What is a "cashless" economy anyway?

I was reading Josh Hendrickson's (good but difficult) paper about Jurg Niehans on the cashless economy (which predated Michael Woodford by a couple of decades). But I can't get this question out of my head.

Tell me exactly where it is on my slippery slope that this economy becomes "cashless", and tell me why it matters.

1. People use paper currency issued by the central bank to buy and sell everything. If you like, you can let commercial banks issue paper currency too, convertible on demand into central bank currency, but I don't think it changes my argument. The central bank changes the stock of currency by buying or selling bonds (Open Market Operations, or "QE" if you like to use the new word for the same old thing).

2. Muggers become a problem, so people put their currency in a box at the central bank with their name on it. It's safer than keeping it in their pockets or under the mattress. If Andy buys $10 worth of apples from Betty, he gives Betty a signed letter ("cheque") instructing the bank to take $10 currency out of his box and put it in Betty's box. When email gets invented the letter is sent electronically instead.

3. The bank notices it is now administratively easier to pay (positive or negative) interest on currency than it was when people kept their currency in their pockets. So the bank now has a second monetary policy instrument, in addition to Open Market Operations.

4. A rumour spreads that a fire at the bank has destroyed all the currency. The bank denies the rumour, but says it wouldn't matter anyway, because the bank keeps real-time back-up records of how much currency is in each person's box, so it can always replace it if there really is a fire.

5. The bank admits the rumour was true, but says it has replaced the currency.

6. The bank admits that it hadn't bothered to replace the currency, since it would have been a waste of paper and ink. The back-up records are all it needs.

7. The bank lets people hold negative amounts of (virtual) currency in their (virtual) boxes. Provided they have sufficient collateral, it will not bounce their (virtual) cheques. The bank charges a higher interest rate on these "overdrafts" than it pays on positive balances. The difference is called the "spread". The bank now has three monetary policy instruments (Open Market Operations plus two interest rates). It could have a fourth instrument if it decided to vary the amount of "haircuts" on collateral.

8. The bank uses Open Market sales of bonds to reduce the total amount of net positive balances to zero, so the sum of the positive balances equals the sum of the negative balances.

9. The bank decides to reduce the spread to zero. It pays the same interest rate on positive balances as it charges on negative balances. (8 and 9 together mean the bank now earns zero revenue from its monetary operations, unless it charges a service fee on transactions, because it pays the same amount of interest on positive balances as it receives on negative balances.)

10. A fairy with an inordinate fondness for representative agent models waves her magic wand to make this economy conform to her favourite model. Each agent produces a different variety of fruit, but there is perfect symmetry, so each agent's inflow and outflow of cash (may I still use that word?) is perfectly synchronised, and each agent always has a zero balance at the bank. If Andy buys $1 worth each of 99 different fruits, 99 other agents are each buying $1 worth of apples from Andy at exactly the same time.

So, where exactly on my slippery slope did this economy become "cashless"? My answer is that it didn't.

Even though at step 10 each individual holds zero cash in equilibrium, an exogenous increase in the rate of interest paid on cash will cause each individual to plan to accumulate cash balances, by spending less cash than he expects to receive from others. But those plans and expectations are out-of-equilibrium plans and expectations. In a representative agent model each individual will be surprised to learn that his receipts of cash have fallen by exactly the same amount as his expenditures of cash. And exactly the same would be true if each individual held positive cash in equilibrium.

The only point on my slippery slope where the Fiscal Theory of the Price Level has any bearing whatsoever is in step 9, where seigniorage revenue goes to zero.

If you want to talk about a world where bonds are used as a medium of exchange and medium of account, then I will reply that your "bonds" are really "cash".

If you want to talk about a barter economy, where any two people can directly swap any fruit with any other fruit, so with n types of fruit there are (n-1)n/2 markets, then OK, that really is a cashless economy. But that doesn't look like where our economy is headed.

Comments

[Mike Sproul]

Nick:

I get that you've just replaced paper dollars with checking account dollars. (Some people might call that cashless.)

But you pulled the Fiscal Theory of the Price Level out of nowhere. As I understand the FTPL, the government's future tax receipts originally backed the paper dollars, and now the tax receipts back the checking account dollars. So the FTPL has a direct bearing throughout.

[Matt Youngh]

Right, we are going to a pure cash system. In a pure cash system the smart credit card simply emulates ink and paper. It makes exchanges with other smart cards, being honest about the digital denominations. Exchanges are encrypted, the smart card is counterfeit proof, the banks guarantee honest accounting in the card chip. The LCD displays the denomination exchanged, etc. Simply pure digital cash emulating every aspect of ink and paper. Digital does to paper cash exactly what digital did to paper files, an exact emulation, no function or feature left un-emulated.

[Market Fiscalist]

Take 2 scenarios:

A: Everyone has an account at the CB but cannot hold cash. Each period their balance goes up or down depending upon their net spending but in aggregate the total of balances remains the same. If the CB wants to change the volume of spending it can do so by either 1) varying the interest rate it pays on balances while leaving the total of balances the same or 2) Using OMO to buy/sell assets and adjusting people's balances.

B: The economy runs on cash issued by the CB as a result of OMO - bank accounts don't exist. In this scenario if the CB wants to change the volume of spending then it can only use OMO. By varying the amount of cash both the volume of spending and the interest rate will vary (and the CB may even use the interest rate as a target). However ultimately the CB only has one tool - OMO.

SO: I would argue scenario A is cashless. Interest rate alone could be used to control aggregate spending. scenario B is cash-only - adjusting the quantity of cash is the only tool. I think under careful examination the differences might be more apparent than real , but I think the comparison is interesting to think about as real economies might be a combination of both.

[Nick Rowe]

Mike: The main reason I mentioned FTPL is because it is central to Niehans' approach, according to Josh's paper. It was just an aside in mine. But let me try to bring it in.

In steps 1 to 8 the central bank is earning profits, so there is a link between monetary policy and the government budget constraint (assuming the government owns the central bank, so gets those profits). But in step 9 those profits drop to zero (except for transactions fees), so there is no link at all between the government budget constraint and monetary policy. The central bank's monetary assets (negative balances) and liabilities (positive balances) cancel out.

Matt: yep. It's really all the same, whether it's paper, plastic, or silicon. Except for administrative questions about which is easiest.

MF: well, there's no reason in principle why paper currency can't have a positive or negative interest rate. It's just a bit more of a hassle to administer. Gessel's stamped money is one example of negative. Or hold a lottery on serial numbers would be an example of positive.

[James Hudson]

Usually your thoughtful and insightful posts are too advanced for me to attempt a comment. But on this post let me offer the following: 'collateral' has two l's.

[Gene Callahan]

Hmm, James, you just spelled it with three l's! (Which I think is correct.)

[Nick Rowe]

James/Gene: Ooops! I thought that was the UK/Canadian spelling, but Google tells me it's Spanish. Fixed.

[Nick Edmonds]

Isn't a "cashless" economy just one where there is no financial asset for which the nominal interest rate is necessarily zero?

[Nick Rowe]

Nick E: well, I suppose you *could* define it that way, because the ability of the central bank to vary the rate of interest it pays on its monetary liabilities is something that does matter. But is "cashless" the right word? Would we say that Silvo Gessel's stamped money plan abolished cash?

BTW, "cashless" sometimes gets translated as "moneyless", and I certainly would not say that an essential feature of money is that it pays 0% nominal.

[Jordan Charbonneau]

Nick, I see the point you're trying to make, but I think you're overcomplicating things. What I imagine most people think of as "cash", and what I think Hendrickson means by "base money", is the paper/plastic bills and metal coins in our pockets/wallets/purses/etc. If everyone used debit cards instead, that would be a cashless economy. Of course, money in the M1+ sense still exists.

[Nick Rowe]

Jordan: Maybe. But in my Step 10 economy, for example, M1+ (or M whatever) is zero. Some would claim it's not just "cashless", but "moneyless". And I'm saying it is not a "moneyless" economy.

[Max]

Imagine a utopia where it doesn't cost *anything* to buy and sell financial assets. You would not hold any cash unless cash paid exactly the same interest rate as non-cash, since a cash balance provides no liquidity service. That's a "cashless" economy. Clearly it's not something that could ever exist, it's just a thought experiment, or a way of simplifying economic models.

[Nick Rowe]

Max: In my Step 10, nobody *holds* any cash. Flows in are perfectly synchronised with flows out. But (I claim) this is not a cashless economy.

[Trevor]

If people's cash flows are perfectly synchronized then isn't that just barter? I mean in real life not in an abstract model where you can just will anything to happen. If I really hold no cash at any given time then I'm just trading goods with people in some giant economy wide trade to maintain that because I can't trade with one person and then at another time trade with another if I'm using cash as the medium of exchange. Making people arbitrarily use cash when they also have to make these giant double coincidence of wants trades to maintain a balance of 0 cash just seems like barter with the arbitrary accounting requirement that everyone has to trade in cash.

[Min]

"So, where exactly on my slippery slope did this economy become "cashless"?"

Answer:

"1. . . . The central bank changes the stock of currency by buying or selling bonds"

Objection:

"If you want to talk about a world where bonds are used as a medium of exchange and medium of account, then I will reply that your "bonds" are really "cash"."

Response:

Flexible definitions let you say anything. You have not distinguished among "cash", "currency", and "money". Besides which, money is not a single thing, either, as we know.

Technically, "money" in particular is a fuzzy concept (as in fuzzy logic). OTOH, "cash"is not; it refers to physical coins or notes, as in Stanley Freberg's song, "Dey done took away my credit cards, Lawd, Lawd, and now I gots to pay cash on de line."

[Nick Edmonds]

I don't think we would say that stamped money abolished cash, because we usually simply take cash to mean bearer money. But, I do think that if all cash was stamped, then the economy would behave in the way that we think a cashless economy would behave. Maybe would then say that it was an as-if cashless economy.

So, possibly "cashless" is the wrong word, but that's how I understand the implications anyway. I don't think anything rests on the distinction between bearer and registered form.

[Max]

Nick, it's cashless in the sense that there's no money multiplier. The central bank just sets the quantity of money at $0.01 and leaves it there forever. And it still sets the price level at whatever (because it controls the unit of account), and sets the interest rate by paying interest on the $0.01.

[Max]

Whoops, that was for Nick Rowe, not Nick Edmonds, in case that wasn't clear.

[Frances Coppola]

Nick, "cashless" and "moneyless" are not the same thing, at least in common parlance. I am nearly always cashless, because I pay for everything electronically, but that doesn't mean I don't use money. You seem to be conflating the two.

Trevor, timing differences are not the only reason why people trade using money. They also do so in order to agree price. Note that in step 10, Andy is trading apples for other fruit. In the absence of some kind of agreed unit of account, whether an apple is worth 1 orange or 10 has to be renegotiated for every transaction. That really is the problem with pure barter systems. Lack of a common pricing standard.

[Frances Coppola]

Following on from that: if all goods can be swapped one for one, money is not required. But such a society would have no means of establishing value. I'd suggest that this would only happen in superabundance. As soon as there were scarcities - real or, more likely, artificial, human nature being what it is - some means of establishing value would re-reemerge, and that would become "money", though not necessarily "cash".

[Nick Rowe]

Max: "Whoops, that was for Nick Rowe, not Nick Edmonds, in case that wasn't clear."

On this blog, there's only one "Nick", which is me. Valued commenter Nick Edmonds in "Nick E".

(Also, there's only one "Frances", which is Frances Woolley. Valued commenter Frances Coppola is "Frances C".)

:-)

[Nick Rowe]

Frances C: "Nick, "cashless" and "moneyless" are not the same thing, at least in common parlance. I am nearly always cashless, because I pay for everything electronically, but that doesn't mean I don't use money. You seem to be conflating the two."

We are basically in agreement, except it is not me who is conflating the two. I am worried that some others (not you) are conflating the two. (E.g. people who say that Woodford's "cashless" economy is an economy without money.) If you want to make a distinction between (physical) "cash" and (non-physical) "non-cash money" I am OK with that. But I think it is a distinction without any relevant difference. Which is the main point I was trying to make in this post.

"Following on from that: if all goods can be swapped one for one, money is not required. But such a society would have no means of establishing value. I'd suggest that this would only happen in superabundance. As soon as there were scarcities - real or, more likely, artificial, human nature being what it is - some means of establishing value would re-reemerge, and that would become "money", though not necessarily "cash"."

I agree with your first sentence there, but disagree with the rest. We do see some barter in the real world, and we can imagine a world where all trade is barter trade, and in which the (relative) values of goods are determined in exactly the same way as they are now (by supply and demand, or whatever). (And in a world with only 2 goods, money would be pointless anyway.) We might see some good emerging as the unit of account in a barter economy (I say my apples are priced at one carrot each, you say your bananas are priced at 2 carrots each, then I swap 2 of my apples for 1 of your bananas.) But there would be no medium of exchange, by assumption.

[Nick Rowe]

Trevor: "If people's cash flows are perfectly synchronized then isn't that just barter?"

I don't think so. Imagine n people sitting in a circle. At exactly the same time, each person buys goods from the person on his right, and pays for it with money. Trade is pairwise; only 2 people need to agree to each trade. Without money, all n people would have to agree to the n-good trade. It's like the UN, only with the requirement that all votes must be unanimous.

Your example assumed 2 people with 2 goods, where it is true that money is not needed.

[Nick Rowe]

Min: "Flexible definitions let you say anything. You have not distinguished among "cash", "currency", and "money"."

I am saying that any such distinction is unimportant for monetary theory. In different circumstances, different forms of money may be more or less convenient. That is all.

Nick E: " But, I do think that if all cash was stamped, then the economy would behave in the way that we think a cashless economy would behave."

I don't get that. Compare two economies. One has a 5% inflation target and 0% nominal interest on money. The second has a 0% inflation target and negative 5% interest on money (stamped money). Both should have the same opportunity cost of holding money.

Max: "Nick, it's cashless in the sense that there's no money multiplier."

I don't get that. There are no commercial banks in my story, so there is no money multiplier. All money is base money. (Or, you could say the money multiplier is always one, by definition).

[Roger Sparks]

Nick,

At item 7, you introduce "negative amounts of (virtual) currency in their (virtual) boxes". This seems to be a substitution for "Open Market Operations" which were introduced in item 1.

I believe that the relationship between money and bonds is like the relation between water and ice.

If this relationship is correct, what is the purpose of introducing "negative money" when depositors could just as easily have "positive money" coupled with a bond (as introduced in item 1)?

Finally, who are the depositors in the banking system that have deposits measured in negative numbers?. The customers of banks ALL consider their deposits to be a positive number (someone had to work to earn that positive number).

[Scott Sumner]

Excellent post. I recall that Bennett McCallum used to complain that Woodford's "money-less economies" actually had money.

In step 9, what assures that the negative and positive balances are equal once interest rates are unified. Consider this extreme example, which makes my point easier to see. Suppose almost no one has collateral that the central bank finds acceptable. How can we assume that desired positive and negative balances are equal at the same interest rate?

(I'm not yet assuming the representative agent model.)

As a practical matter, when the cashless nirvana arrives we'll probably have a regime with positive net balances, and hence the quantity theory still holds.

[Nick Edmonds]

The point about the nominal rate on money being necessarily zero, is not that it is zero, but that it is a rate that is fixed outside of the system. If stamped money was always stamped at negative 5%, a rate that was determined by physical considerations, say, it would be comparable to having cash with a necessarily zero rate. The difference comes when you have a stamp rate that can be set as a policy rate or determined in the market, i.e. where there is no instrument with a rate that cannot be set this way.

[JF]

In point 8, the sale of a bond by the CB is used to remove excess cash.

Where did this excess come from to make point 8 necessary (and why are they removing it?)? If someone's incoming = some others outgoing, in the aggregate, that sounds like there is no growth, just turn-stiling, and no added cash would even need to be withdrawn.

But someone, somewhere creates extra money because there is a growing economy with many more goods and services and stable prices (not wacky hyper-inflation or hyper-deflation or de-population from warfighting), or are you just assuming that prices adjust and velocity/speed just covers everything up.

Endogenous money creation by the banking sector is real (assuming the CB is exogenous). I think you need another Point in there, or several points in there, to explain when and how and why the CB injects more cash/money as an economy grows, in a stable fashion.

After 2007's Minsky it seems that we really shouldn't ignore the financial community in economic models.

[jonathan]

(I *think* the below is right, but I wouldn't bet too much on it)

First, I think it's important to distinguish between Woodford's cashless *limit*, and a cashless economy. The former is a model with base money -- it just takes the limit as the quantity of base money approaches zero. The central bank still implements its interest rate target by adjusting the (infinitesimally small) quantity of base money.

By contrast, a cashless model is one where transaction technology has completely eliminated demand for base money. I *think* that this happens at step (9) of your described slippery slope to a cashless economy. When there is no difference between positive and negative balances, people don't care how much they have in their accounts, only their total wealth. This is the point at which the quantity of base money stops mattering, and open market operations have no effect.

Of course, monetary policy can still work by changing the interest rate on reserves (which equals the interest rate on bonds). Also, if you think that bonds are just money in this economy (basically true), then the fiscal authority can raise the "money supply" through new borrowing.

[Max]

"There are no commercial banks in my story, so there is no money multiplier. All money is base money. (Or, you could say the money multiplier is always one, by definition)."

Yes, wrong term. I was thinking of the V in MV=PY. M can be infinitesimal for any PY, so varying M has no traction. Which also means that the central bank wouldn't really be a bank. Its job would only be to manage the unit of account, not to produce money.

[J.V. Dubois]

Very good article. However I have some proposal of what cashless can mean. Imagine following situations:

1. I am Canadian buying stuff in USA. Prices are in US dollars so US dollar is medium of account. However I pay by credit card so my bank first has to sell my Canadian Dollars for US dollars and then hand those to the buyer. What is medium of exchange here?

2. Imagine that everybody wears google glasses, those that can translate stuff in real time. Except that these glasses will be programmed so that I will see all prices in all currencies automatically as if they are in Canadian dollars. What is "medium of account" from my perspective?

3. Imagine that people eventually find out that they don't have to use currencies. I can actually pay on gold. Or wheat. Or oil. The way it works is that I have my "crude oil" account because I really like crude where at all times I hold certificates for oil in warehause. When I want to buy something my bank sells these certificates on the market and buys whatever the opposing side wants: dollars, euors, bitcoins, gold or whatever. I program my google glasses so that I see everything priced as gallons of crude oil.

The last is I think true cashless economy, it is closest emulator of barter. What was the crucial step that changed it from our own economy in point 1?

[Nick Rowe]

Roger: "At item 7, you introduce "negative amounts of (virtual) currency in their (virtual) boxes". This seems to be a substitution for "Open Market Operations" which were introduced in item 1."

I don't think so. Suppose you and me both have $0 in our accounts. Everyone else has +$100. Then I buy $10 worth of apples from you. Your account goes to +$10 and mine goes to -$10. No OMOs were harmed in making this negative money.

Scott: Thanks!

"In step 9, what assures that the negative and positive balances are equal once interest rates are unified. Consider this extreme example, which makes my point easier to see. Suppose almost no one has collateral that the central bank finds acceptable. How can we assume that desired positive and negative balances are equal at the same interest rate?"

Fair point.

Assume interest rates on positive and negative balances are the same. For a given inflation target (or NGDP target) there will be an equilibrium curve between the net stock of money (positive minus negative balances) and the rate of interest paid on money. A tightening of colateral requirements shifts that curve to the right (bigger net money stock for a given interest rate). With extremely tight colateral requirements, that trade-off cannot cross the zero net balances axis, so it will be impossible for the central bank to make net balances zero using OMO, while keeping inflation on target.

Nick E: OK, I agree that the ability of the CB to choose the rate of interest paid on money does make a qualitative difference to monetary policy, since it now has a second instrument.

JF: You lost me. The stock of central bank money cannot grow unless the central bank lets it grow, by doing an OMO, for example. (And just because an economy is growing doesn't necessarily mean the stock of money must be growing, because the price level could be falling, or velocity could be rising.)

jonathan: good comment. I'm not sure whether you are right either. Let me try to think it through:

Define "gross money" as the sum of absolute values of positive and negative balances, and define "net money" as positive balances minus negative balances.

The central bank can always set net money equal zero, by using OMO (as long as collateral constraints aren't all binding). But a reduction in transactions costs would reduce gross money, for any given level of net money.

I need to mull this one over, but it's related to my answer to Scott's question above.

Max: Ah! I misunderstood you. Yes, "money multiplier" can mean dY/dM as well as dM/dMB. (I think we normally call the first the monetary policy multiplier and the second the money supply multiplier.)

Well, in the limit, as V goes to infinity, dY/dM goes to infinity too. But that is holding constant the rate of interest paid on money.

JV: Thanks!

By the time we get to your step 3, I think we are talking about something very close to a barter economy. One way to define a barter economy (paradoxically) is to say that in a barter economy *all* goods are money. And in your step 3, certificates for lots of different goods can be used as money.

Except: "When I want to buy something my bank sells these certificates on the market and buys whatever the opposing side wants..."

Be explicit. What *exactly* is your bank selling those certificates for, and what exactly is it buying those other things with? Only if it's a direct swap is this barter.

[Kaleberg]

When people talk about a "cashless" society they are talking about a society in which no one deals with paper money or coinage. There was a charming article on this in Business Week back in the late 60s complete with an illustration of daily life without "cash", but including a beggar asking for spare "electro-credits" and holding a device not all that unlike the portable credit cards readers popular outside of the US.

Economists may have other definitions for "cash" and "cashless", but they have nothing to do with the popular definitions of the terms. It's sort of like arguing with a linguist about whether sign language has "words" or not or getting stuck arguing with a CS major who keeps insisting that it doesn't matter if it is a Mac or PC because everything is Turing computable.

[Gary Anderson]

I have been reading the comments on this thread and I can say with certainty that a cashless society is dangerous to anyone who gets in trouble with his bank, who has to put food on the table for his kids, or is traveling in a foreign nation. The risk of error due to a number of reasons is so great as to be something that should concern rich and poor alike, although the most vulnerable will be the poor or people with toxic loans. Knowing this, people may borrow less and retreat, rather than expand their credit.

I honestly think that economists have become apologists for banks, formatting all the positive reasons why we should have a cashless society. But they are doing their fellow man a big disservice, IMO.

[J.V. Dubois]

Nick Rowe: "Be explicit. What *exactly* is your bank selling those certificates for, and what exactly is it buying those other things with? Only if it's a direct swap is this barter."

Of course your Bank will buy electronic version of whatever commodity or commercial paper the opposing side needs. It may be Mitsubishi corp. stock, paper gold, electronic voucher to get free "iPhone 6SP lus 16GB, black color" in any Apple Store or whatever and sends it over to their bank. Now we may argue if it is really a barter if noone physically touched those goods upon exchange, but I say that it for sure is cashless exchange. Unless you have a weird definition that there are tens of thousands of types of cash in the economy (which you may very well apply to barter economy as well)

[Nick Rowe]

JV: If I directly swap my apple certificates for your banana certificates, that is barter.

If I sell my apple certificates to Harry for dollars, and then a minute later buy your banana certificates for dollars, then dollars are being used as the medium of exchange.

The subsequent swapping of 100 apple certificates for 100 apples is just delivery on an IOU.

Gary: Margaret Atwood's "Handmaiden's Tale" had a great example. First they abolished currency, then took away women's credit cards, so they had to use their father's or husband's. But this post is not about the desirability of abolishing currency.

Kaleberg: "Economists may have other definitions for "cash" and "cashless", but they have nothing to do with the popular definitions of the terms."

I think that's correct. And accountants also have a very different definition of "cash". IIRC, it's any safe promise to pay that has a term of less than one year. But I'm focussing on what definition is appropriate for macroeconomists. What's a useful definition for one is useless for another.

[Min]

Moi: "Flexible definitions let you say anything. You have not distinguished among "cash", "currency", and "money"."

Nick Rowe: "I am saying that any such distinction is unimportant for monetary theory. In different circumstances, different forms of money may be more or less convenient. That is all."

If you are claiming that there is a slippery slope between a cash economy and a cashless economy, you need to distinguish between the two, or else there is no slope, slippery or otherwise.

[Tom Brown]

Nick, on Nick Edmond's point again: Are there any examples in the world of your step 3 happening (i.e. cash being completely replaced with something bearing a policy-determined & thus potentially time-varying nominal interest rate)?

[Nick Rowe]

Min: I think it's the other way around. My slippery slope forces us to think about what (if anything) "cashless" might (usefully) mean.

Tom: not that I know of. It's a matter of degree. But the Bank of Canada does have two policy rates: the deposit rate on positive balances and the bank rate on negative balances. (Similar for other central banks.)

[Min]

Nick Rowe: "I think it's the other way around. My slippery slope forces us to think about what (if anything) "cashless" might (usefully) mean."

You still need to define your endpoints. :)

One way to look at what you are doing is to delineate a fuzzy concept, "cashless". There is not a clean distinction between a cash economy and a cashless economy. You say, "Tell me exactly where it is on my slippery slope that this economy becomes "cashless". Given your point of view, the answer is nowhere. There is no exact boundary. The boundary is fuzzy. But there is not a continuum between the two, either. You identify different stages between the two, which is one way of describing a fuzzy boundary, one which is appropriate when the boundary is not a continuum. But when you are describing or exploring such a fuzzy boundary, you still need to be clear about your endpoints.

In history, or even prehistory, I don't think that there has been a pure cash economy, but we can say what it would be. All trade involves special tokens called cash that are exchanges for goods and services. This cash may or may not have value outside of exchange. There are cashless non-civilized societies in which people exchange goods and services in a reciprocal manner, but the exchange is typically not immediate, nor is there a well defined tit for tat. In civilized societies I think that there have been almost cashless enclaves. For instance, in the antebellum American South, neither slaves nor slave owners made much use for cash on an everyday basis. (And much of the slave owners large transactions made use of credit.) Also, on the American frontier, I think that there were small towns that for some time existed without much cash. The general store extended credit to nearly everyone for long periods of time, and would often adjust the ledger according to IOUs between customers, acting like a bank in that regard. So I think that for civilized societies, we can define another endpoint as a pure credit economy. All transaction affect the general ledger, none involve cash.

[Nick Rowe]

Min: "The general store extended credit to nearly everyone for long periods of time, and would often adjust the ledger according to IOUs between customers, acting like a bank in that regard."

Suppose the manager of the general store can't keep records. So he writes each customer's name on a box. If the customer gives him something worth $10, he puts 10 bits of green paper in the customer's box. If another customer takes something worth $5, he puts 5 red bits of paper in the customer's box. If he sees a red and a green bit of paper in a box, he takes both out. Those green and red bits of paper are just like currency. It's all just different ways of keeping records.

See my next post.

[Andrew Baldwin]

Am I missing something? I would have thought that at this particular time in history the crucial distinction between a cashless society and a society with cash is that the central bank can only make its policy rate slide a little below zero in an economy with cash (the Bank of Canada recently estimated that it might take the overnight rate to -0.5%) but could probably go a lot lower if we were a cashless society. To my mind, this would be a good reason to keep cash around.

[Nick Rowe]

Andrew: it is administratively more difficult, but not impossible, to pay (positive or negative) interest on currency. Many economists say that is a good reason to eliminate currency. We could target 0% inflation, if we wanted, without ever worrying about the ZLB.

[Oliver]

Some random thoughts from the parallel universe of French circuit theory:

I think you have to start before step 1 if you want a real 'moneyless / cashless' society. And if you don't, then your 10 doesn't make any sense (to me).

One can specify two ways by which money / cash enters the economy:

1. The first is by (central) bank credit. The debtor sells his (real or unreal) bill to a bank which discounts it and exchanges it for cash by means of which the debtor obtains real goods from the creditor / new money owner.

2. The second is effectively the same, except that the steps are taken in sequence, as opposed to simultaneously, thus turning an originally bilateral financial agreement into a tripartite monetary agreement. The (central) bank buys a specific bill from a creditor after the fact and exchanges it for an unspecific bill of its own as part of an overall liquidity operation. Call it OMO or QE.

If the bills involved are such that they constitute a monetary promise, i.e. promise of a future cash flow from one agent to the other, then money is already implicit in the original bilateral agreement.

For an economy to be moneyless / cashless on the other hand, the bills have to be in kind, e.g. 1 cow today for 1.2 cows tomorrow, as opposed to money's worth of cows.

That is the perspective from which your fairy in 10 operates. But from that perspective, bank operations make no sense (to me), in fact banks don't make any sense. Banks would effectively have to consume or at least keep the products they buy and be able to produce more of the same products in future to repay the creditors upon redemption of their bills. That's not what banks do in my world. The only in-kind relationship involved is that between debtors and creditors, namely at the time when a circuit is opened and again when it is closed. Everything that happens in between is monetary in nature. Money = disequilibrium = when the exchange of real stuff does not happen at the same time.

From the above point of view, and getting to the actual question, there is no distintion between money and cash. And monetary policy is basically the sum of all credit / OMO decisions which, to the extent that more than one bank is involved, can be applied both directly by banks through their respective credit decisions as well as indirectly by the central bank , either by influencing the cost of business for member banks or via the financial markets through liquidity operations. That holds even if there is a 0 requirement for reserve balances. There are other ways of charging banks to do business. But monetary policy in a world without money, i.e. where all bills are in kind, makes no sense to me whatsoever. The other convenient thing about circuit thery is that you don't need separate concepts for red and green money. There is always a creditor and a debtor so they're just two sides of the same operation. I find that more elegant and it squares better with the accounting.

[chuck]

In step 2, is all of the currency in the box or only some of it?

"2. Muggers become a problem, so people put their currency in a box at the central bank with their name on it."

I would put it on a spectrum:

Pure cash: no cost to holding currency. zero bound on rates is hard.

Middle: theft, fires, etc. currency is still a substitute for deposits, but rates can go negative

Cashless: no currency substitute for deposits. unclear if there is a lower bound on rates