Three (maybe four) main questions about money

What are the main questions in monetary economics? This post isn't really about the answers to those questions. It's about the questions themselves.

This is how I think about dividing up monetary economics into a small number of main questions, and how those questions are related to each other. Others may want to divide it up differently. Sometimes the questions you want to ask may depend on what sort of answers you have. And how those questions are related to each other may depend on how you answer them too.

This is partly a response to Robert P.  Murphy's EconLib article. I think about it a bit differently. In particular, I see the equation of exchange and the cash balances approach as just different ways of formulating the same question.

Here are my three (or maybe four) questions:

1. Why do people use monetary exchange? Why don't they use barter, or a central clearing house where all goods can be exchanged between all people at once (a Walrasian auction), or some other method of exchanging goods? Why do we (mostly) converge on a single good to use as a medium of exchange? Closely related questions would be: if there are two different media of exchange, what determines where one will be used rather than the other; how does one medium of exchange get replaced by another?

2. How much money will individuals choose to hold (what determines the demand for money)? How quickly will individuals choose to turnover money (what determines the desired velocity of circulation)?

Those sound very different, but I see them as being two ways of asking what is fundamentally the same question. Both are asking about the relation between stocks of money and flows of monetary expenditures. Both are asking: why don't people spend new receipts of money instantly so that velocity is infinite and the stock held is zero? Why is the desired inventory of money strictly positive and desired turnover of inventory of money finite?

This second question assumes the first question has already been answered. If people didn't use money, they presumably wouldn't want to hold a stock of money.

3. What determines the value of money? (What determines the price level?) What determines the rate of inflation (the rate of change of the price level)?

If you are a quantity theorist (understood in the broadest sense), your answer to this third question is closely related to your answer to the second question. Your answer to the second question gives you a theory of the demand for money. You add in a theory of the supply of money (which will depend a lot on what particular good is used as money and on the particular monetary institutions of time and place). Putting demand and supply together you get a theory of the value of money (the reciprocal of the price level). But if you are not a quantity theorists you may see the second and third questions as unrelated.

4. (I can't decide if this is really a separate question or just a part of the third question). What ensures that the equilibrium value of money is positive? If a good has a value of zero, people can't use it as money (question one), and so there will be no demand to hold a positive stock of it (question two), and so it won't have a positive value (question three).

If the good had some other use, apart from being used as money, it might still have a positive value regardless (provided the non-monetary demand exceeded supply at a price of zero). And yet most of us nowadays use an intrinsically useless good (like paper) as money. Does paper money have two different equilibrium values, and if so, what stops us observing the equilibrium in which the value is zero?

Those aren't the only questions you could ask. What are the effects of monetary policy? What is the best monetary policy? How valid are economic theories which abstract from monetary exchange (which is most of economics)? Etc.

Comments

[Ian Lippert]

Another question I have is under what conditions do financial assets start acting like money substitutes and how does that affect the above 3 questions?

[Nick Rowe]

Ian: Depends on what "money substitute" means. Some financial assets (like bank deposits) get used as media of exchange. That is part of question 1. "Under what circumstances will a promise to pay money itself get used as money?". Other financial assets don't get used as media of exchange, but might affect the demand to hold media of exchange. That's part of question 2.

There's maybe other ways of thinking about it.

[Tom]

5. How does expected lifetime net wealth affect V, the velocity of money?

I am convinced that the huge drop in wealth because of the housing price drop has had an anchor like drag on the economy -- and will continue to do so.

The difficulty in stimulating more "demand for purchases" so as to be closer to the prior trends is because so many individuals have purchase constraints based primarily on income but also influenced by wealth. The wealth effect has been too under-studied -- and will remain so, because there have been so few cases of big national wealth drops.

[Mike Sproul]

"But if you are not a quantity theorists you may see the second and third questions as unrelated."

Something we agree on! If I wanted to explain the value of corporate bonds, I certainly wouldn't be interested in the velocity of circulation of those bonds, or even the demand/supply of those bonds. I'd be interested in the assets backing those bonds, and nothing else. Same goes for money.

[Nick Rowe]

Tom: I see that question as part of 2. What determines desired V (or what determines the demand for money)?

[M.R.]

Nick, I think Ian's point may be a bit more subtle than that. It is sensible to say that the exchange-medium attribute is the defining characteristic of money. But it is of course widely acknowledged that other instruments function as near substitutes, even though they aren't a medium of exchange. Nor is this just a matter of "liquidity."

I happen to think that the terms "money market" and "cash equivalent" are not accidents. Is this just semantics? What are the characteristics of instruments that are close substitutes for the medium of exchange?

Somewhat relatedly, I would add one more large question for monetary economics: How should our monetary institutions be designed? I consider this question to be distinct from (and to analytically precede) "what is the best monetary policy," as policy must be implemented within a particular institutional framework.

[Nick Rowe]

Mike: yep! I wrote that bit just for you (OK, and a few others)! Same goes for refrigerators and all assets?

[Nick Rowe]

M.R. You lost me on the first bit. Sorry.

"How should our monetary institutions be designed? I consider this question to be distinct from (and to analytically precede) "what is the best monetary policy," as policy must be implemented within a particular institutional framework."

Good point. I like that question.

[Mike Sproul]

Nick:

"Same goes for refrigerators and all assets?"

Not sure what you mean. I sure don't think the value of refrigerators is affected by the velocity of refrigerators.

[Nick Rowe]

Mike: agreed. But does it depend on demand and supply, or just on the assets backing those refrigerators if you want to sell it back to the manufacturer?

[dlr]

Can I beat Mike to the answer? He will say of course refrigerator value depends on demand and supply. And so does money except the cost of the part of money that is consumed is close to zero because the cost of printing more money (unlike refrigerators) is close to zero. If refrigerators were free to produce they would be free, unless they came with assets in the freezer and then their value would be determined by backing and demand and supply would only contribute to the near-zero cost of their consumption value. The counter to this is that the cost of competing with the government's monopoly on money is actually enormous due to network effects and switching costs and some special government advantages, much like the cost of competing with incremental near-costless incremental copies of Microsoft Office but even worse. Mike is not persuaded by this.

[Lord]

Price level seems a little too limited. Not only are we interested in consumer price levels and inflation but asset price levels and appreciation. Yes, a fall in asset prices will increase money demand, or would you say the reverse as the assets lose their money like qualities? Asset prices are fundamentally expectations of future income flows, and while we rarely consider the asset price of our expectations of future earnings, those expectations must change significantly. So if the question is the value of money we also have to consider in terms of what. Even if money is stable in consumer terms, it may be volatile in assets terms.

[Nick Rowe]

dlr: I will leave Mike to respond!

Lord. This is a case where the questions you ask depend on your answers to those questions.

Suppose you believe money is neutral. (Most economists believe in long run neutrality). Then you would say that the ratio of (say) house prices to the CPI is determined by real variables independently of nominal variables like the overall price level. In that case, you can separate out the relative price of assets as something that is not part of question 3. (On the other hand, the relative price of assets may well affect the demand for money and the overall price level. Causation runs one way, but not the other.

In the short run, it's all different.

[M.R.]

"M.R. You lost me on the first bit. Sorry."

Yes, I wasn't very clear at all. Here's another way of posing the question about money-substitutes. It's a basic question about the design of monetary institutions. Suppose we like the idea of chartering special issuers of monetary instruments. We can call them "banks," and they issue "deposit obligations," under conditions established by the monetary authority. Any entity without a special charter is legally prohibited from issuing these instruments; it is a privileged activity.

Suppose we think this is a good way to design a monetary system. (Not everyone does, but some of us do ...) Does the same rationale also apply to the issuance of near-monies? I.e., should their issuance be confined to a special set of licensed entities, under conditions established by the monetary authority? Or should anybody be free to issue them, just like stocks and bonds? How should the line be drawn?

This isn't a metaphysical question about "what is money," but rather a practical question about the design of the monetary system. But I suppose it isn't really a first-order question, conceptually speaking.

[Nick Rowe]

M.R. OK. I get it now. Not really a first order question though, theoretically. (It's maybe a bit embarrassing for us theorists though if the most practically important questions aren't always the most theoretically important!)

[Joe]

Nick,

Here's two weird questions that have been bothering me.

This one was once brought up by Adap P. Here in the USA we have a recession and therefore an excess demand for cash balances. Yet we also have a little bit of monetary information, 2%. How is that possible? According to basic micro price theory if demand goes up, prices go up. Yet we have an excess demand for cash balances yet the price of money is falling because inflation is going up. How can we have an excess demand for money and simultaneously have the price of money falling>

Here's the same question in an another form. An increase in the money supply works through the HPP. People get the new money, have more than they need, and then spend it. However, we are in a recession and therefore everyone has an excess demand for cash balances. People are hoarding money. Doesn't this imply that an increase in the money supply during a recession will simply be hoarded and not spent? THIS implies that the HPP only works once the economy is at full output because only at that point is every individual's money demand satiated. In a recession if the CB increases M it will keep getting hoarded by everyone and the HPP will only occur once everyone's money balances are satiated. Yet, we know for a fact that an increase in the money supply in the middle of a recession does cause the HPP.

These two questions have driven me nuts.

My hunch, is that the answer lies in one of three possibilities, 1) stocks versus flows, 2) aggregate decisions versus individuals, 3) the fact that the price of money is sticky.

Best regards,

JoeMac

[Mike Sproul]

Nick and dlr:

"does it depend on demand and supply, or just on the assets backing those refrigerators if you want to sell it back to the manufacturer?"

Still not quite clear what you are asking, but let me switch from refrigerators to gold.

First of all, gold has value because of supply and demand, and part of that demand comes from its use as money. This creates a slight inefficiency as the monetary premium on gold makes gold too expensive for other uses. As paper IOU's start to displace gold, this premium shrinks. Thus, right off the bat, the backing theory looks wrong, because the backing theory says that issuance of adequately backed IOU's won't cause inflation, and there they are, causing gold to lose value, i.e., causing inflation. But once gold has lost its monetary premium, further issuance of adequately backed gold IOU's can't cause gold to lose any more value, and furthermore, as long as each 1 oz. IOU has 1 oz. worth of assets backing it, the issuance of IOU's can't cause the IOU's to lose value relative to gold.

Second of all, paper IOU's pegged to 1 oz of gold are not commodities, and hence are not subject to the laws of supply and demand. Those paper IOU's are liabilities of their issuer, and are therefore subject to the rules of accounting. Those rules say that an IOU that is backed by assets worth 1 oz of gold will be worth 1 oz of gold, and that will remain true whether there are 100 IOU's in existence or 100 million.

[Nick Rowe]

Joe: I think those are two different questions.

1. If there's an excess demand for money, and that causes people to stop spending (negative hot potato), so there's a recession, why don't prices fall immediately, or at least start falling immediately?

That's a question about the short run Phillips Curve, sticky prices, and sticky inflation. Nobody really knows the answer to that question. We figure it's got something to do with the costs of changing prices, the difficulty of coordinating millions of individual price changes, and expectations, and maybe money illusion. All we know is that this is what seems to happen. And we try to cook up different theories of how prices get set to try to fit the facts as they appear to be.

2. Start in equilibrium. Then assume prices are fixed (or at least sticky). Then assume something changes to cause an excess demand for money. So people want to spend less to hoard more money (negative hot potato). Because prices don't fall, the quantities of goods sold falls instead. As the quantities of goods sold falls, people's incomes fall, and their demand for money falls too, until we get to a point at which income is low enough that they don't want to try to hoard more money, and (on average) want to spend what comes in. Is that point an "equilibrium"? Is there an "excess demand for money" at that point? Well, yes and no. People can't sell as much stuff as they want to sell (unemployed can't sell all their labour, firms can't sell all their goods, etc.), so it's not really an equilibrium. But, given the quantity of stuff people are actually able to sell it is an equilibrium, with no excess demand for money.

Let me put it another way. People want to sell more stuff in exchange for money, but can't, so it looks like an excess demand for money (disequilibrium) when you want to sell things, but people don't want to buy any less (or more) stuff for money, so it looks like supply=demand for money (equilibrium) when you want to buy things.

So what you want to do to get out of the recession is to create an excess supply of money given that reduced level of income, so people buy more things and you get a hot potato, so people can also sell more things and income increases.

Don't know if that helps. Hope it does.

[Nick Rowe]

Mike: "Second of all, paper IOU's pegged to 1 oz of gold are not commodities, and hence are not subject to the laws of supply and demand."

Suppose I had a theory which said that the prices of all goods, whether refrigerators, bonds, money (backed or unbacked), was subject to the laws of supply and demand? Wouldn't that be a better more general theory? And that is my theory. The only difference is that sometimes some goods may have perfectly elastic supply curves or demand curves, making them a limiting case of that general theory.

A commercial bond (unless there's something special about it) will have a perfectly elastic demand curve.

A convertible money, 100% backed, will have a perfectly elastic supply curve.

[Max]

Does MV=PT imply that if the central bank sets (base) money to zero (by eliminating paper money and reserve requirements), that the economy collapses?

Presumably nobody believes it would...we would conduct business as usual, just without having paper money as a payment system option. The black market would have to use something else.

So why is it so exotic to think of money purely as a reference unit, not as a *thing* that circulates?

Is leaving paper money out of economic models any different than leaving out Visa, Mastercard, PayPal, etc? They are all competing payment systems.

[Mike Sproul]

Nick:

Yes; in a world where R=5%, a commercial bond promising $105 payable in 1 year will sell today for $100, and you could say that the demand for that bond today is horizontal at $100. But it's the laws of finance and accounting that put that demand curve there. Is your supply and demand theory of bonds better than the theory of finance? I don't think so, and I'm pretty sure the entire finance and accounting professions would line up behind me on this one.

"A convertible money, 100% backed, will have a perfectly elastic supply curve."

Ooooh! Be careful what you admit to there. If each dollar is backed by various assets worth 1 oz, then the money supply curve is horizontal at 1 oz/$ and no amount of change in money demand will move the dollar from 1 oz./$. It's the 1 oz. of backing that put the supply curve there, so you're perilously close to admitting that backing determines the value of the dollar.

Now, you'd probably object that you only admitted this for the case of a convertible money. But what does convertible mean? Here are some possible answers:
1. The bank gives you 1 oz for each of its dollars 24/7 with 100% certainty
2. The bank gives you 1 oz for each of its dollars during normal business hours with 99.99% certainty
3. The bank will give you 1 oz for each of its dollars in 60 days, with 99.9% certainty
4. The bank will give you 1 oz for each of its dollars in 60 years, with 95% certainty
5. The bank will liquidate in 2 centuries or so, and probably redeem each of its dollars for whatever it has in its vault at the time.

[Nick Rowe]

Max: "Does MV=PT imply that if the central bank sets (base) money to zero (by eliminating paper money and reserve requirements), that the economy collapses?"

Not if M /= base , because there's an alternative money out there: demand deposits.

"Is leaving paper money out of economic models any different than leaving out Visa, Mastercard, PayPal, etc? They are all competing payment systems."

Visa and Mastercard are not media of exchange for most of us. They just give you credit, so you can postpone final payment. If an individual can get both debits and credits on MC, so you both buy and sell things with MC, and only have to pay the difference, then they are starting to become media of exchange. I'm too old to understand PayPal.

Mike: "Yes; in a world where R=5%, a commercial bond promising $105 payable in 1 year will sell today for $100, and you could say that the demand for that bond today is horizontal at $100. But it's the laws of finance and accounting that put that demand curve there. Is your supply and demand theory of bonds better than the theory of finance? I don't think so, and I'm pretty sure the entire finance and accounting professions would line up behind me on this one."

Yes! And the entire finance and accounting professions shouldn't be allowed out on the streets without adult supervision!

When you say "...in a world where R=5%,..." you are (implicitly) assuming that this bond too will have a demand price (demand rate of return) exogenously fixed at 5%, regardless of quantity supplied. So you are (implicitly) assuming that that bond has a perfectly elastic demand curve at 5%. That would only be true if this bond were a perfect substitute for other bonds, and if this bond were a small part of the bond market. Yep, if Swift Mattawa canoes were a perfect substitute for all other canoes, and were a small part of the total canoe market, then the demand curve for Swift Mattawa canoes would be perfectly elastic too, and the price would be independent of supply.

"Now, you'd probably object that you only admitted this for the case of a convertible money. But what does convertible mean? Here are some possible answers:.."

The demand for refrigerators depends on the expected future price. Expected future price depends on expected future demand and supply. Suppose the manufacturer of the refrigerator says that maybe, just maybe, it might buy the fridges back at some future date at some fixed price, if anyone wants to sell. That would affect expected future (stock) supply of fridges, which would affect expected future price, which would affect current demand, which would affect current price.

Same with money as with fridges.

Demand and supply, all the way!

[rsj]

Demand and supply, all the way!

OK, and when the CB is offering to lend money at 0%, and still there are no takers, what do your supply and demand curves tell you? Do they suggest an excess demand for money? Are the financial markets not clearing?

It seems to me that you are assuming an excess demand for money because you rule out all other possible causes by assumption.

And that creates the situation in which you conclude free money (leaving total wealth unchanged) is still too expensive.

Me and another guy in the back are saying that households want more wealth, not more money per se. That is, given their current portfolio, they don't really want to convert more of it into deposits, or whatever cash/cash equivalent you happen to care about. They are able to create as much money as they need.

But they are not able to create more wealth. In aggregate they are, buy bidding up assets, but individually they are not. Individually they would like their nest eggs to be greater, and their discounted labor income to be higher, and they want to stop worrying about getting laid off.

Even better, what they really want is a better distribution of wealth as well as a better distribution of income, so we don't have a situation in which a small group of people with enormous net worth prefer to sit on assets with zero return rather than spend, while everyone else is worried about paying rent and can't save anything at all.

If you really believe that households have the right amount of wealth, but too little money, then all you need to do is a portfolio swap. Supposedly the financial markets aren't clearing.

But if you believe that the average household has the right amount of money, but too little wealth, then you want to start redistributing wealth and income on the double. That's another portfolio swap, in some sense.

[rsj]

FYI, here is a nice article "Why do the rich save so much"?
http://ideas.repec.org/p/jhu/papers/388.html

Of the top 1% (by wealth), wealth to income ratios rapidly increase with age.
Among the findings -- elderly wealthy couples without children save more than elderly wealthy couples with children

And there is Plato's discussion of "false pleasures", in which wealth is a fetish object that is an end to itself, rather than a means to an end.

[Nick Rowe]

rsj: there's still plenty of wealth. The human capital, physical capital, land, and technology, are still all there. No earthquake destroyed any of that wealth. The problem is that the goods that wealth could produce are not being traded. Why not? Don't unemployed car workers want houses, and unemployed construction workers want cars? Why don't they just do a swap? "You build me a house and I'll build you a car!" Hmmm. Maybe it's because we don't do barter, and use monetary exchange instead?

"If you really believe that households have the right amount of wealth, but too little money, then all you need to do is a portfolio swap. Supposedly the financial markets aren't clearing."

Here's the answer to that question.

[rsj]

"The human capital, physical capital, land, and technology, are still all there."

Yes, if we lived in an ideal riskless world with perfect future foresight and infinite calculational capacity, we would have the same potential, and would be continuously realizing that potential.

But if I had a house worth $200,000, that I was planning on selling when I retired. That house is now worth $100,000, so I just lost $100,000 of wealth. Now I need to save more in order to fund my retirement.

Telling me that I have the same house doesn't help, right?

At the same time Zuckerberg just saved $19 Billion. It is sitting in a vault somewhere, so he is not spending either. If he would "only" save, say, 1 Billion, his consumption habits would be unchanged.

But for 50 million people whose net-worth was fell from say $100,000 to $50,000, their consumption habits are certainly changing.

And add to that a lot of people who are not worried about being laid off, so that they discount their future earnings at a greater rate.

So while we have the same "stuff", those who are counting on being able to sell their labor now believe that in the future, they will be able to sell it for less. And those looking to sell their property, are more uncertain as to the future price of that property. That means that their wealth is lower today.

The above would be true in a barter economy or a monetary exchange economy -- it would be true in any economy in which there is uncertainty and investment.

[Nick Rowe]

rsj: you know it's not excessive saving that causes a drop in AD. It's excessive hoarding of the medium of exchange. Ive been over this so many times.

[K]

We have. But rsj is making a really good point that was also brought up last time we discussed this. It didn't seem like we concluded.

[Alex Godofsky]

Does paper money have two different equilibrium values, and if so, what stops us observing the equilibrium in which the value is zero?

We do observe that equilibrium. Every time you see an object that could potentially be used as money but isn't, you are observing that equilibrium. It's similar to the anthropic principle; you're only going to ask that question of a potential money that happens to be in the upper equilibrium.

That still leaves open the question of why this potential money instead of any of the others.

[Dan Kervick]

it's not excessive saving that causes a drop in AD. It's excessive hoarding of the medium of exchange.

What's the difference, Nick? The medium of exchange is just one kind of asset; it's an undifferentiated claim on the output of the future. The value of that claim is fluid and unpredictable, but if someone chooses to hold onto it for the time being it is because they are convinced that the value of that fluid claim is greater than the value of the most valuable of the various kinds of stuff they could exchange it for immediately. That's saving.

If the collective result of all of that saving is to let capital resources go to rot and diminish the value of future output below the level of the savers' vague estimation of it, then the saving is collectively self-defeating. That's excessive saving.

[Alex Godofsky]

Dan: because saving in any type of asset other than money results in the production of consumption or investment goods. Only saving in the medium of exchange does not.

[Dan Kervick]

Alex, I don't see at all why that would be true in general. But apart from whether it is true or not, I don't see the logical connection with my claim. If money is just one kind of asset, then "hoarding" it is just one way of saving an asset. It seems to me that if you possess any kind of asset at T0, and you neither consume it nor exchange it nor invest it in production between T0 and T1, then you will still possess it at T1 and thus will have saved it through the period T0 to T1.

So excessive hoarding of the medium of exchange is a form of excessive saving, and it makes no sense to say that the demand shortfall is caused by the former but not the latter.

[Nick Rowe]

Alex: "It's similar to the anthropic principle; "

Neat!

[K]

Alex: "Only saving in the medium of exchange does not."

I don't get it. Demand deposits fund bank loans just like savings deposits, bank bonds and bank stocks. Or do demand deposits not qualify as "medium of exchange?"

[Saturos]

I agree with the idea that the Equation of Exchange and the supply and demand for money-balances are closely related, almost "flipsides" of each other. But I disagree with the idea that velocity is infinite if people don't want to hold any stocks of money. People might get rid of balances instantly, and money flash around the economy in exchanges at lightning speed - but then have to come to a rest as all the year's production is bought up and allocated to its final users. The quantity of money limits how high prices can rise, as you can't bid for goods at prices that exceed the quantity of money in circulation, can't bid for goods with money that hasn't circulated back around to you yet, except on credit (but borrowing constrains other spenders). In other words, though people may have an endless desire to exchange money for goods and deplete cash balances, and may always be able to do so, they are limited by people's willingness and ability to exchange goods for money, which is finite in a given period. So you could have the money race around the economy and then come to rest, not because people want to hold it but because there is nothing left to buy this period, giving a finite average turnover per period.

"If a good has a value of zero, people can't use it as money"

But doesn't the value of the good come from the fact that people are willing and able to use it as money?

K, I think he means that hoarding money balances does not promote circulation of money and spending. In a sense holding demand deposits is not hoarding because the fractional-reserve banking ensures that the reserves circulate even while you are "holding on" to them.

[J.V. Dubois]

Nick: I think you forgot one important question: "What is money"? Many other questions (and answers) depend on how you think about this.

[ezra abrams]

quote
why don't people spend new receipts of money instantly so that velocity is infinite and the stock held is zero? Why is the desired inventory of money strictly positive and desired turnover of inventory of money finite?

Is this a joke, or some inside econ sarcasm ?
Is this some pedagogical point, put up on the web without warning that it is a teaching device for students ?
Every day, if you are at least middle class, you are told, by all sorts of experts, to save.

quote
Why do people use monetary exchange? Why don't they use barter,
Again, is this some sort of inside econ humor ? what, exactly, am I supposed to trade to the car dealer with ?

a few weeks ago I posted on Brad Delongs website an excerpt from a history of hte civil war, and the horrible inconvience people faced without money - the problem the housewife had in going to buy some "fixins"

If I'm missing the point, I apolgize.

[Alex Godofsky]

I don't get it. Demand deposits fund bank loans just like savings deposits, bank bonds and bank stocks. Or do demand deposits not qualify as "medium of exchange?"

Because you had to exchange the medium of exchange for the asset, someone else has now got it. That person has to then either hoard it or exchange it for someone else. If it keeps getting exchanged eventually it should end up having to be spent hiring someone to produce a consumption or investment good - presumably it won't indefinitely be used solely to purchase existing assets.

Now, saving in the form of certain asset purchases frequently looks very similar to hoarding the medium of exchange, because as a pretty direct result of the purchase you cause the other party to hoard the money you gave them (e.g. bank deposits).

[JP Koning]

JV makes a good point. Usually one starts with the "what is money" question which evolves into the classical medium of exchange, unit of account, and store of value triumvirate.

[Nick Rowe]

Hmmm. JV does make a good point. Maybe the definition of "money" should be question 0, because it's logically prior to any other questions? (Though again, how you answer that question may depend on your other answers, because some emphasise medium of exchange, and others medium of account?)

[Nick Rowe]

ezra: you are missing one point, but getting the other.

"a few weeks ago I posted on Brad Delongs website an excerpt from a history of hte civil war, and the horrible inconvience people faced without money - the problem the housewife had in going to buy some "fixins""

Yep. That's exactly how I would start out answering that question. But I want to know more about that inconvenience. Because people do sometimes use barter, so it can't always be more inconvenient. And maybe barter isn't the only alternative. And which particular good will they use as money? And how can lots of different individuals all coordinate to "agree" on using that particular good? Your answer is just the beginning.

You missed the point about the demand for money. We can save in many ways, not just by holding more money. We can even hold more money without saving, by borrowing, or by selling an asset. And people hold money even when they are not being exhorted to save.

[Nick Rowe]

ezra: this is slightly off-topic, but I would be interested in hearing a little more about what people did without money in/after the (US?) civil war. Did they barter? Did they start to use some other good as money (like WW2 POWs sometimes used cigarettes)?

Saturos: "But I disagree with the idea that velocity is infinite if people don't want to hold any stocks of money. People might get rid of balances instantly, and money flash around the economy in exchanges at lightning speed - but then have to come to a rest as all the year's production is bought up and allocated to its final users. The quantity of money limits how high prices can rise, as you can't bid for goods at prices that exceed the quantity of money in circulation,..."

You lost me there. Production is a flow, and the stock of money is a stock. If both are finite, and non-zero, we get a finite V when we divide one by the other. And an infinitesimally small stock of money could purchase annual production at very high prices if it circulated fast enough.

"But doesn't the value of the good come from the fact that people are willing and able to use it as money?"

Yes, but there's also a second equilibrium (or seems to be) in which the good has no value and so can't be used as money and so isn't demanded.

[Ritwik]

Nick, I think the best answer to the 0th question of 'what is money' is in Perry Mehrling's framework of the hierarchy of money. http://www.ieor.columbia.edu/pdf-files/Mehrling_P_FESeminar_Sp12-02.pdf

[Mike Sproul]

Nick:

Naturally, if fridge makers offer to buy back their fridges at some future date, that increases the price of a fridge now. More backing=more value.

And if some paper IOU is backed by 1 oz worth of stuff, then you can draw the supply curve of IOU's horizontal at 1 oz. If the IOU lost half its backing, the supply curve would then be horizontal at .5 oz. It sure looks to me like "supply and demand all the way" could just as well be written as "backing all the way".

Also, thanks to you, I am starting to appreciate the incomparable awesomeness of Swift Mattawa canoes. It won't do me much good in the deserts of Los Angeles, but I can dream.

[rsj]

Savings does not create investment. Savings is the result of investment. When savings demands exceed investment demand, nominal incomes adjust so that savings demand is equal to investment demand at the lower level of income. When investment demand exceeds savings demands, nominal incomes adjust so that the two are equal.

That also means that real incomes adjust.

That is true irrespective of how you allocate your savings among various forms of paper claims, the purchase of none of which causes investment to occur.

[Alex Godofsky]

Mike: but we observe people use currencies that aren't backed by anything at all. I don't mean regular fiat currencies, which you might say are backed by the CB's balance sheet; I mean the little local scrips that pop up when the supply of the normal medium of exchange becomes too constrained.

[Mike Sproul]

Alex:

Give me an example. Every one I've ever heard of was either expected to be redeemed for something of value sometime in the future, or else it was an outright scam that lost value as soon as the suckers realized it had no backing.

Anyway, every major currency I know of is clearly stated to be the liability of the bank that issued it, and as such, it is just as clearly backed by the bank's assets, just like bonds are backed by their issuers' assets. People see that dollars are not convertible into gold, and jump to the conclusion that the dollar has no backing. But 'inconvertible' does not imply 'unbacked'. They might just as well claim that inconvertible bonds are unbacked.

[Peter N]

Two things:

1) on the original topic, it seems to me the first item of business would be a definition of money, so we can all be talking about the same thing. This should be a useful definition, that is one which you are committed to use whenever you use the word money. Otherwise the definition is useless. I'll second J. V. Dubois who posted while I was composing this.

2) "you know it's not excessive saving that causes a drop in AD. It's excessive hoarding of the medium of exchange. I've been over this so many times."

This is much too glib. It and the article you refer to seem to be based on two theories, neither of which I believe are true. First:

"The supply of antique furniture is fixed. Either the price of antique furntiture rises to equilibrate the market, or it does not. If it rises, then the quantity of antique furniture demanded falls back to its original level, and people decide to buy newly-produced furniture instead, so there's no recession. If it does not rise, then people will be unable to buy the antique furniture they want to buy, because nobody wants to sell. Unable to buy antique furniture, people have to buy newly-produced furniture instead, so there's no recession."

If you apply that to the housing bubble, you can see the problem. People will buy because the price has increased, taking it as a signal that the trend will continue. You are confusing antique furniture you sit in with antique furniture as an investment. A collector probably never let's people sit in the furniture at all, lest it be damaged. I have lived with these sorts of collectibles, and I know whereof I speak. For instance my mother had a number of antique pitchers. They never held fluid. We had ordinary pitchers for milk or lemonade.

"Land, old houses, government bonds, whatever. Either the price rises until people stop wanting to buy it, or it doesn't rise, and they can't buy it, so they buy something else instead. And no matter what they try to buy instead (unless it's hoarding money) the only thing whose supply can expand to meet that demand is newly-produced goods and services. "

The supply doesn't need to expand, because

a) the price can rise instead. And this is good in an investment?!

b) supply can come out of the woodwork. There is a huge stock of assets which are available if the price is right. Think oil. The supply is unlimited; only the supply at a given price is limited. Raise the price, and marginal wells become profitable.

Finally, one person's loss is not another person's gain.

You may not have realized this, but the entire reduction in consumer debt hasn't come from increased repayment. It's come from default. The defaulter's gain is nominal, and in a non-recourse state, it's 0 by definition.

Unrealized losses aren't unreal losses. If 50 million houses have lost a large part of their value, that doesn't mean there have been 50 million sales at a loss. Such losses are still very real. For instance home equity is one of the chief means of financing new businesses. And a fall in prices reduces sales, because either owners can't afford to sell (they're under water) or they refuse to accept their loss. By supply and demand a massive decrease in prices should spur demand. It hasn't and rarely does.

I think part of the problem is using GDP to mean spending. It isn't, and the difference has consequences.

Finally, unlike energy, there is no law of conservation of money.

[Nick Rowe]

Ritwik: from my previous (lighter) reading of Perry Mehrling on that subject, you are probably correct.

Mike: if there is less than 100% backing, the supply curve may be reverse-L-shaped. In fact, there could be many different shapes of supply curves, depending on how much and at what price the issuer would be able and willing to buy back notes. (The amount of backing, whether 100% or less, would determine the able part of the able and willing.) Current willingness and ability to buy back affects current supply, and current price. Expected future willingness and ability to buy back affects expected future price, and hence current demand. It's supply and demand all the way! Backing only matters insofar as it affects supply and demand. Same as Marginal Cost and Marginal Benefit, which only matter for price insofar as they affect the supply and demand of canoes.

I met a guy while paddling the Gatineau river near my home who said he had paddled the Colorado. I test-paddled the demo Mattawa on the Oxtongue River, and my own new canoe on Meech Lake (a constitutionally important lake for Canadians). But haven't done a real trip yet.

rsj: "Savings does not create investment. Savings is the result of investment."

Whoah! That's linear thinking! Desired saving and desired investment are co-determined simultaneously in equilibrium along with interest rates and NGDP. The exact nature of that co-determination depends, inter alia, on the monetary policy regime.

[Victor Galis]

In response to this bit:

"This second question assumes the first question has already been answered. If people didn't use money, they presumably wouldn't want to hold a stock of money."

Actually, it seems to me that one of the primary reasons we want to use to money is to have a store of value. I want to be able to exchange my labour today for food on Sunday. (This is harder in a barter society.) Furthermore holding a stock of money is a way of dealing with uncertainty about future needs. This doesn't go away in a barter society.

Holding a stock of money guarantees that we have the means to purchase whatever we need when unforseen circumstances arise. In a barter society without a perfect market where I could immediately exchange any good for any other good, I would have to either stockpile all sorts of emergency supplies or stockpile the goods most readily tradeable for other goods (which is just a less efficient version of holding a stock of money).

[Peter N]

The unit of account is the key attribute of money because there are numerous historical examples of unit of account only money. Either the money is notional like Carolingian shillings and pounds, or it is symbolic or temporary like the stone money of Yap. The larger pieces of stone money never circulate, since they are much to heavy to move, in fact, that is part of what makes some stones valuable.

The store of value and medium of exchange aspects are, past a certain point, in conflict, because of the opportunity costs of holding one form against another.

[rsj]

he exact nature of that co-determination depends, inter alia, on the monetary policy regime.

It depends on the distribution of wealth, monetary policy regime, fiscal policy, desire of debtor households to increase borrowing, whether firms are willing to understake risky investment, why people save, expectations of future profits, the birthrate, and what the Chinese are doing :P

And in some cases, the monetary policy regime is either impotent, or less effective than the others.

[Peter N]

You have to be careful appealing to supply and demand.

1) There's no necessity for the supply and demand curves to be independent or stable.

2) There's often a possibility of short circuiting, that is manipulating the curves. Business schools teach courses in this.

3) Supply and demand can fail in the face of perverse incentives where cost has been shifted (like medical costs).

4) Supply - a stock can behave differently from supply - a willingness to supply, which is a function.

5) The shape of a demand curve doesn't have to be very well behaved (e.g. monotonic), nor need it be solely a function of price (even allowing for supplies with different utilities).

and this list isn't exhaustive.

[Nick Rowe]

Peter N.: "The larger pieces of stone money never circulate, since they are much to heavy to move,.."

Are you sure that's right? *Ownership* of the stones can circulate, even if the stones themselves don't. We don't always have to physically put money in our pockets. Simply saying "I will pay you my stone, the third one on the right is mine, for your cow." should do it. Much like a cheque.

Houses don't move either, but they still "change hands". IIRC, the German Imperial authorities once confiscated all the Yap stones, by painting a black cross on them, to signify ownership.

Victor: there are lots of different stores of value. Money is a small part of our total wealth.

[Peter N]

The Murphy article is a beautiful distillation of much of what's wrong with modern economics as taught.

"The original and more intuitive version of the equation is MV = PT, where M stands for the total quantity of money in the economy, V is the "velocity of circulation" (meaning how many times, on average, a dollar bill changes hands for the time period in question), P is the average price of a transaction ("price level"), and T is the total number of transactions."

"The problem with stating the equation of exchange in this form is that the term T includes all transactions, including the sale of previously-produced goods and even of assets... However, economists typically don't welcome this result because they want to exclude financial transactions from the analysis in order to focus on production."

"Consequently, it is more common nowadays to see the equation expressed in the form MV = PQ, where Q stands for newly-produced goods and services ("real output") during the period in question, and where both V and P are restricted to transactions involving newly-produced goods and services. In this updated form, the equation of exchange can be very useful to illustrate certain relationships."

So M = PQ / V and M = PT / V and PT / V = PQ / V. But what does this mean? Nothing. The assumption that you can separate the "productive" economy from the total economy and deal with it in isolation is nonsense. It's no more reasonable than trying to define the economy of all left handed people and thinking that the result was of any use.

"For example, the equation shows that if the quantity of money in circulation doubles while velocity and real output remain constant, then the price level doubles, too."

It does nothing of the kind. Given the implicit assumption here that M and V in the two equations are the same, we get P T = P Q, in which case we can't be "restricted to transactions involving newly-produced goods and services". So M and V must be different, but then why does an accounting identity in M and Vp imply an actual economic relationship with another identity in M and Vt. It doesn't.

Unfortunately, there's more to come.

"In the cash-balance approach, the economist recognizes that in equilibrium—though it may be a fleeting condition—it has to be true that every individual is satisfied with the amount of money he holds."

"Having thought through the necessary and sufficient conditions for a typical individual to be in equilibrium with respect to his cash holdings, the economist finishes the analysis by saying that the sum of all desired cash balances must equal the total quantity of money. In equilibrium, it must be the case that the community collectively desires to hold the exact amount of money in existence at that moment—no more, no less."

This is underwear gnome logic. The economist begins the analysis and ends the analysis. What about the middle part where all these fleeting equilibria occur at the magic moment when we aggregate?

Equilibrium is one of the most abused concepts in economics. Realistic models never produce a single stable equilibrium. No one ever has or will ever prove that the real economy is in equilibrium, because, of course, it isn't.

Given this result, the author shows how various economic theories might be demonstrated. However, since it is well known that given a single false statement in logic, it is possible to prove anything whatsoever, it's not a very impressive showing.

This is undergraduate toy economics, a genre that has outlived its usefulness (if it ever had one) and has become outright dangerous.

In the words of Einstein ""Everything should be kept as simple as possible, but no simpler."

[Peter N]

The stone money of Yap seems to be a form of special debt money, of a sort discussed by Graeber. It doesn't circulate, it acts as a debt marker. The debt is repaid, and the money ends up back with the original owner. Graeber has all sorts of examples of unusual forms of money.

[Greg Ransom]

Some core questions are missing.

1. How does the changing flow or demand for money effect production & consumption choices?

2. Why do different polities have different monies, and what establishes the relations between these different monies and what consequences do these changing relations have on production and consumption choices.

3. How do laws governing various money regimes come into being, and who do these laws benefit, and who do they harm?

These seem to go beyond your 1-4, but are of central interest when one turns ones attention to question raising patterns involving money.

[Greg Ransom]

A classic sort of pattern giving rise to scientific inquiry and causal explanation is the attempt to account for the origin of a thing, eg the source of the origin of species or the origin of biological adaptations.

This question is a bit different, and closer to a genuine empirical scientific problem than is your own #1.

Often it is helpful to go back to the original patterns that raised questions & to the theoretical constructs that helps us see those patterns, or that made patterns problematic, to uncover and highlight and flesh out the deeper, empirical scientific problems to be explain, ie the problems that a capable of genuine explanations in terms of causal mechanisms.

As much as anythings, Darwin's great achievement was in perceiving and recasting the problems to be addressed in biology -- problems which came to light in tandem with the causal mechanisms that could account for the patterns imagined without reference to magic or divine intervention.