green money). It has to circulate as a medium of exchange to count as money.

[Nick Edmonds]

The credit risk issue shows why the liquidity position is different. In the red world, what matters is not the buyer's holding of money, but rather the willingness of his bank to allow him to accept more money. And the seller's balance of red money shouldn't constitute a liquidity constraint, because the creation of new red money to facilitate the sale should not depend on the credit of the seller.

[JKH]

I agree central banking can be interpreted as a real world case of the red/green idea in operation, with corresponding interest rates.

[Jussi]

"The very centre of the Canadian monetary system is a red/green system. Commercial bank accounts at the Bank of Canada are red/green accounts. The red interest rate is the Bank rate; the green interest rate is the deposit rate. Clearing houses operate on a red/green basis, I think."

"For ALL OF YOU: not all financial liabilities are red money (just as not all financial assets are green money). It has to circulate as a medium of exchange to count as money."

I think one source of confusion is that not all money is equal in all respects. E.g. clearing house accepts collateral or (green) money against negative positions. But are negative (mark-to-market) positions red money or just a sort of liability? Do they circulate - it is just a transfer wihtout a corresponding (goods) transaction, it just pays the reprincing of positions? Does that count as medium of exchange? Yet the asset pledged as collateral can be rehypothecated (https://en.wikipedia.org/wiki/Rehypothecation) - to cover many kind of liabilities (and surely not all of them are red money) - and transactions! Is collateral green money if used as medium of exchange or is the respective short position red money? What if there wasn't any short position to start with? Or is collateral actually a sort of loan (along paying with red money)? I think the definitions are not that clear cut.

[Market Fiscalist]

I was thinking more about the red world/green world stuff.

If you imagine a world where everyone has a bank account and where all transactions are electronic then then a bank could record people's positions in terms of either credit or debt. A sale from A to B could be seen equally as either a movement of credit from B balance to A that increases A's credit balance and reduces B, or a transfer of debts from A to B that reduces A's debt balance and increases Bs. The first is like green money the second like red money. Both are different ways of recording the same transaction and as long as the bank allows overdrafts of either credit amounts (first scenario) or debt amount (second scenario) there is no problem. A $100 credit balance would be the same as a -$100 debt balance (-$100 = negative debt), and a sale of something worth $20 by the account holder could be either a $20 inward transfer of credit increasing his credit balance to $120, or an outward transfer of $20 of debt reducing his debt balance to -$120 of debt (he now owes negative $120).

If you move from electronic transactions to one using physical tokens, then I think things map pretty much exactly. A sale from A to B could either be a movement of green tokens from B to A, or of red tokens from A to B. If people run out of tokens then the bank will (if the appropriate overdrafts are in place) handover more on demand. In this the tokens that people hold, plus the overdraft records that the bank hold, would correspond exactly to the electronic records in the cashless version.

If a bank chose to track people's account value in debits rather than credits and issued red rather than green cash then any interest payment it made to those in credit would appear as an increase in their negative balance of debits.

[Nick Rowe]

Nick E: my head is not clear on this yet.

"In the red world, what matters is not the buyer's holding of money, but rather the willingness of his bank to allow him to accept more money. And the seller's balance of red money shouldn't constitute a liquidity constraint, because the creation of new red money to facilitate the sale should not depend on the credit of the seller."

In the green world, the bank will not create $100 green money to let me buy a house worth less than $100 (unless I have other assets).

In the red world, the bank will not create $100 red money to let me sell a house worth less than $100 (unless I have other assets).

It looks symmetric.

JKH: Yep. What about clearing houses? And who sets the limits on red money? Any thoughts?

MF: Yep. That was pretty much my thinking too. (The only problem with red tokens is it's very hard to stop people throwing them away, or accumulating too many).

[Market Fiscalist]

I'm not following: "In the red world, the bank will not create $100 red money to let me sell a house worth less than $100 (unless I have other assets).'

Why not? I have a house worth $50 that I want to sell for $100. In the green world this is clearly a great deal d and poses no risk to the bank. In the red world the same thing surely must apply. Suppose the sale happens electronically - even if I have no red money at all, why would the bank not be happy to allow me to send $100 of debt to my buyer and change my balance to reflect an extra -$100 of debt. I'm $50 richer and have $100 less debt, and that's not a risk to the bank in the red world any more than the green one.

If I wish to make it a cash transaction - I go the bank and get them to give me $100 in red notes that I use to make the sale. When they give me the $100 in red notes they must change my balance with an extra -$100 of debt. This seems counter-intuitive but must be true since the cash transaction has to end up the same as the electronic version as far as my balance is concerned.

[Nick Rowe]

MF: by "worth less than $100" I mean you can't sell it for $100.

[Market Fiscalist]

OK.

If I have more than $100 in debt in my red bank account then my red bank will have no problems (if they trust me) of giving me $100 in notes to try and make the purchase and if I only sell for $50 I can either keep the extra notes or pay them back into my account.

If I have a negative debt balance (that is: I have a positive credit balance) why would they act differently ? They give me $100 in red notes and increase my negative net balance by a further $100. If I sell my house for only $50 they get to pass on $50 of debt to my buyers bank, and even if I just keep the notes under the bed it doesn't mater to them if they choose to charge interest on red money holdings.

[Frank Restly]

Nick,

In a red / green monetary world can the central bank successfully set the net spread between the value of red notes and the value of green notes? Yes they can set the interest rates on the various note types but I don't think they can ensure that a positive $1 green note is always equivalent to a negative $1 red note.

A premium may develop between the two types of notes based upon group preference. For instance, the central bank may pay 5% on green notes and charge -5% on red notes, but individuals may prefer to either pay $0.95 in green notes or receive $1 in red notes when purchasing a good (or vice versa). People may discount one note versus the other even if the interest rates on both are equivalent.

That premium may arise as the result of fiscal action (taxes are always paid in green notes not by accepting red notes), limitations on transfers (green notes are inheritable, red notes are not), inflation / deflation expectations, or other factors.

[Nick Rowe]

Frank: read the post. I said:

"The bank ensures that one red note is always worth exactly minus one green note, in exactly the same way that banks ensure that one $20 note is always worth exactly two $10 notes. The bank will freely swap one green note plus one red note for nothing, and vice versa. Just as banks will freely swap one $20 note for two $10 notes, and vice versa."

Do not reply.

[Roger Sparks]

Nick,

You write

"Green" money has positive value; "red" money has negative value. (An overdraft in your chequing account is red money.)"

and

"In the red world, the bank issues red bits of paper worth minus $1 each. If you sell $10 worth of apples, the seller must give the buyer 10 red bits of paper."

If we assume that the overdraft in the chequing account is $10, then we could have the apple seller giving a buyer $10 plus title to the overdrafted chequing account. Some apple buyers would be willing to make that trade.

This would be very similar to my example: "sometimes a home owner with debt-owing-on-the-home will sell the home and debt to a new buyer." I think this is also a red world transaction.

But here is the important point: In both examples, the red money (always linked to collateral) has a link to a ledger. The ledger is held by a third party who retains a record of how much collateral has been pledged and retains a right to that collateral should the red money not be repaid.

Now you might consider this question: If the red money is reduced to a "bit of paper", is the link to the third party (probably a bank) and collateral preserved? I would assume the answer would be "yes"

[Jussi]

"The bank decides every individual's red money limit."

"In the red world, the bank will not create $100 red money to let me sell a house worth less than $100 (unless I have other assets)."

I think we should define red money better. It is either prearranged credit line (credit card comes to mind) or it can be case by case "issued".

The latter is IMO problematic from medium of exchange point of view - as every transaction needs third party credit analysis (e.g. whether sold house is fair valued or not).

The former might not be fraud proof?

[Tel]

In the red world, the bank issues red bits of paper worth minus $1 each. If you sell $10 worth of apples, the seller must give the buyer 10 red bits of paper.

It doesn't make any difference if people keep their red bits of paper in a cardboard box at the bank, instead of in their pockets. It doesn't make any difference if the bits of paper and cardboard boxes disappear, if the bank keeps a record in a ledger.

I reckon it does make a difference... what with me storing both gasoline and matches in the same cardboard box as the red paper... yeah, could make a real big difference to my bottom line.

Bits of paper, gosh anything could happen!

But what you are really saying is that using the bank ledger is 100% compulsory (for red note holders) because giving away red notes would be crazy, you must get the receiver to sign off on receiving those red notes and be sure to update the bank ledger (and that's what really matters, and ultimately is authoritative). In reality then, the paper notes don't actually exist, the bank ledger exists, and only the bank ledger.

... put a limit on how many red notes an individual can hold, otherwise an individual would want to accumulate an unlimited number of red notes, and die holding more red notes than his wealth.

Certainly, and whatever that limit should be, every individual would be attempting to die with that amount of red notes. Also, whether the red limit is the same for all people, for corporations (I can always whip up a few extra corporations to hold my red notes for me), etc.

The green interest rate (which may be positive or negative) it pays on holdings of green notes. Increasing the green interest rate increases the demand to hold green notes. The red interest rate (which may be positive or negative) it charges on holdings of red notes. Increasing the red interest rate reduces the demand to hold red notes.

And of course, whether those interest rates are equal for all people. For example, regular folks don't get to borrow from the Fed at 0% interest, if I could I'd certainly keep borrowing.

Please don't talk to me about risk premiums, you can have 10x the collateral in assets and still you get the same "market" interest rate that surprisingly isn't 0% interest. That's kind of what you expect from a cartel system... but until economists start to model that, they won't be anywhere near to how a modern economy operates.

[JKH]

Nick,

"In the green world, the bank will not create $100 green money to let me buy a house worth less than $100 (unless I have other assets).

In the red world, the bank will not create $100 red money to let me sell a house worth less than $100 (unless I have other assets).

It looks symmetric."

I agree.

Although here is one nuance:

In the green world, the bank simultaneously creates a term loan and green money. The reason the bank needs collateral is that it is left with the term loan with credit risk after the customer spends green money on the house.

Compare that with the red money case. First, I think I was right in suggesting above that the corresponding banking transaction has to be the simultaneous creation of red money and a term deposit, respectively issued and held initially by the same customer. When he sells his house, that customer transfers his red money to the house buyer.

Unlike the green money case, it is not the bank that created the red money that needs the house as collateral. That bank can use the term deposit as collateral in effect - because it can simply extinguish the deposit liability and write it up to equity if the customer fails to deliver red money at the maturity of the deposit. It's pretty weird.

It is the house buyer's bank that now needs the house as collateral - because it has no such deposit arrangement with the customer who bought the house. That bank only has a red money asset.

[JKH]

Not sure about your clearing house question.

The central bank typically undertakes the clearing house function.

[Nick Edmonds]

"In the green world, the bank will not create $100 green money to let me buy a house worth less than $100 (unless I have other assets).

In the red world, the bank will not create $100 red money to let me sell a house worth less than $100 (unless I have other assets).

It looks symmetric."

In the green world, you do not first need $100 assets to buy a house, because the bank does not just give you $100. They release the $100 against the transfer of title to the house. You are never in a position where you can change your mind and spend the $100 on something else. The bank does not care what assets you hold as collateral to cover the period between giving you $100 green notes and you owning the house, because commercially this period is a fiction.

In a similar way, in the red world the bank does not care what assets the seller has provided that the red notes arise only on sale (which is effectively what happens with overdraft banking). As long as the red notes are only created when the sale is unconditional, the seller is never on the hook for them.

The difference is that in the red world, all sales can be done this way, whereas in the green world it only works for purchase of assets that can be used as collateral (like houses).

(Having written this, I just read JKH's comment on this and I think it might be along the same lines.)

[Jussi]

"if the customer fails to deliver red money at the maturity of the deposit."

Good point. Yet that shouldn't be the case even (respectively in the green world no one is giving out free assets).

"It is the house buyer's bank that now needs the house as collateral - because it has no such deposit arrangement with the customer who bought the house. That bank only has a red money asset."

In the red world money can be created without collateral (if banks are trusted) but transactions need to be supported by collateral. Is red money then a real medium of exchange?

[Nick Rowe]

Tel: In the green world you need a rule that only the bank can create (its) green money.
in the red world you need a rule that only the bank can destroy (its) red money.

In the green world you cannot take green money from someone without permission.
In the red world you cannot give red money to someone without permission.

See my old post I linked in the post above.

[Nick Rowe]

JKH and Nick E may well be right.

The underlying question is this: does the green world economise on information compared to the red world?

My hunch is that the green and red world are perfectly symmetric when everybody trusts everyone else. But the green world requires less trust than the red world.

If my hunch is correct, that would explain why green is more common than red. And it would also tell us that a red/green world is "unstable", and would get more unstable the more red it contains. Because trust can break down.

My hunch is too unclear at the moment.

[JKH]

Just thinking more out loud here about what has become a very interesting thought experiment:

Households in the red world use red money liabilities as the medium of exchange.

Suppose you took all of the residential mortgages outstanding and chopped them up into units called red money and used these units as the medium of exchange.

These units would be collateralized by claims on houses. Divisibility becomes an issue if the bank wants to sell its collateral, but that's a moot point in the context of the weirdness of the entire thought experiment.

Problems happen if people accumulate too much red money relative to their pledged real asset collateral. This can happen if they spend too much and pay for it by receiving more red money in exchange. But they can literally work their way out of this by transferring some of their red money to their employer on pay day.

Red money is always on the move. The question is where does the risk lie?

Somehow the answer must involve interest payments.

Red money pays a rate of interest to the bank.

So the risk is that the bank won't allow a red money holder to pay his interest in the form of more red money that the bank itself would have to create (which creates bank equity at the same time). That becomes Ponzi after a while.  It's a basic bank credit adjudication function. Just that it's on red money as a bank asset.

But its kind of weird to consider what actually becomes the binding constraint in that process. The customer doesn't need to find green money to pay back the bank in a pure red world. What makes the bank say enough is enough?

That seems a strange part in itself. 

Beyond that, and to the degree that there is such a credit risk function involving a highly mobile medium of exchange, such a collateralization attachment must be mobile as well and transferable to successive recipients in the turnover of red money.

..........

[Jussi]

"Problems happen if people accumulate too much red money relative to their pledged real asset collateral. This can happen if they spend too much and pay for it by receiving more red money in exchange. But they can literally work their way out of this by transferring some of their red money to their employer on pay day."

In the pure green world they can also "work their way out" - until they can't. That seems obvious risk and "trust can break down". It looks obvious risk to me. I think Nick's unclear hunch is right.

[Jussi]

"What makes the bank say enough is enough?"

If the banks' asset side (most of it is red money) doesn't move in value, banking seems too weird business - can they charge interest premium (red money interest minus term deposit interest) from reverse maturity transformation? I think a green bank would emerge and take over all the business.

"Beyond that, and to the degree that there is such a credit risk function involving a highly mobile medium of exchange, such a collateralization attachment must be mobile as well and transferable to successive recipients in the turnover of red money."

Having such a mobile collateral system it would be more straightforward to use the assets as medium of exchange.

[Market Fiscalist]

@JKH,

I have a number of questions/comments on your comment:

- "Suppose you took all of the residential mortgages outstanding and chopped them up into units called red money and used these units as the medium of exchange." If I understand this then in theory a bank could create red money via red OMO - "buying" illiquid debt like mortgage debt, and creating liquid red money that people could use for selling stuff. Unlike green OMO where the loan as an asset is bought in this case the loan as a debt is involved.

- "So the risk is that the bank won't allow a red money holder to pay his interest in the form of more red money that the bank itself would have to create (which creates bank equity at the same time). That becomes Ponzi after a while. It's a basic bank credit adjudication function. Just that it's on red money as a bank asset.". I'm not following why this is any different from the normal green world. Banks will put limits on people who they consider have too high a positive red balance, just as they put limits on people with too high a negative green balance in a green world. Why is the red world different ?

"The customer doesn't need to find green money to pay back the bank in a pure red world. What makes the bank say enough is enough?". The customer doesn't need green money, but will need to find a way to keep their red money balance to a level agreed by the bank (by selling as much as they buy), or the bank will stop accepting any more red payments ,and may even go after their collateral if thy can't cover the interest payments. This is exactly analogous to a green bank bouncing checks and bankrupting people when they exceed their overdraft limits and fail to reduce it fast enough.

So overall I think the red and green worlds are rather similar. In an electronic-only banking world they would just be 2 identical ways of looking at the same set of transactions. In both worlds no transaction goes though unless the buyers bank approves, the only difference is the direction of the payment and how balances are thought of (red where debt is recorded as positive, or green where debt is recorded as negative). In a world of hand-to-hand currency things are more awkward because people have to hold debt (rather than assets) in a liquid form. This is challenging for banks as they almost certainly will have to charge interest to hold "liquid debt" red money and this is complicated. (In a world of deflation where the real interest on cash is positive, banks might actually prefer red money that pays only 0% ?). And unlike in a green world where you may assume your buyers cash is good, in a red world you will need to make sure the buyers bank will accept the additional red money before competing the sale, and if the red notes pay interest your bank will have a stake in that decision as well

Finally , it strikes me that no-one has to be a net debtor in a red money world. Everyone could have negative red money balance and it still works. If people with negative red money balances at the bank need red cash - they arrange with the bank to "lend" it to themselves, and in return the bank increases their negative red money balance further. The banks will charge more interest on the loan than they do on the negative balance so will make a profit on people's. .

[Majromax]

@Jussi: In a green world, people are never "in" to have to work their way out. Financial liabilities in a green-only world do not circulate, so someone in particular accepts the counterparty risk.

We can kick-start a green-currency world by having a central bank issue pieces of paper in exchange for a real asset. This is exactly how we got to paper money from a specie system, via gold and/or silver certificates initially redeemable on demand. Green pieces of paper in circulation are liabilities of the central bank, against which it holds assets.

We can't kick-start a red-currency world in the same way, since nobody would be willing to take an initial allotment of red notes. To get to circulating red notes from nothing, we need to force someone to take red notes. This can happen either if the central bank has an initial endowment of real assets that people want (giving us our general store ledger), or it can happen if the central bank is backed by a government (giving us a chartalist "money is for taxes" view.)

Another asymmetry is that nominal wealth is unbounded in a green money world, but not in a red money world. I can always accept more green money, and it is always to my benefit to do so. However, my nominal wealth in a red money world is given by my capacity to accept red notes, and that capacity cannot increase without a transfer of some illiquid asset for collateral.

That collateral limit ultimately means that we can't have a red-money-only world. Something else has to circulate as collateral.

Imagine that I'm a red-money Warren Buffet, and I wish to make a penniless person fabulously wealthy. I can't do this: I can take any red money they're carrying with them, but I can't through purely monetary charity give them any additional capacity. I can achieve my aim if I also give him a binding promise to accept red money in the future, but then "a promise to accept red money in the future" is a securitizable asset that could circulate like green money -- and we're suddenly in a red/green world.

The equivalent to the collateralized red money world is a green money world where there is a hard limit on the amount of currency that one can hold.

[Majromax]

@Fiscalist:

> In both worlds no transaction goes though unless the buyers bank approves, the only difference is the direction of the payment and how balances are thought of

That's only true if debt is involved in the green-money transaction. If the paying party has green notes, then there is no risk that the buyer's bank can disapprove of the transaction. Getting into the nitty-gritty, the risk with debt is that the buyer's bank refuses to issue new green notes on the spot to the buyer.

In the green money world, everyone holds a strictly nonnegative value in the most-liquid asset.

In the red money world, everyone holds a strictly nonpositive value in the most-liquid asset.

I think this is ultimately contradictory, since "most liquid asset" means the thing that can be used for transactions with the least encumbrance. But needing the bank's approval for each transaction is by definition an encumbrance: it is possible to increase one's liquidity in a red-money system by issuing tokens authorizing the transfer of red money on demand against the issuer's credit limit.

[JKH]

I'm a bit cautious on the notion of symmetry between the green and red worlds. But symmetry is always a great question for the testing of thinking.

Green banks are organized to transform credit risk - from risky loans to less risky green money. Any notion of credit risk on green money is oriented to the natural reduction of it via green banking.

The nature of credit risk on red money is still not clear to me. Red money is a red bank asset. But that doesn't mean there can't be red bank loans. And the credit risk on those loans intuitively should be the same, independent of the medium of exchange.

So the full balance sheet structure for the two types of banking could be quite different. One would think that would open the door to some non-trivial asymmetries somewhere along the line.

[Market Fiscalist]

Majromax,

"In the green money world, everyone holds a strictly nonnegative value in the most-liquid asset.". Isn't an overdraft in a green world a negative value in the most-liquid asset?

[Nick Rowe]

Good thinking out loud from JKH.

[JKH]

Additional "thinking out loud" - way out loud.

When the bank earns interest on an asset which is the medium of exchange, that is inherently Ponzi in the bank's favor. The only way interest on red money can be paid with red money is to increase the red money balance by the same as the amount of interest. That's obviously not usually the case when a bank is paid green money interest on a regular loan. In that case, the payment of interest on a loan destroys a green money deposit. There is an effect similar to red money interest compounding in the case of interest paid on green money deposits. But the net loan/deposit interest margin effect is a destruction of green money and an increase in bank equity. So net green money Ponzi is precluded by the profitability of green banks.

Something must constrain bank Ponzi in the red world.

And I guess it must be regulatory/credit limits.

Maybe the true underlying risk in all of this is inflation risk at the macro level.

I'm not sure how individual risk would become binding on banks otherwise given the open ended Ponzi opportunity for both banks (earning compound interest) and their customers (spending like crazy by accepting red money in exchange). It's not even clear to me how the rationale works for collateral seizure and sale when limits are busted.

[Nick Edmonds]

I'd agree that trust is at the heart of this.

Assuming a single monopoly bank, a payment is a tripartite arrangement involving payer, payee and the bank. The advantage of green money is that the bank is indifferent to the payment, so it need impose no restrictions on payer and payee. Green notes therefore have a negotiability that is one of the key benefits of money.

Some degree of negotiability can be achieved with red money, but the bank will want to impose limits. Payer and payee may be free to novate red money, provided neither ends up holding more than an amount specified by the bank. These limits are crucial in the red world, but are not needed in the green world.

[JKH]

I normally try to read other comments closely and respond where appropriate. But this topic is sufficiently challenging and mind blowing that it requires substantial up front investment just to try and work out one's own starting views.

[JKH]

Nick E.,

I lied. I just read your comment.

"These limits are crucial in the red world, but are not needed in the green world."

In fact, banks are known to have deposit concentration limits foe individual customers - a liquidity risk mitigation exercise.

As a corollary to that, I suspect the academic economics community would be surprised to know just how critical liability management is to bank liquidity management.

[Min]

Nick Rowe: "not all financial liabilities are red money (just as not all financial assets are green money). It has to circulate as a medium of exchange to count as money."

Right. I'm still waiting.

[Nick Edmonds]

JKH,

That is true. In the interbank, banks will also sometimes restrict borrowing from specific counterparties so as not to tie up lines. But I think that goes way beyond the issues here.

[Oliver]

Is my understanding correct that you are distinguishing between money which enters the economy via an overdraft / borrowing and money which enters via an asset purchase by a (central) bank?

If so, then some thoughts:

If I borrow money via an overdraft, then I have green money plus a red debt (the money I borrow is not red, at least once it changes hands. The payee couldn't care less whether I borrowed money to pay him / her or whether I earned it). So I pass on the green money in exchange for something of equal value but the debt remains between me and my bank. I can only get rid of my debt by repaying the bank and in doing so, cancel out both my debt as well as the corresponding asset of the bank. To the extent that I pay back with green money, the banks is left with green and red money which cancels out / disappears, thus diminishing the stock of gross money. To the extent that I pay with another asset, say a house, the red money disappears but the green doesn't, so gross money is displaced by net money.

I can see how asset purchases allow the bank to exogenously determine the amount of net money extant. And I can see how that would be important from a quantity theory perspective. But I'm personally not (yet) convinced it's that important for the economy. A decreased demand for monetary savings is more than offset by the increased demand for non-monetary savings.

The other thing that struck me, which you mention two posts before, is that you do not distinguish between things that banks buy. I assume that holds in the red & green world as well? Government bonds, corporate papers, real assets (say bridges) as well as consumption goods and services are all the same. What matters is the total stock of money, red + green.

What would that mean for the banks, I thought? Real world banks can only expand their balance sheets by purchasing financial assets. For most central banks, purchases are restricted to assets of their owners. No banks own and run bridges or highways, nor do they issue new money for staff haircuts or crates of bananas for board meetings. The simple explanation for the latter two examples being, imo, that potential returns do not accrue to them / they would make a loss. Banks don't tax, nor do they collect tolls or other revenue. They do earn money with financial assets they purchase, though.

So, financial asset purchases mean banks earn money with existing assets while expanding their balance sheet. The earning would otherwise have accrued to the non bank sector. Other purchases mean they lose money, all else equal. Red money / credit means banks make money with new assets (the loan did't exist before). So, even from a quantity theory perspective, there is reason to argue that what banks buy, matters from both a stock as well as a flow perspective.

I seem to have veered off topic. But I hope it makes some sense nonetheless.

[Nick Rowe]

I've learned a new word: "Verb 1. novate - replace with something new, especially an old obligation by a new one"

JKH: "In fact, banks are known to have deposit concentration limits foe individual customers - a liquidity risk mitigation exercise.

As a corollary to that, I suspect the academic economics community would be surprised to know just how critical liability management is to bank liquidity management."

I'm surprised. But it makes sense now you say it.

But that's liquidity risk, whereas the main risk we are worried about in red and green (a)symmetry is (I think) default risk.

Aside from the purely intellectual side of this discussion, there is also (I think) a fairly immediate real-world application: most clearing houses (I think) run on a red/green system where net money is very small (maybe zero) relative to gross money. But if the participants fear that one party may hold too many red notes at the end of the day, the whole thing could seize up.

[JKH]

Nick E., Nick R.,

Quite right. Somewhat peripheral.