Money as store of wealth?

This is something that always bugs me with the textbooks' definition of money.

They start out well. They provide a functional definition of money; "money is what money does". If people in a given time and place use cigarettes as money, then cigarettes are money, in that time and place. Moneyness is not a property of a good; it is a property of how people use that good.

Then they list three functions of money: medium of exchange (we buy and sell all other goods for money); medium of account (we quote the prices of all other goods in money); and store of wealth (or store of value). It's that third function of money that is totally out of place. We should ditch it.

Most goods are not used as a medium of exchange. A good that is used as a medium of exchange is special.

Most goods are not used as a medium of account. A good that is used as a medium of account is special.

(And, with rare exceptions like hyperinflations, the good that is used as a medium of exchange is also used as a medium of account. Because if you are going to be paying with cigarettes it's nearly always a lot more convenient to quote the price in cigarettes.)

But most goods are used as a store of wealth. There is absolutely nothing special about a good that is used as a store of wealth. The chair I am sitting on is a store of wealth. So are the desk and computer in front of me. If I rented them, instead of owning them, my wealth would be lower than it is.

If I were a visiting anthropologist, wanting to discover what Canadians use as money, I would look at what good they use as a medium of exchange, and as a medium of account. That tells me all I need to know. I wouldn't go looking for what Canadians use as a store of wealth. A list of all the thousands of goods that Canadians use as a store of wealth would be an interesting list; but it wouldn't help me narrow down my search for what they use as money.

Now, it's neverthless true that money is a store of wealth. There is nearly always some delay, however short, between receiving money and spending it again. The velocity of circulation is not infinite. If the value of money fell to zero before you could spend it again, it couldn't be used as a medium of exchange. And if the value of money fell too quickly, it would not be a good medium of account either. But being a store of wealth is not what makes it money. That's not part of a useful definition of money.

Money is a very peculiar store of wealth.

How much would I have been willing to pay for my house if the seller had imposed a condition that I could use it as long as I wanted but could never sell it again, or rent it out to someone else? Less than I paid for it, but still a positive amount. It yields a flow of services even if I can't sell that right.

How much would I have been willing to pay for the S20 note in my pocket if the seller had imposed a condition that I could use it as long as I wanted but could never sell it again, or rent it out to someone else? Nothing.

When you buy a good, you buy the option to use it, and the option to sell it again. For most goods, both options have value to the buyer. For a good that is only used as money, the first option is worthless; only the second option has value. Because if I can't sell it again, I can't use it as a medium of exchange.

If the government made it illegal for me to sell my assets, my wealth would drop, but I would still be wealthier than someone who had no assets. I could still live in my house, drive my car, sit on my chair, and use my computer. Even my financial assets would still pay dividends, or might be redeemed at some future point. I wouldn't throw them away. But the irredeemable paper money in my pocket would be worthless to me. I would throw it away. It is only a store of wealth because it is a medium of exchange.

Some people are said to use money for lighting their cigars. But I'm not going to list that as one of the functions of money.

Money is what money does. There are two functions of money that define what is and what is not used as money: medium of exchange; and medium of account. That's it.

This post is a reflection of John Quiggin's post on David Graeber. Here's John:

"So, if we had the kind of disciplinary modesty richly merited by our performance as a profession over the past few years, economists would recognise that we owe an intellectual debt to Graeber. From now on, we can treat money primarily as a store of value, and stop worrying about how it works as a medium of exchange."

I disagree. We need to start worrying a lot more about how money works as a medium of exchange. We need to understand a lot better than we do how money works as a coordinating device in a decentralised economy. And we need to understand a lot better than we do how money can sometimes fail as a coordinating device. Because, outside a very simple economy, people can't barter their way back to full employment if the monetary exchange system fails.

We need to stop thinking of money as a store of wealth, just like all the others. And let's start by changing the textbook definition of money, by deleting that bit about money being a store of wealth. It only serves to confuse visiting anthropologists.

Comments

[Min]

"It's that third function of money that is totally out of place. We should ditch it."

But, but, but what about those huge stones on the South Sea islands? What about those cows in Africa?

;)

[Determinant]

The Bank of Canada has a Yap Stone in its Currency Museum. I am always delighted to visit the Currency Museum because it's free. I always get a chuckle out of visiting a free museum in the Temple of Capitalism.

[Nick Rowe]

That's supposed to be part of the BoC's foreign exchange reserves! One of the great things about Yap stones is they are hard to steal.

[Nathan Tankus]

Yeah Graeber and anthropologists have come to different conclusions about money ergo, the textbook has confused them. That's not condescending at all...

[Lord]

Money is the most liquid store of wealth. Assets change in value, often sharply. Money does so more slowly, at least if it is to retain its value as money. That house may be worth half what it was, but cash is nearly worth what it was.

[Nick Rowe]

Nathan: IIRC (and my memory may be wrong on this) somewhere in the beginning of Graeber's essay he talks about the economics textbook definition of money. Is it "condescending" to think that an anthropologist might look at an economics textbook definition of "money"? And that doing so might confuse him about how economists view (or should view) money? OK.

Lord: that isn't true in hyperinflations. But yes, sticky prices (and intertial inflation), because money is nearly always the medium of account too, do tend to make the value of money more predictable than the value of most individual assets.

If we line up all assets in terms of liquidity, then money is at the extreme end of the spectrum. And because of network externalities, the liquidity race is a "winner-takes-all" race. The winner is used in a qualitatively different way from all the other assets.

[Determinant]

Money doesn't depreciate (much) like other capital goods. Real Estate is like money in that it is a good whose value is not strongly tied to its depreciation. On the other hand my car depreciates ever so predictably.

Part of the problem with real estate is that it is used as a store of value instead of depreciating like other consumer or capital goods. That leads to odd impacts in the financial sector. What if we didn't expect real estate to keep its value and to profit form the sale of residential homes?

[JW Mason]

But this just isn't true. Suppose I expect to need to make money payments in the future. i way to take steps today to ensure I am able to make those payments when they come due. Do I buy another desk? No, of course not! What I do is refrain from buying desks, in order to increase my holdings of (or access to) money.

You are talking about the role of money in what is basically a barter economy. Or, I guess as a monetarist, you are assuming the only demand for money is the transactions demand. but once we acknowledge a precautionary or speculative demand for money, it becomes clear that "store of value" is quite central to its function. ("Means of payment" would be more precise, but we can let that go.)

[JW Mason]

i way s/b I want. (Note to self: type slower.)

[Nick Rowe]

Determinant: paper money doesn't physically depreciate (and if it does, you can swap the note at the bank for a new one), but it does depreciate in price. 2% per year in Canada, and sometimes a lot more in other countries and at other times. Depending on time and place, money depreciates more quickly than some assets, and less quickly than others. Nothing special there.

JW: "Suppose I expect to need to make money payments in the future. [I want] to take steps today to ensure I am able to make those payments when they come due. Do I buy another desk? No, of course not! What I do is refrain from buying desks, in order to increase my holdings of (or access to) money."

Au contraire. I expected I would need to make money payments. So first I bought a PhD ;-). More recently I have bought stocks. I bought the house to avoid making money payments (rent). It's only when I need to make a money payment very soon that I stop buying things, and hold money instead.

"You are talking about the role of money in what is basically a barter economy."

You totally lost me there. I am talking (primarily) about the role of money as a medium of exchange, which is the precise opposite of a barter economy.

[K]

Is it a necessary property of money that you can hoard it? Cause you could easily design a medium of exchange that you can't hoard. Lets say securities trades settle instantly into cash at the ATM, but the cash disintegrates after one hour. Useless as a store of value, can't be hoarded, but fine as a medium of exchange. Is it money? Can it still cause suboptimal AD?

[Nick Rowe]

K: that's a question I keep trying to wrap my head around, and failing. I wanted to say something about it in this post: to imagine a medium of exchange that could not be used as a store of wealth. I was thinking:

Suppose the market opens once a week, and we have a computerised trading system that lets you buy and sell for computer credits, which net to zero in aggregate, and must be at zero for each individual before the market closes.

Yes, I think we can imagine such a system. But it's a weird system. And all trades would have to be bunched in time, which isn't very convenient.

[K]

I think it's better to imagine the credits expiring in short order unless spent. Let's say you borrow them at cb against liquid securities. The cb will expire them in 10 minutes so you have to spend them and then the seller buys securities from someone who cancels the money by reverse repoing said securities from the cb. That we we don't need coordinated auctions. And the holding period for money can be arbitrarily short. This is almost like using capital assets as the medium of exchange, but it retains the current method by which the cb fixes the value of the unit of account. I don't think it would be weird either. The only requirement is instant settlement of securities.

[Dan Kervick]

I think you're right Nick. Money is a store of wealth, but it is a store of wealth mainly because it is a medium of exchange. If you receive some money, and that particular kind of money stops functioning as a medium of exchange, then you are no longer storing wealth, you are storing garbage.

K's point fits into this picture. If I store some lettuce, I am storing wealth. But lettuce is a rather poor store of wealth because its value depreciates rapidly. Similarly if there were a form of money whose exchange value always depreciated very rapidly, it would be a very poor lettuce-like store of wealth.

One addition to this picture. Money has a legally institutionalized status. By law it is established as a means for final discharge of obligations: both tax obligations and private debt obligations. By law, it is the only thing that an obliger must be willing to accept from the obliged. If I owe someone a cow, and I have no cows but have some money, then a court will require myy creditor to accept the money from me instead. And taxes are explicitly obligations to deliver money. So one of the things we receive in exchange for money is relief or liberation from an obligation.

[greg]

Money as a store of value? For an individual, perhaps. An individual can save it up and buy things. For a society? For an economy? Money has no intrinsic value. A society cannot take a vault and fill it up with $10 Trillion dollars, and save it for tomorrow.

And it should have no intrinsic value. The more worthless it actually is, the easier it is to keep account of its motion, the more effective it is as a medium of exchange. You don't have to subtract its 'intrinsic' value from what you actually buy with it. Any money with intrinsic worth, that is it had a use, greater than its nominal value, would soon be converted to that use. It was only when silver was worth less than its nominal value as coinage, was it useful as money.

Anyway, the operation of Gresham's Law has made sure that the money we have is worthless, even to the point of credit, money you have to pay to use, driving out (a lot of) cash.

Double the money supply, you have not made society any richer, but you have enriched the ones you gave the newly minted money to, and impoverished those you did not.

[Lorenzo from Oz]

Money is a transaction good: it is used to do transactions. Hence, an economy with three goods (output, money and assets) has two markets. The output market and the asset market with money as the good used in transactions in either (and so both) markets. And anything that can be used as a transaction good is money.

Since money is something used for transactions, you accept payment in money because you can use it in future transactions. It is a store of value in that sense, but, as you say, a peculiar one. For its value is purely due to its potential use in future transactions.

All of which is why fiat money is money in its purest form. It is not "distorted" money or "false" money or "perverted" money: it is money in its purest form for it has no value apart from its value in transactions.

Which leads to lots of interesting questions. Those about what determines the value of money in transactions and what determines the range of transactions money in general (or a specific money) is used for. These are related questions, but not the same question. And the ambit question is not only why US$ works in Red Square (even in the old Soviet Union) but why there are a whole lot of transactions we engage in which are so personal that using money would be inappropriate.

[Mandos]

Nathan: IIRC (and my memory may be wrong on this) somewhere in the beginning of Graeber's essay he talks about the economics textbook definition of money. Is it "condescending" to think that an anthropologist might look at an economics textbook definition of "money"? And that doing so might confuse him about how economists view (or should view) money? OK.

I haven't read Debt yet (and I think neither have you?) but I think the point is that Graeber's critique---from the essay and the CT discussion---is precisely focused on the view of money as a medium of exchange, so if you delete the idea of "store of wealth", most of the critique would still stand.

It's hard to overestimate the importance of this issue. If you want to rest the economics intellectual enterprise on a notion of "medium of exchange", then a view that an exchange-based society is mostly contingent on social history means that economics as a whole is only contingent on our particular social history.

[Kevin Donoghue]

Nick: If we line up all assets in terms of liquidity, then money is at the extreme end of the spectrum. And because of network externalities, the liquidity race is a "winner-takes-all" race. The winner is used in a qualitatively different way from all the other assets.

If that was true we wouldn't have had all those tedious discussions in the 1970s (and maybe earlier) about which assets should be considered part of the money supply. If you colour-code the assets in your line with red as the most liquid and violet as the least, the arguments about whether the orange and yellow ones count as money will never end.

But there is a grain of truth in what you say nonetheless. Because barter is hopelessly impractical there will always be an asset that is the 'winner' for purposes of day-to-day shopping. Usually it will be the stuff printed by the central bank, but in a hyperinflation it could be some hard currency or whatever. Still, I get the sense that you're putting the cart before the horse. You say money is special because it's the most liquid asset. Keynes says (GT Ch 17) that the most liquid assets inevitably play the role of money.

[Ryan]

Nick, I have to say that on this one you are not just wrong but very wrong.

What you are suggesting is that money is no more a store of *value* than any other good, which is true. But _value_is_not_wealth_. Wealth and value are two different things.

If they weren't, then people would be paid their wages in chairs. If they weren't, people would be fully ambivalent between payment in money versus chairs. You are right that money is "special" but you seem to want to reject the common understanding of wealth.

If I add up the value of all my assets, how do I express that number? I'd express it in dollars, wouldn't you? I'd express it in the denomination of the medium of exchange. This is because if I had to pick up and leave town immediately, I would need to convert a large portion of my assets into value and (*AHEM*) transfer my wealth from Point A to Point B.

I could theoretically transfer my wealth in chairs, but this would require significant additional monetary expenditure. I would need to pay for a big moving truck and a crew to help me load and unload. This costs me money, i.e. a part of my wealth. It does not merely cost me "value." There is a value associated with a big moving truck and a crew, but if I choose to spend my money on that, I am no longer transferring "wealth" from money to trucks and crews. Instead, I am consuming my wealth.

I can only transfer my wealth using the medium of exchange. I cannot do it in chairs. Wealth and value are two separate concepts. We should not conflate the two.

[Nick Rowe]

K: "I think it's better to imagine the credits expiring in short order unless spent."

Then I have to think about credits not being fungible, because you wouldn't swap one credit for another with a shorter expiry date.

Dan: "If I owe someone a cow, and I have no cows but have some money, then a court will require myy creditor to accept the money from me instead."

I've wondered about this. I thought I remembered someone saying that in some forward contracts, the buyer could insist on physical delivery. And if I buy a car, for example, I expect to see that particular car, with that VIN, and not cash, or any car. I could see a court saying I had to take cash compensation, if the seller were unable to deliver that car. But then cars, and cows, aren't really fungible. This is part of that whole "legal tender" thing, that I've never really gotten my head around.

greg: yes, I forgot to mention that whole wealth to an individual/wealth to society distinction. In one respect though, money is wealth to society, because it makes the economy function better. "Money is none of the wheels of trade, it is the oil that helps the wheels..." Was that David Hume?

Lorenzo: "All of which is why fiat money is money in its purest form. It is not "distorted" money or "false" money or "perverted" money: it is money in its purest form for it has no value apart from its value in transactions."

I like that. Vaguely remember Georg Simmel saying something similar.

"Those about what determines the value of money in transactions..."
which is the question monetary economists answer (in different ways).

"...why US$ works in Red Square (even in the old Soviet Union).."
a question we rarely address, but do sometimes try to say something about.

"..why there are a whole lot of transactions we engage in which are so personal that using money would be inappropriate."
a question we rarely think about at all, but perhaps should.

[Nick Rowe]

Mandos: I haven't read Graeber's Debt. I skimmed the Essay. For some reason it didn't grab my interest (maybe him; maybe me).

"It's hard to overestimate the importance of this issue. If you want to rest the economics intellectual enterprise on a notion of "medium of exchange",.."

I only want to rest monetary economics on the notion of money's being a medium of exchange, not the whole of economics. Money is only a part of economics. Nearly all of micro ignores money. At least half of macro ignores money. Even that half of macro that does sort of include money doesn't really include it intrinsically (but it should, IMHO). I argue with New Keynesian macroeconomists about whether their models do and should include money as a medium of exchange. Many say it doesn't, and doesn't need to.

"... then a view that an exchange-based society is mostly contingent on social history means that economics as a whole is only contingent on our particular social history."

I'm comfortable with the idea that a society based on exchange is contingent on social history. Other forms of economic organisation are possible. I like to think there is some sort of tendency towards markets, under most conditions (though, if pressed, I couldn't say much about those conditions), but it's not inevitable.

Economics as we teach it is mostly (not all) about market economies, because: that's what we mostly live in; that's what raises interesting questions; those are the questions we have a comparative advantage in answering. But: some economics applies to all economies, whether market or not; fields like comparative economic systems, economic institutions, economic history, etc., still hang in there. E.g. Frances does work on the economics of the family, and Coase got a Nobel for the internal economics of the firm. It's not all markets.

It's tempting to think that economics is all about money. But it isn't. It ought to be a bit more about money, IMHO.

[Mandos]

Right, but it's hard to deny the effect, at least, that monetary economics and its particular conceptualization has had on the public discourse at the very least. Remember, perhaps, when you recently told me that the language of deserving is very foreign to mainstream economics. (To which I was going to respond but waited too long.) Sure, fine, but the way the apparent underlying assumptions are structured fits very well with the "deserving" thought-template. "Undeserving" is simply the moral valuation of having wants (needs) that don't doubly coincide...

[Nick Rowe]

Kevin: One of these days I must do a post about winner takes all in the liquidity spectrum. Still getting my head clear on it.

"Still, I get the sense that you're putting the cart before the horse. You say money is special because it's the most liquid asset. Keynes says (GT Ch 17) that the most liquid assets inevitably play the role of money."

I see a snowballing effect of 2-way causation. This goes back to Menger. People are more likely to accept a more liquid asset in trade, because they can easily sell it again if they decide they don't want to keep it. And the fact that more people are willing to accept it makes it even more liquid. So it spirals. When all other goods, no matter how liquid, get traded for one good, the most liquid, there is a qualitative difference between that good and all the others.

[Nick Rowe]

ryan: we normally measure both wealth and value in units of money. But we don't have to. Sometimes we adjust for inflation, and measure both in CPI baskets. I have known farmers measure wealth and value in acres of land, or tons of wheat. We can still talk about both wealth and value in a barter economy. We can even talk about Robinson Crusoe's wealth, and the value goods have to him, even though he can't sell anything.

"I could theoretically transfer my wealth in chairs, but this would require significant additional monetary expenditure."

When people move house, sometimes they move their furniture, and sometimes they sell it and buy again. It depends on transportation costs vs liquidity costs. Yes, money is very liquid and very easily transported.

[K]

Nick: "Then I have to think about credits not being fungible, because you wouldn't swap one credit for another with a shorter expiry date."

"Expiring credits" was just a thought experiment to motivate the limit of zero quantity. In practice, you would just create the credit at the moment of the transaction, for the purpose of the transaction, and then back to capital assets immediately after the transaction. In such a system, nobody would even think of how much medium of exchange they held, since they'd be holding their liquid assets in the form of securities. Nobody would think of money as anything but an accounting device.

If we were really going to create such a system, we wouldn't actually have expiring credits. We would have the CB support the entire transaction mechanism directly (like LVTS does for the commercial banks, just faster) and nobody would ever notice the medium of exchange.

[Nick Rowe]

Mandos: "Right, but it's hard to deny the effect, at least, that monetary economics and its particular conceptualization has had on the public discourse at the very least."

It's certainly true that "money" is the symbol that represents almost everything economic. But monetary economics is a field in economics that the majority of economists stay well clear of. I think it's fair to say that the majority of economists don't feel comfortable with teaching monetary economics even at the first year level. It just feels "weird" to them. And it should. Because the medium of exchange is a very weird good. As John Quiggin said:

"Both as a unit of account, and as a store of value, money appears all the time in economics, from the introductory textbooks to the most advanced work. By contrast, money as a medium of exchange is something of an embarrassment. It doesn’t fit will into standard models, and theoretical attempts to formalize the “double coincidence of wants” idea, while useful as a five-finger exercise for grad students, haven’t yielded much in the way of insight."

He's (mostly) right about that. (I think it does yield insight, but it's hard to formalise to get those insights.)

""Undeserving" is simply the moral valuation of having wants (needs) that don't doubly coincide..."

That doesn't seem right to me. Maybe I don't understand you.

[Nick Rowe]

K: As I think I've said before, I think you are onto something there, but I can't get my head around it to say anything useful in response. Maybe someday I will.

[Ryan]

Nick, CPI is an arbitrary government construct specifically designed to conflate the difference between what Ludwig von Mises termed "money" versus "fiduciary media." As for farmers' lingo, they are talking about assets, i.e. real wealth. Acres of land and their agricultural yield are real assets, real wealth. The hoe the farmer uses is not an asset, nor is it wealth. It is a good, a dead-end to what we call wealth. The point at which you decide to take your wealth out of storage by exchanging money for a hoe is the point at which you consume your wealth in exchange for an act of speculation. ("If I use this hoe, I will be able to produce more wealth.") You seem to be obliterating the difference between an asset and a good.

Note that although a new car is not currently within my budget - and therefore definitely not a part of my wealth - I value it at exactly the same price as I do when I buy one. Wealth and value are entirely separate economic concepts, as are assets and goods. We use these words for a reason.

When we talk about economic concepts (or any definition at all, really), we list the features that define the concept in question. The fact that both vehicles and Planet Earth transport human beings across space does not mean that "transportation of human beings is a non-unique property of vehicles." Transportation of human beings is the most important feature of vehicles. That makes all the difference. You can sit in your wealth-transferring chair all day long and still not make it to Columbus, Ohio no matter how far a distance Planet Earth takes you. In order to move across space in an individually meaningful sense, you have to take a vehicle.

Moreover, we would never list "transportation of human beings" among the essential properties of Planet Earth. That characteristic is entirely beside the point.

[Ryan]

Whoops! Sorry - forgot to close my HTML tag!

[Mandos]

That doesn't seem right to me. Maybe I don't understand you.

What do people mean by "undeserving" (particularly when they're referring to poverty)? Someone is undeserving when there is a set of stimuli to which they have the (apparent) capacity to respond in a way that would better their condition, and they don't appear to be responding to those stimuli. In the current "Western" culture, those stimuli are the wants and capacities of others, and money-as-exchange gives us the ability to barter these wants and capacities without actually directly encountering the stimuli. Therefore, someone is undeserving if they do not appear to be performing the indirect-barter of wants and capacities to improve their status.

Why would someone not be performing this indirect barter? It is either because the things they want are not offered, or they do not *want* to exchange some apparent capacity for them. The double coincidence is broken. We can view this situation in neutral terms, as what I think some would call a "market failure". But why would we? It "costs" the rest of us to adjust for this market failure. Because we don't want to pay the price for this market failure, we view it as an attack on us, and therefore we assign it strongly negative moral rating. "Undeserving."

[Mandos]

Libertarian tag-fail is attacking my socialist screed. I view this as class oppression!

[Stephen Gordon]

But *monetary* economics is a field in economics that the majority of economists stay well clear of. I think it's fair to say that the majority of economists don't feel comfortable with teaching monetary economics even at the first year level. It just feels "weird" to them.

Yep. I took the graduate monetary theory sequence at both UWO and U of T, and I wrote the comp in the field. But all I really remember is that monetary theory is a nasty, thorny field in which it's way too easy to get lost.

And yet it's also the field that attracts the most cranks.

[Frances Woolley]

Random thoughts - in reply to Stephen's comment's above - actually monetary econ was always one of my favourite subjects, but I couldn't take real business cycle theory 1980s-style macro, and in lots of places money=macro.

On Nick's post - whenever there's a post like this that seems really simple, straightforward and sensible, I always suspect that there's some deep insight buried there that I'm just not seeing. In this case, it seems to be " how money can sometimes fail as a coordinating device" - that our confusions about what money is and does cause our confusions about how money fails to do what it's supposed to do?

[Jeremy Fox]

Thanks Nick, I was hoping you'd do a response to Quiggan's post.

Re: money as a medium of exchange, I was imagining that you'd say something like the following. In order to understand how money functions as a medium of exchange *in a modern economy*, we need to imagine what that economy would look like if there were no money but everything else were held constant. A modern economy without a medium of exchange is a barter economy. So when anthropologists (or Quiggan, relying on the anthropologists) say that, as a matter of historical fact, non-monetary economies aren't barter economies but rather are based on norms of reciprocity and sharing, that's simply irrelevant. Asking how money functions as a medium of exchange in a modern economy is a conceptual question, not a historical one. And the historical answer is the wrong answer to the conceptual question, because non-monetary economies don't just lack a medium of exchange, they also differ from modern economies in other ways, such as being based on norms of reciprocity and sharing.

Is that gloss a correct paraphrase/summary of your response?

Seems like there might be analogies here to social contract theories in political philosophy--most social contract theorists did not claim to be telling a historical story about how 'pre-contract' societies became 'post-contract' societies. But it's too early for me to think this through, so this thought could be totally off base.

[Scott Sumner]

Nick, You make a very good point, but I'd put it slightly differently:

1. The definition of money is ambiguous, because it doesn't address what happens if the medium of exchange and the medium of account become separated. That's probably because they virtually never become completely separated.

2. That which is used as the medium of exchange is generally defined as money.

3. Economic models of the macroeconomy also implicitly define money as that which is used as the medium of account. If the medium of account and medium of exchange ever became separated (theoretically possible), then we'd have to rethink the definition.

4. That which is the store of value is not defined as money. That's the point you are making.

5. I'd prefer the definition of money be reduced to "that which is the medium of account." But I know you see the medium of exchange as the essential characteristic. Since money always seems to do both functions, I don't think it's a disagreement that is currently of much practical importance.

6. Imagine the dictionary definition of canoes was "a. Small narrow boats used in rivers and lakes, and b. A store of value." Most people would consider that a silly definition. So you are absolute right that it's insane to define money in terms of it's role as a store of value. However many textbooks list three "roles of money" and it's perfectly fine to call "store of value" one of the roles of money. But the definition should be either the medium of account, or medium of account and medium of exchange.

7. If the medium of exchange and account ever became separated, I'd strongly oppose defining money as the medium of exchange. That's because the assumption at the heart of monetary models is that price level changes are equivalent to changes in the value of the medium of account, not the medium of exchange.

[Min]

K: "Is it a necessary property of money that you can hoard it? Cause you could easily design a medium of exchange that you can't hoard."

Nick Rowe: "I wanted to say something about it in this post: to imagine a medium of exchange that could not be used as a store of wealth."

What about Gesell's Free-Money?

Gesell: "Money is an instrument of exchange and nothing else. . . .

"Only money that goes out of date like a newspaper, rots like potatoes, rusts like iron, evaporates like ether, is capable of standing the test as an instrument for the exchange of potatoes, newspapers, iron and ether. For such money is not preferred to goods either by the purchaser or the seller. We then part with our goods for money only because we need the money as a means of exchange, not because we expect an advantage from possession of the money.

"So we must make money worse as a commodity if we wish to make it better as a medium of exchange." ("The Natural Economic Order" at http://www.silvio-gesell.de/en/neo/part4/1.htm )

Free-Money self destructs at a rate of about 5% per year (although that rate could be adjusted, OC).

[marcel]

On whether money is the most liquid good or the most liquid good becomes the one use as money, Kiyotaki and Wright (1989?) seems pertinent.

About the different functions. Whether or not money is a store of wealth, that function is key, not to its definition, but to what makes it interesting. The defining function of money is the medium of exchange one. Another function that is typically unique to money is the unit of account function. The final conventional function, store of wealth, is mentioned because it is thought to be why we have a field called monetary economics.

Without it (store of value fn), it would be difficult to make sense of variations in velocity, which seem closely tied to business cycles. Given the disruptive effect of business cycles on society, we wish to understand them. So, we list the store of value function not as a defining function but as an important one, along with unit of account (which is also not defining) and medium of exchange (which is).

We can imagine a situation in which all exchange had to involve cash/currency (pieces of paper) but this had neither unit of account or store of value functions.

1) Unit of Account: Prices could be quoted in heads of cattle, and before each exchange, it was necessary that the 2 parties check the price of money (in cattle) so they could determine how much cash was needed for the exchange.

2) Store of Value: Each day, people have to go to the bank and get cash. The person withdrawing it must endorse it with a thumb print that is matched to the account owner's for verification, and it is stamped with the date. In the course of the particular exchange for which specific pieces of paper are used, the two transactors must endorse the paper with their thumb prints, and the buyer's is compared to the original one from earlier in the day. Any cash that is not returned to the bank before midnight loses all value. If it is not returned by the seller in the exchange, it cannot be returned, and when returned the depositor's thumb print is compared to the seller's. If the 2 thumb prints of the buyer and the one who originally withdrew the cash do not match, it cannot be returned. When returned to the bank, the depositor can choose which of a variety of assets (accounts) to put it in. Cash/Currency/Money is not a store of value.

Very artificial, but it does allow for us to separate the functions. Store of value is mentioned only because it is thought to be central to what makes money interesting to monetary economics (or, as Lucas tried to rechristen it (I think), the study of business cycles).

I just noticed that Scott Sumner made related points just before me. I disagree with him about his point 7, since I think the concern with inflation (price level changes) stems from the aggravations it caused both to calculations (unit of acct) and savers (store of value).

[K]

Stephen,

I think you mean *therefore* it attracts the most cranks.

Scott: "I'd prefer the definition of money be reduced to 'that which is the medium of account.'"

Agreed, since it's easy to imagine a system without any quantity of medium of exchange. All we need is capital assets assets as a store of wealth, instant settlements, and a *unit* of account to avoid n^2 exchange rates, i.e. a classic pure exchange economy like Lucas '78. But without any relevance for the quantity of the *medium*, then what is the meaning of "monetarism"?

Min: I was, in fact, thinking about Gesell.

[Min]

On the matter of disappearing currency: Before the American Revolution, a number of colonies had their own fiat currencies. Pennsylania created its money by lending it. Could something like Gesell's Free-Money be created by 0% 20 year loans?

[K]

Gesell, btw, had good instincts, but I don't think his idea would have actually done anything. Interest rates would just have risen 5% compared to gold standard money. His idea, in fact, is just Friedman's 5% rule, which is just a constant velocity version of Scott's 5% (NGDP) rule.

The problem with money, IMO, has nothing to do with with its medium of exchange properties. It's inherent in the unit of account. The paradox of thrift, and its dual, the consumption/asset bubble, is every bit as aggravated by nominal debt (and other nominal rigidities) as it is by nominal currency. (I'm just reiterating the standard Keynesian/NK view here). Eliminating the medium of exchange has a lot of desirable properties, but restoring Say's law, in my opinion, is not one of them.

[Patrick]

K: wouldn't you want to add 'equivalent of a printing press' to your list? Money is also special in that the CB can make more of it whenever it wants, and people (within limits) still think of it as 'wealth'. I have a hard time imagining a the equivalent of a printing press for capital assets.

[Nick Rowe]

Jeremy: "Seems like there might be analogies here to social contract theories in political philosophy--most social contract theorists did not claim to be telling a historical story about how 'pre-contract' societies became 'post-contract' societies. But it's too early for me to think this through, so this thought could be totally off base."

That is exactly on base. At least, you are on exactly the same base as me (oops, I think my baseball metaphor just backfired). See my old post here

"Is that gloss a correct paraphrase/summary of your response?"

Yes. Except:

"So when anthropologists (or Quiggan, relying on the anthropologists) say that, as a matter of historical fact, non-monetary economies aren't barter economies but rather are based on norms of reciprocity and sharing, that's simply irrelevant."

Actually, it is relevant, but in a different way than they intended. If he's right about the history, it suggests to me that barter exchange is so very difficult that it's rare. It suggests to me that exchange is only workable in practice if it's monetary exchange, so that money is even more important than I thought it was. If we don't have monetary exchange, the practical alternatives are non-exchange economies.

[Nick Rowe]

Damn! I've forgotten how to embed links AGAIN! Can somebody please help me?

[Frances Woolley]

Nick, I'll send you the info via email, the code doesn't show up when I post it here. F.

[Frances Woolley]

<a href=http://worthwhile.typepad.com>best website ever</a> should show up as best website ever

[Frances Woolley]

Shoot, sorry, Nick, that should be href not ref. Please edit!

[Nick Rowe]

Thanks Frances! edited (mine and yours). How the hell do you remember this stuff? Plus, what you did in that comment above, where you managed to rig it so the href stuff appeared as written??

I'm feeling like I should retire, and take up a simpler life.

[Jeremy Fox]

Nick wrote, referring to pre-modern economies not using money as a medium of exchange:

"Actually, it is relevant, but in a different way than they intended. If he's right about the history, it suggests to me that barter exchange is so very difficult that it's rare. It suggests to me that exchange is only workable in practice if it's monetary exchange, so that money is even more important than I thought it was. If we don't have monetary exchange, the practical alternatives are non-exchange economies."

Good point.

[Frances Woolley]

Nick, I remember how to do it the same way I remember how to bake a cake - that is, I don't remember, I just follow the recipe. Which, in this case, I found by googling 'how to display html code'.

You, on the other hand, are capable of making a cake without using a recipe, which is a truly awesome and amazing accomplishment.

[Jeremy Fox]

Nick wrote, re: the analogy with social contract theory:

"See my old post here"

That old post is *exactly* what I had in mind. I really should quit being so lazy and start using the search function before I comment. ;-) Because half the time I am effectively just asking you to point me to an old post.

Nick also wrote:

"At least, you are on exactly the same base as me (oops, I think my baseball metaphor just backfired)."

*Snort!* You are on fire today, Nick!

[Phil Koop]

I'm surprised at the flak you got here, as I thought you were making an obvious and uncontroversial point. I've been living under a rock.

I have to support Kevin Donoghue: there are degrees of moneyness, it is not a binary property. That is why we can imagine K's evaporating money thought experiment: the "securities" upon which it depends are just another kind of money (they are "ATM options", ha ha ha!) Inferior, in that the money value of a security is uncertain until the moment we exchange it, superior in that it can be stored indefinitely. But possessing the same properties of guaranteed liquidity at all times and places, and no intrinsic value.

[Nick Rowe]

Scott: thanks! I agree with your 1 to 6, but disagree with your 7 (as you might expect!)

"That's because the assumption at the heart of monetary models is that price level changes are equivalent to changes in the value of the medium of account, not the medium of exchange."

And that is precisely what is wrong with so many monetary models! My old post on bling again!

[Sina Motamedi]

Amen.

[Account Deleted]

I think there is a lot to learn from the Bitcoin experience.

Bitcoin started out as digital money and the protocol allows only a certain number of coins to be created.

There have been attempts to create money that would inflate(it is trivial to recode and launch your own coin), but most people in the community don't support such attempts, because no one wants to end up supporting an inflating coin.

The value of bitcoins has fluctuated wildly vs the dollar. People don't have any guarantee that anyone is going to accept a bitcoin. But there is a community accepting bitcoin today. There is no force as in fiat money, very little tradition compared to precious metals and no use for the digital hashes that are called coins. There is value, because there is a supporting community. There is a supporting community because they expect the bitcoin to be a store of value.

[JP Koning]

"How much would I have been willing to pay for the $20 note in my pocket if the seller had imposed a condition that I could use it as long as I wanted but could never sell it again, or rent it out to someone else? Nothing."

That's the same question a value investor asks before buying a stock.

The answer usually comes to something like, if I can pay $50 for a stock that is worth $100, then even though I can't resell it in the market I'll still buy it. Because the stock can't be resold, that $50 in value has to be realized through dividends. But if the stock is prohibited from ever paying a dividend, this value can still be realized by the firm repurchasing and canceling shares at higher prices.

I'd say roughly the same applies to central bank notes. If I can buy a note for far less than it's worth and hold it till the central bank begins to mop up the supply notes and cancel them, then I'll go ahead with the transaction. Since central banks are less opportunistic than firms and therefore less likely to announce buy backs, I'd only buy at a huge discount. A huge discount to what? The value of its bonds, bills, gold, buildings, and forex. In sum, the price I'd be willing to pay for non-transferable bank notes is not "nothing" but some number >0.

[J.V. Dubois]

Prakash Chandrashekar: It seems that your post was truncated by the system. Should it not be the following: "There is value, because there is a supporting community. There is a supporting community because they expect the bitcoin to be a store of value. And they expect the bitcoing to be a store of value because they expect to use it as medium of exchange sometime in the future (or to sell it)"

Basically I see little Difference between bitcoin and (some versions) of modern art or collectibles. There are people willing to offer millions of dollars for some piece of art with really low "real" value. They are willing to pay that much, because they believe that they would be able to sell it for similar ammount. Maybe there are people that really love to look upon that piece of art and that derive so much "utility" from this act. But then we have to admit that there are people willing to look upon masses of green papers (or sums on internet banking accounts), or who think of money as the best form of lighter and experience the similar strong emotions "worth the money"

[jsalvatier]

Yes! The way I have put it is that while "store of value" is clearly an important feature of money, it's not a feature unique to money, so we shouldn't expect the unique features of a theory of money to originate there. If you think there's something special about money, then it has to come from the differences it has with other goods. I wrote about this when I was trying to write an introduction to thinking about money, trying to make money as unmysterious as possible (http://goodmorningeconomics.wordpress.com/monetary-economics-1-money-as-a-good/).

[Ryan]

Scott Sumner said: "7. If the medium of exchange and account ever became separated, I'd strongly oppose defining money as the medium of exchange. That's because the assumption at the heart of monetary models is that price level changes are equivalent to changes in the value of the medium of account, not the medium of exchange."

I think this is fascinating. This point highlights my objection to monetary models precisely because I think money's role as a medium of exchange is more defining and important than its role as a medium of account.

But then again, I'm nuts.

[Nick Rowe]

JP: "I'd say roughly the same applies to central bank notes. If I can buy a note for far less than it's worth and hold it till the central bank begins to mop up the supply notes and cancel them, then I'll go ahead with the transaction."

Hmmm. Like under the gold standard, when convertibility was temporarily suspended. If we had a redeemable (convertible) paper money, then it would retain its value even if we didn't use it as a medium of exchange. It would become like a Tbill.

[Andy Harless]

My first thought upon reading this last night was that Quiggin has gone over to the dark side. Once you declare "store of value" to be a quintessential feature of money, you invite demands that money should be a good store of value, and it becomes that much more difficult to recover from (or prevent) a depression. As Nick says, there are plenty of other stores of value but no other media of exchange or account (usually, anyhow). When you require the medium of exchange and account to be also a good store of value, you invite depression. It is a temptation of Satan to tie this feature to money specifically. The love of money as a store of value is the root of all evil business cycles.

But in the comments section of his own post, John Quiggin makes a point that I find hard to answer. If you allow the items being exchanged to have time subscripts (as with babysitting), then how is it possible to conceive of a medium of exchange except as a store of value? To the extent that we're talking about exchanges that have a time component as their essence (as when babysitting is the only good, so the only property that distinguishes the items exchanged is their time component), storing value is what exchange is all about.

FWIW I think Scott Sumner undersells the medium of account idea above. It is possible to separate the medium of account from the medium of exchange, as for example when contracts are written with COLA clauses. In that case, controlling the medium of account becomes critical, and part of the reason that the 1980-1983 double dip was so severe was that the medium controlled by the Fed was no longer the only medium of account -- so in effect there was not just "tight money" (and Scott would say there was not tight money at all) but also "tight oil" influencing widely used units of account.

Having said all that, though, I should say I am starting to come around to Nick's view of the medium of exchange as a critical element in business cycles. The question is why, when money becomes precious (in excess demand), people are still willing to part with it in exchange for items priced at their usual prices. And the reason is that money is the medium of exchange. Essentially, people are willing to pay a premium (i.e. buy things with money despite the fact that its shadow value is greater than its exchange value) because they would have to pay an even higher premium if they used some other medium (the premium in that latter case being the cost of searching for someone willing to accept the other medium). So it is the medium of exchange feature of money that allows people to receive it, and thereby keep hoarding it, even when it is in excess demand.

[Min]

K: "Gesell, btw, had good instincts, but I don't think his idea would have actually done anything. Interest rates would just have risen 5% compared to gold standard money. His idea, in fact, is just Friedman's 5% rule, which is just a constant velocity version of Scott's 5% (NGDP) rule."

Isn't Gesell's rule the opposite of Friedman's? It decreases the money supply by around 5% per year instead of increasing it (if the Treasury or Central Bank does nothing).

X: "The problem with money, IMO, has nothing to do with with its medium of exchange properties. It's inherent in the unit of account. The paradox of thrift, and its dual, the consumption/asset bubble, is every bit as aggravated by nominal debt (and other nominal rigidities) as it is by nominal currency."

Gesell seems to have been quite aware of the paradox of thrift. Later in "The Natural Economic Order" he states:

"But what must happen if everyone brings produce worth $100 to market, and buys produce for only $90 - that is, if everyone wishes to save $10. How can this contradiction be resolved, how can all men be enabled to save ? The answer is given, the contradiction is resolved, by Free-Money. Free-Money applies the Christian maxim: whatsoever ye would that men should do to you, do ye even so to them. It says: If you wish to sell your produce, buy the produce your neighbour wishes to sell. If you sold for 100, buy for 100 in return. When everyone acts in this manner, everyone will be able to sell his whole produce and to save. Otherwise savers mutually deprive one another of the possibility of carrying out their purpose."

He seems to think that Free-Money in itself allows everyone to save money. But how, unless the gov't increases the money supply? (That's true of every monetary system, right?)

[Nick Rowe]

Andy: "Essentially, people are willing to pay a premium (i.e. buy things with money despite the fact that its shadow value is greater than its exchange value) because they would have to pay an even higher premium if they used some other medium (the premium in that latter case being the cost of searching for someone willing to accept the other medium)."

Neat.

Ryan: "But then again, I'm nuts."

Join the MOE-nutter club!

[Nick Rowe]

Andy: "To the extent that we're talking about exchanges that have a time component as their essence (as when babysitting is the only good, so the only property that distinguishes the items exchanged is their time component), storing value is what exchange is all about."

Imagine the goods all being produced and sold at the same time, but being arranged around a circle in space, not in time. Each person wants to buy the good produced one house over in a clockwise direction.

[Nick Rowe]

marcel: "Without it (store of value fn), it would be difficult to make sense of variations in velocity, which seem closely tied to business cycles. Given the disruptive effect of business cycles on society, we wish to understand them. So, we list the store of value function not as a defining function but as an important one, along with unit of account (which is also not defining) and medium of exchange (which is)."

That seems very sensible to me.

(Your comment got delayed by being caught in our spam filter, sorry.)

[Nick Rowe]

Min: "Gesell seems to have been quite aware of the paradox of thrift. Later in "The Natural Economic Order" he states:.."

That is a lovely quote from Gessell. But I disagree on it being about the paradox of thrift. I would say it's the paradox of hoarding (the medium of exchange). I think he is not using "saving" in the normal NIA sense, but in the popular sense, of saving in the form of money.

[Min]

Andy Harless: "But in the comments section of his own post, John Quiggin makes a point that I find hard to answer. If you allow the items being exchanged to have time subscripts (as with babysitting), then how is it possible to conceive of a medium of exchange except as a store of value? To the extent that we're talking about exchanges that have a time component as their essence (as when babysitting is the only good, so the only property that distinguishes the items exchanged is their time component), storing value is what exchange is all about."

But does it need to be a perfect store of value? Suppose, a la Gesell, that every Monday morning at 3:00 a. m. the gov't levied a tax of 0.1% on money deposits above $10. Money would still be a store of value, but one of decreasing value. Bad idea?

[K]

Min:

"He seems to think that Free-Money in itself allows everyone to save money. But how, unless the gov't increases the money supply?"

Right, and if I remember correctly, Gesell envisioned a citizen's dividend equal to stamp rate and distributed equally (to all mothers, maybe?).

"Isn't Gesell's rule the opposite of Friedman's?"

Gesell's stamp tax, by itself, does little. By itself it just results in 5% less money, and 5% lower prices, so no change in real wealth for money hoarders. Presumably everyone would just adjust prices down by 5% on the date of the stamp tax. It is the 5% citizen's dividend which causes the loss to the money hoarders, equivalent to an inflation loss, but without the impact on nominal prices, because of the stamp tax. The scheme is really just a citizen's dividend financed by a stamp tax on money holders. So from a store of value perspective it's equivalent to Friedman's rule, but it's the citizen's dividend that has the real impact and it's obviously very different from Friedman from a distributional perspective.

And I take back what I said about Gesell's scheme not doing much. The citizen's dividend makes it quite powerful.

[Donald A. Coffin]

I should probably read all the comments before I add my 2-cents-worth...but...

I agree that of the "functions of money" stuff, "Money as a store of value" is a distant, distant third. But, I would argue, something that serves as a medium of exchange will serve *better* as a medium of exchange is it *also* serves as a store of value. The argument I make in class for this is, essentially, the argument that some things that serve as media of exchange stop fulfilling that function in hyperinflations, precisely because they no longer serve as even temporary stores of value.

Being a store of value does not make something money. But, for something that *is* money--that serves as a medium of exchange, which I regard as *the* core function--being a store of value is a feature, not a bug.

[Declan]

I agree 100% with Scott and his 7 points above.

One ramification of seeing a recession as a problem with the unit of account rather than with the medium of exchange is that a problem with the medium of exchange would simply be a coordination problem, whereas a problem with the unit of account is a collective action problem (i.e. Prisoner's Dilemma).

Getting everyone to agree to, say, a 20% devaluation of all economic claims is a signficant coordination problem to be sure, but what makes it particularly challenging is that if I can get everyone *else* to revise the value of their wages/debts/etc. downwards while keeping mine unchanged, I will come out ahead. The situation with insufficient demand represents a standoff where people are hoping someone else will go first in discounting their claims to reflect the shortage of the unit of value. Over time, the people with the lowest level of power will capitulate and the economy will adjust - unless instead of capitulating, they default on their debts, since that will aggravate the shortage of the unit of value and lead to a debt-deflation spiral.

[K]

I think it's worth noting that in Gesell's economy, the nominal interest rate would be unaffected because prices are unaffected by the tax/dividend (assuming it's done fairly continuously as Min suggests). So while it ruins the possibility of hoarding money as a store of value, it has no impact on debt. So it serves as a really good thought experiment in paradox of thrift vs Nick's paradox of hoarding. If it's the paradox of hoarding that causes instability, Gesell's tax should fix it. Otherwise it wont.

Andy: "I should say I am starting to come around to Nick's view of the medium of exchange as a critical element in business cycles"

Now you're the one going over to the dark side!

[Nick Rowe]

Don: "The argument I make in class for this is, essentially, the argument that some things that serve as media of exchange stop fulfilling that function in hyperinflations, precisely because they no longer serve as even temporary stores of value."

I agree. What amazes me though, is that this effect is so weak. It takes a truly horrendous hyperinflation to get people to abandon using a medium of exchange. We are prepared to pay a very high price indeed to keep on using the same money we have always used and that everyone else uses. That shows how costly barter is. And it shows how costly it is to coordinate on switching to a different money, without the money-issuer itself doing this coordination.

Declan: "a problem with the medium of exchange would simply be a coordination problem, whereas a problem with the unit of account is a collective action problem (i.e. Prisoner's Dilemma)."

I see what you are saying, but I disagree. This is hard for me to explain clearly. Assume an economy with monopolistically competitive firms (like in NK models) but a simple monetarist AD function, like MV=PY.

Start in full Nash equilibrium. That equilibrium is a PD. If all firms cut their prices, all would be better off, because Y would increase (towards the competitive equilibrium). But if they all did cut their prices, each firm would want to defect.

Now assume sticky prices, and halve M. P stays the same, and Y halves. They are all worse off. This is more of a coordination failure. If each firm thought the others would halve their prices, it would want to halve its price too. None would want to defect.

Sorry. That wasn't clear.

K: "If it's the paradox of hoarding that causes instability, Gesell's tax should fix it. Otherwise it wont."

Good point.

[marcel]

I just recalled a vague distant memory - IIRC, at some point in the years after the US Revolution, the standard unit of account (in the US, not in that vast benighted area to our north) was dollars. Because there were few of these in many parts of the country (in part because of the scarcity of issuing institutions, i.e., banks), the medium of exchange tended to be other things, sometimes currency of one sort or another (British, Spanish, etc.), at other times commodities like whiskey, tobacco or cotton.

During the period that followed the 2nd Bank of the US and preceded the establishment of the Fed, the unit of account was the dollar, but there were many media of exchange that traded at discounts with each other; this was especially true of the Free Banking period before the US civil war.

[Donald A. Coffin]

Marcel--

It's correct to say that that, in the early national period,, the U.S. economy was not fully monetaized, and so things were messy. That messiness took various forms, one of which was the issue of banknotes by commercial banks. (This was, in general, how banks made loans then. Lend you banknotes, receive more back from you in repayment in banknotes. Very few people had readily transferable--"checkable"--bank accounts.) Another form was that many (the literature is unclear on what percentage this accounts for, but there's some evidence that, pre-revolution, it may have been as much as 80%) consumer transactions were carreid out in terms of merchant credit, with that credit extending (in some cases) for 6-12 months, and with repayment made in a number of ways. The fundamental cause of this messiness, in my judgment, was the lack of a widely accepted medium of exchange.

[marcel]

Apropos the unit of account, Declan wrote (February 23, 2012 at 03:45 PM): "... since that will aggravate the shortage of the unit of value and ... "

I think this shows a misunderstanding of the concept "unit of account" (which Declan appears to use interchangeably with "unit of value"), and I don't think I am merely being pedantic here.

We would not talk about there being a shortage of inches or ounces or watts. But money, as a unit of account, is in exactly the same category. It is what we use to measure something, to specify a quantity. We can have a shortage of so many inches of something in particular, or so many ounces of gold, or so many watts of electricity (or some other type of power), but we cannot have a general shortage of inches (well, maybe in Wonderland). Similarly we cannot have a shortage of the unit of account. The unit of account can exist independently of any actual good.

It has been 15-20 years since I made any attempt to keep up on this, but Robert Hall wrote an article* many years ago about constructing a unit of account that would be a bundle of specified; all prices would be quoted in terms of this, and IIRC, the goal of monetary policy would be to stabilize the value of the medium of exchange in terms of this account. In this case, money would not be the unit of account, but (and this is redundant) it is the medium of exchange, and remains a store of value. Call this unit of account "The Bundle". We would not talk about shortages of the Bundle; we would be able to measure prices in terms of it, whether or not there are "enough" Bundles in the economy, so long as markets for each of its constituents continue to function.

And so long as these markets continue to function, it is not clear what a shortage of any constituent means: supply falls, and cet par, price rises and the shortage disappears. This may have affects on various economic sectors, and eventually on the macro economy. Consider what happens if oil is one of the goods used to define the Bundle. Supply falls, the price of oil rises in terms of the Bundle, but monetary policy keeps the value of the dollar (medium of exchange) constant, effectively forcing down prices of all other goods in terms of the Bundle.

This made more intuitive sense before we moved to a fiat money, when we had a gold or gold exchange standard, and we could talk about prices implicitly in terms of a quantity of gold (because the dollar was fixed in terms of quantities of gold). However, moving to a fiat system does not change the underlying logic, it just makes the measuring unit more obviously variable.

*Just found it: Hall (1982) Explorations in the Gold Standard and Related Policies for Stabilizing the Dollar"

NBER Reprint No. 416
Issued in October 1983
NBER Program(s): EFG
No abstract is available for this paper

Published:
Hall, Robert E. "Explorations in the Gold Standard and Related Policies for Stabilizing the Dollar." Inflation: Causes and Effects, edited by Robert E . Hall, pp. 111-122. Chicago: University of Chicago Press, 1982. ,
Explorations in the Gold Standard and Related Policies for Stabilizing the Dollar, Robert E. Hall, in Inflation: Causes and Effects (1982), University of Chicago Press

[marcel]

In my last comment (just above as I type this), I made a key typo.

"would be a bundle of specified; all prices" should be "would be a bundle of specified goods; all prices"

Oops

[JP Koning]

"Like under the gold standard, when convertibility was temporarily suspended. If we had a redeemable (convertible) paper money, then it would retain its value even if we didn't use it as a medium of exchange. It would become like a Tbill."

Yes, like that. Wicksell, Fisher, and Laughlin had an interesting debate on this with respect to the suspension of convertibility of the greenback.

The store-of-value vs MOE argument is interesting. I just don't know why it has to be conducted on a either-or, us-not-them way, as opposed to figuring out how to configure the two together as best as possible. They are both somehow true.

[edeast]

Marcel, "The unit of account can exist independently of any actual good"
As long as it includes at least one actual good.
pedantic pt*2.

The unit of account has to be a division of an exchangeable 'commodity' otherwise, it is meaningless noise, or semantic 'bottom'.

Just as metric is convertible into weight of water on earth(I think) Celsius depends on properties of boiling water...

[edeast]

Which is the assumption you were going under, talking about market failures in the basket of goods. I just made it explicit. I just wanted an excuse to say semantic bottom. Everyone carry on with comments.

[Peter T]

Hmm. May have missed something in the posts. Try this:

Define "money". M 3 includes cash, bank deposits, money market deposits and term deposits. Money - as far as consumers go - also includes credit card balances. The comment by Kevin hits it - you can line up all your non-physical assets from the most to the least liquid and argue about which are "money". Then you can order them by degree of acceptability (cash is taken anywhere within the sphere of its circulation, credit cards almost everywhere, gold bars only at specialised dealers, my IOU only at my regular bar and so on). Look at this and you can only conclude that anyone who can persuade someone else to take a note has created money. If that is too broad, take the narrower case where the note is transferable (this is the documented origin of money) - that is, someone takes the note or other token knowing that someone else will accept it. If the community of acceptors is small, the money is narrow. If the note is liable to be discounted, the money is soft. At the other end, the community may be wide, and the note always good for the face value. then the money is broad and hard. But it's all money. And it's all a bridge between what physical wealth I have now, and what I hope to get in the future. In that sense, it's a debt. It's a store of expected wealth. The OP is right in that if you can't exchange (realise) this expected wealth, the "money" is worthless. The unit of account is an independent property - it's a measure, but not an absolute one.

So the bottom line is that "money" is ONLY a store of future wealth. The distinction among the different kinds is the expected speed of exchange and the expected degree of acceptance.

[mb]

May I suggest a re-reading of Alchian's "Why Money?" JMCB, Feb 1977.

[Nick Rowe]

mb: Hmmm. I wonder if my memory still works?

IIRC, the main message of Alchian 77 is that the good for which everyone has low costs of estimating its value will become used as the medium of exchange. Didn't he also add something about specialist traders, who are good at estimating the value of other goods? I don't remember him talking about the store of value function.

How did my memory do?

[marris]

Nick, have you tried thinking of expiring money as "out-of-the-money" "buy-anything" options? They have exchange value up until expiry (because there's still a shot that they will become in the money). Their price falls as they approach expiry. Their price increases in periods of volatility [which makes sense because volatility increases the chance that you may be able to exchange them for something valuable].

You're one of the few economists I've seen knocking down Graeber's "arguments." Well done! I think you're analysis of Graeber's errors is correct. They are basically category errors. He is confusing *counter-factual* economic arguments [what you call disequilibrium arguments] (e.g. how a barter system would work) with historical ones. No doubt he is more comfortable with the historical analysis, but writing "huge barter economies never existed!" is an invalid counter-move.

Once you throw out the CMC/MCM mumbo-jumbo, Debt shrivels to another New York Times flavor of the month. It's that book that came out between between Blink and Keep Blinking.

[Mandos]

Why is that a category error?

[Nick Rowe]

marris: thanks! I haven't really read David Graeber, so I can't really say if I'm really knocking down his arguments. Bob Murphy did a more proper and thorough job a few months back. God only knows how Bob finds the time and energy. I have the suspicion that David Graeber is like a lefty version of Neill Ferguson? Knows his own stuff, but doesn't really get economics? If you told a political scientist that Hobbes etc. is obviously rubbish because we can't find a historical example of the State of Nature, he would laugh, then start speaking very slowly.

[Nick Rowe]

Mandos: I first learned about "category mistakes" in undergrad philosophy. It's from Gilbert Ryle. The visitor is shown the library, the classrooms, the offices, the students, the profs, and then asks "OK, I've seen all that, now can you show me the university?"

http://en.wikipedia.org/wiki/Category_mistake

[Nick Rowe]

Someone who took out his screwdriver to see if windows or Linux had been installed on a computer would be making a category mistake.

[Mandos]

I'm aware of the definition of a category error, I just don't see how marris' example qualifies as one, or even your example about political science qualifies as one. I would suggest that lack of a historical example of a State of Nature significantly diminishes its relevance to a discussion of anything in real life.

[Jon]

Nick and K,

I think you guys are a on strong tilt with your discussion of hoarding. Money being a store of value is precisely why it is hoarded and why there can be excess demand for money.

So I disagree with the thrust of the main post here, each part of the definition has a role:

1) unit of account: this means that value of money has a role in sticky prices. Since money is involved in every contract (possibly only in the breach), the current value of money versus the expected value of money has a macro effect.

2) medium of exchange: this means that money is in every market, and (nearly) the only markets are the ones with money and another good. Because of #2, an excess demand for money can cause a decline in output even without contracts embedded with expectations about the value of money.

3) store of value: money can be usefully hoarded. Thus there can be excess demands for money. Transactional demand for money (#2) is actually counter-cyclical. A decrease in transactions, ceteris paribus, leads to a decrease in the value of money versus expectations and vice-versa. Because of (#1), a change in the value of money changes output.

Conversely, hoarding demand for money (#3) is pro-cyclical. Money being a store of value is precisely why we need a CB to equilibrate the supply of money against the demand to neutralize the effect a fluctuation in the value of money would have given its role as the unit of account (#1).

[Jon]

Min writes:

But does it need to be a perfect store of value? Suppose, a la Gesell, that every Monday morning at 3:00 a. m. the gov't levied a tax of 0.1% on money deposits above $10. Money would still be a store of value, but one of decreasing value. Bad idea?

Why would you have the tax levied on deposits? Deposits are not base money. It is clear where the tax should be: it should be on reserves and on currency. We have just that tax, it is called inflation.

This is precisely why the interest on reserves policy is such a severe blunder. Suppose the Fed sticks to its original policy of keeping IOR 1% below the Fed Funds rate. Once the Fed Funds rate rises above 1%, the implicit inflation rate tax starts to decline and eventually goes negative (Hoarding earns a positive return).

[Nick Rowe]

Mandos: "I would suggest that lack of a historical example of a State of Nature significantly diminishes its relevance to a discussion of anything in real life."

OK. It depends on how you interpret Hobbes et. al. Is Hobbes writing history? Or is he explaining why we don't live in a State of Nature?

How would you explain why Lake Ontario is the same height on the Canadian side of the border than on the US side? I would say: "suppose it weren't. Suppose it were higher on the Canadian side. Then the greater pressure of water on the Canadian side would cause it to flow from the Canadian to the US side, until the levels were equal." And if someone said that it had never been observed to be higher on the Canadian than on the US side, I would say they are misunderstanding my explanation. I am explaining *why* it wouldn't be observed.

Jon: OK. But are you saying that store of value is one of the attributes of money, rather than a defining function?

[Mandos]

OK. It depends on how you interpret Hobbes et. al. Is Hobbes writing history? Or is he explaining why we don't live in a State of Nature?

Will have to punt on this question. I'm a SMURFLESMURFLE MUMBLEMUMBLEologist, I've never read Hobbes!

How would you explain why Lake Ontario is the same height on the Canadian side of the border than on the US side? I would say: "suppose it weren't. Suppose it were higher on the Canadian side. Then the greater pressure of water on the Canadian side would cause it to flow from the Canadian to the US side, until the levels were equal." And if someone said that it had never been observed to be higher on the Canadian than on the US side, I would say they are misunderstanding my explanation. I am explaining *why* it wouldn't be observed.

Yeah, I recall you used this example before, and I recall thinking it was very off, though I may not have objected loudly (or at all). I suspect that it would never occur to physicists to ask this, because there's a massive prior, well, category error in the question itself that no one would ever make: that political boundaries should have any effect on physical properties.

So, OK, you are going to perhaps get rid of the political boundary issue and ask why it isn't...what? What sort of boundary would make this a plausible way to hypothesize about physics? The 49th parallel? Just an abstract line. Some pair of diametrically opposite rocks? Why would you choose those?

And that's the point here. If money-as-exchange doesn't really form part of the money-origin story from a historical perspective, then it puts into question any sort of scientific hypothesizing about real economies based on the counterfactual. It becomes nonverifiable and very strange indeed. Most "real" science doesn't build hypotheses on counterfactuals, but prior plausibility. When the presuppositions of a line of reasoning have been demolished, science typically abandons that line of reasoning unless something comes around to resurrect it.

[Benjamin Cole]

Excellent post. Only Theo-Monetarists and Econo-Shamans worship paper currency as a store of wealth (when they are not genuflecting to gold).

Forget money as a store of wealth.

Market Monetarism offers a practical way forward.

I prefer that which works. We know deflation is a crappy way to live (see Japan), and we know the gold standard is for fools and their braying mouthpieces.

A sensible, disciplined, yet growth-orented monetary policy is best, in the real world. Where I live.

[anon]

"If money-as-exchange doesn't really form part of the money-origin story from a historical perspective"

But it does. The "historical perspective" is one where gift-exchange economies (but it would be better to call them equity exchange economies, as opposed to debt) broke down due to a combination of political and social factors (breakdown in trust, etc.). So money was used in order to replace gift exchange. Where's the difficulty?

[Mandos]

But it does. The "historical perspective" is one where gift-exchange economies (but it would be better to call them equity exchange economies, as opposed to debt) broke down due to a combination of political and social factors (breakdown in trust, etc.). So money was used in order to replace gift exchange. Where's the difficulty?

That, of course, is a different question, whether Graeber's (or Graeber-proxy's) definitions and distinctions make sense. Like Nick, I haven't read the book so I can't really say...

[Ravi]

If you stop thinking about money as a store of value, then its alleged "debasement" ceases to be a problem - unless you're a Theo-monetarist (to steal Benjamin Cole's phrase).

[Nick Rowe]

Mandos: OK, replace Lake Ontario with a U-shaped tube. Why is water the same level in both sides? I vaguely remember this question from hi skool physics: why does water find its level?

A quick Google gives me answer #2 from "Dr H" (so must be a real scientist) on Yahoo Answers.
http://answers.yahoo.com/question/index?qid=20080206072225AAdpmpq

The only way I can think of to test this is with an experiment.

The only experiment I know of is this one natural experiment. If an economist hadn't been there right on the spot, we would never have observed the very short barter phase.

http://worthwhile.typepad.com/worthwhile_canadian_initi/2010/11/the-economic-organisation-of-a-pow-camp-remembered.html

[mb]

from Alchian (p. 139) "This analysis explains the use of money, which good becomes money, why it is not necessarily also the store of value....."

[Lee Kelly]

This is a very interesting discussion about which I have much to say. It seems to me that monetary economists constantly forget to include money in their models of the economy, which is really very strange. However, it's very hard for me to pin down what this intuition is driving at--it's probably something to do with confusing money and credit, an implicit assumption that there is no qualitative difference between money and near monies. But whatever.

What I really want to say here is something very trivial: STOP CALLING IT A 'MEDIUM OF ACCOUNT'! It's 'medium of exchange' and a 'unit of account', surely!? That's all. Please continue.

[Nick Rowe]

Lee: it's a very picky distinction. If gold was money, for example, gold would be the medium of account and an ounce of gold might be the unit of account ;-)

[Min]

Nick Rowe: "But I disagree on it being about the paradox of thrift. I would say it's the paradox of hoarding (the medium of exchange)."

Yes, that's what I had in mind. Just my trick brain. Sorry.