Draft notes for teaching Intro Money

I always suffer self-doubt when I teach the Money part of Intro Economics. Perhaps I'm over-thinking it, and it would be better for my students if I told myself to shut up, and just give them some simple clear story. But how to get it simple and clear, yet not horribly wrong or incomplete?

I'm writing these notes mostly for myself. (To try to organise my thoughts, because I don't actually use notes when I teach.) But I might as well do it in public. Read at your own risk. Because I haven't got it right yet.

Preamble.

When my kids were little they asked me what Economics was. I thought for a bit, then lied to them. "It's all about money" I said. That simple answer works very well at one level. And works very badly at another level.

It is now nearly half way through the second term of Intro Economics. We have spent a lot of time talking about the demand and supply of different goods. When we demand something we nearly always buy it with money. And when we supply something we nearly always sell it for money. So money has been in the background implicitly all along, as the other side of almost every exchange. But we haven't actually talked about the supply and demand for money. Which is very strange, when you think about it.

What is money?

Different peoples at different times have used very different things as money. Shells, metals, cigarettes, bits of paper, etc.. If people use something as money, it is money, for them. That's a "functional definition" of money.

People use money as:

A medium of exchange. They sell everything else for money, and then use that money to buy everything else.

A unit of account. They talk about prices in terms of money.

Nearly always, those two defining functions go together. It's easy to understand why. It's more convenient. If the supermarket wants you to buy food with Canadian Dollars, they will put Canadian Dollars on the price tags. If they put US Dollars on the price tags, you would need to look up the exchange rate and do some extra arithmetic to figure out how many Canadian Dollars you need to pay.

Textbooks normally add a third function: people use money as a store of wealth. It's true that a medium of exchange is always used as a store of wealth to some extent, because there's always a time-lag between getting money and spending it again. But lots of other things are used as a store of wealth too, and many are better than money at this job, so this function doesn't help us define which good is used as money. Being a medium of exchange and unit of account is what makes money different from all other goods.

Why monetary exchange?

Suppose you were visiting some unknown country, and you wanted to figure out whether the people there used a medium of exchange, and what particular good they used as a medium of exchange.

You could make a table, with a list of all the different goods for the columns, and the same list of all the different goods for the rows. So it's a square table, with n columns and n rows, if there are n different goods. If you observe someone exchanging apples for bananas, you put a tick in row A column B. (Obviously you can ignore the main diagonal, because people don't exchange apples for apples, unless they are different varieties of apples, and you can ignore everything North-East of the main diagonal, because it's just a duplicate.)

In a pure barter economy, where any good can be exchanged against every other good, you would have ticks everywhere. With n different goods, there are n(n-1)/2 functioning markets.

In a pure monetary exchange economy, where apples (say) were the medium of exchange, there would be ticks everywhere in the A column or row, and no ticks anywhere else. With n different goods, there are n-1 markets, where in each market apples are exchanged for one of the other goods.

A monetary exchange economy has a lot fewer markets operating than a pure barter economy, for n > 2.

Most economies are a lot closer to the pure monetary exchange economy than they are to the pure barter economy. That's a puzzle. Why? If you were a banana grower who wanted to swap bananas for carrots, why wouldn't you just do that? Why would you first sell your bananas for money, and then sell that money to buy carrots? Why use two exchanges when one exchange would do the same job?

The reason is that you would need to find someone else who wanted to swap carrots for bananas, and that might be difficult. You can even imagine circumstances where 2 person barter is impossible. Like if the apple growers wanted to eat bananas, and the banana growers wanted to eat carrots, and the carrot growers wanted to eat dates,...and the Z growers wanted to eat apples. You would need to get all 26 people in the same place at the same time to do a 26-way swap.

Or you could use money.

A pure barter economy is a very unstable arrangement, so it is not surprising we rarely observe barter, just like we rarely observe eggs standing on end. Suppose one good, say apples, is just slightly more liquid (easier to buy and sell) than any of the others. People would tend to use apples as a medium of exchange, if they couldn't do a direct swap. And that would make apples even more liquid, because more people would want to buy and sell apples. So it snowballs, until apples get very easy to buy and sell, and everything else is bought and sold for apples.

Which good gets used as money?

Almost any good could be used as money, but some would be easier to use than others. So those goods are more likely to get the snowball rolling, and be used as money. Used cars would not be easy to use as money. Each used car is different, and it takes a skilled mechanic an hour to figure out the value of any particular used car. So used cars aren't very liquid. Portability, storeability, divisibility, etc., could also be a problem with used cars.

Historically, shells have been one of the most common forms of money. And gold and silver. And IOUs. Nowadays most money is IOUs.

Suppose I wrote "IOU 1 apple, signed N. Rowe" on a bit of paper. And suppose everyone recognised my signature, and trusted my promise. My IOU is a valuable, marketable, asset. It might work very well as money. It would probably work better than apples as money, because it's more convenient than real apples. If apples were already used as money, my IOU, as a promise to pay apples, would be even more likely to be used as money.

Intrinsically worthless money.

Apples, pretty shells, and metals, could all be valuable even if they weren't used as money. But how can intrinsically worthless bits of paper be used as money?

The difference isn't as big as it first appears. If there is a demand for apples, shells, and metals, to be used as money, that increases the total demand for those goods, and makes them more valuable than they otherwise would be. Intrinsically worthless bits of paper are just a more extreme version of the same thing. But it's still a puzzle how it all got started in the first place. If the supply is limited, and there's a demand for the bits of paper to use as money, they will have a positive market value, and so can be used as money. But if they didn't have value in the market, they couldn't be used as money, so nobody would demand them, so they wouldn't have value in the market. How come we are in one equilibrium, and not the other?

Maybe the bits of paper started out as IOUs, promising to pay real goods that did have a market value. And people started using those IOUs as money. That's "convertible money", because the bits of paper can be converted into real goods by the person who signed the IOU. And then convertibility is temporarily suspended, but the IOUs are still valuable, because people expect convertibility to be restored soon. So they continue to be used as money and people continue to demand them. And then the expectation of restoration of convertibility fades away, but people keep on using them as money, and demanding them, because everyone else does.

Words don't really mean anything either. But people use "cat" to mean cat, because everybody else does too. Sometimes which equilibrium we are in depends on history.

The supply of money.

If we had a movie camera we would watch money circulate from one person to another as people bought and sold goods. If we only have a still camera we can take a snapshot which shows who owns the total stock of money at any point in time. What determines that total stock of money? What causes it to increase or decrease?

If we used apples as money I would now start talking about orchards. If we used shells as money I would now start talking about people collecting shells at the seaside. But we use IOUs as money, and most of those IOUs that are used as money are signed by things that call themselves "banks". So I'm going to talk about banks.

Some IOUs are used as money. Other IOUs are not used as money. A "bank" is something that has (mostly) monetary IOUs as liabilities, and (mostly) non-monetary liabilities as assets, on its balance sheet.

Canadian dollar banknotes are IOUs of the Bank of Canada. But the Bank of Canada isn't really a bank, because its IOUs aren't really IOUs. You can't go to the Bank of Canada and demand it convert your $20 note into gold, or US Dollars, or anything.

But the Bank of Canada does try to keep CPI inflation close to 2% per year. So you might say that Bank of Canada IOUs are indirectly convertible into CPI baskets of goods, at a sort of fixed price. Because if the Bank of Canada printed too many it would be forced to buy them back to prevent CPI inflation rising above 2% (the value of banknotes falling faster than 2% per year). But the Bank of Canada might change its mind about targeting 2% inflation.

Suppose the Bank of Canada wanted to increase the supply of Bank of Canada money. How would it do it?

There are only two ways: 1. print money and give it away; 2. print money and buy something with it. Either way you increase the stock of Bank of Canada money in circulation, because someone gets the money the Bank of Canada gave away or used to buy something.

The Bank of Canada never gives money away, though it could in principle. (But it can and does buy IOUs from the Canadian government, and it gives its profits to the government, and the government can and does give money away, and since the Bank of Canada is owned by the government, it amounts to the same thing.)

The Bank of Canada uses the second method: it increases the supply of money by buying things.

It makes no difference what the Bank of Canada buys. If it prints money and uses it to buy a bicycle the seller of the bicycle now holds more Bank of Canada money. It's the same if it buys cars and computers. But the Bank of Canada doesn't buy many bikes, cars, or computers. It does buy a lot of IOUs. Mostly government IOUs.

And if the Bank of Canada wants to reduce the supply of its money in circulation it sells something. If the Bank of Canada sells a bicycle, and gets a $20 banknote in return, the supply of its money in circulation falls by $20. But the Bank of Canada mostly sells IOUs, rather than bikes, cars, or computers.

Interest rates (a digression).

When I lend you money you give me an IOU in return. And when next year you repay the loan I give you back your IOU. We could redescribe the same thing as me buying your IOU, and then next year you buying it back.

When a bank lends me money to buy a car, it is buying my IOU. When I repay the loan it sells me my IOU back.

Suppose a bank wanted to increase the supply of its money. It needs to buy something to do this, and it normally buys more IOUs. If it lowered the rate of interest, that would persuade people to want to sell it more IOUs (borrow more), so it would be able to buy more IOUs, and increase the supply of its money. If it raised the rate of interest, that would persuade people to buy back more IOUs (pay back loans), so it would be able to sell more IOUs, and reduce the supply of money.

Banks' lowering interest rates is just a way for banks to buy more IOUs and so increase the supply of money. Banks' raising interest rates is just a way for banks to sell more IOUs and so reduce the supply of money.

That is true whther we are talking about a central bank like the Bank of Canada, or a commercial bank like the Bank of Montreal.

So when you read that "the Bank of Canada has reduced its target for the trend-setting overnight rate of interest", that's what's happening.

Central Banks vs commercial banks.

I use Bank of Canada IOUs as money. I also use Bank of Montreal IOUs as money. Bank of Canada monetary IOUs are printed on bits of paper. Bank of Montreal monetary IOUs are not printed on bits of paper. Electrons in a computer somewhere keep a record of how much the Bank of Montreal owes me in the balance on my chequing account. But it's just the same as if it were recorded on paper.

Sometimes it's more convenient to pay for something by taking a $20 Bank of Canada note out of my pocket and putting it in your pocket. Sometimes it's more convenient to write a cheque or use my debit card to instruct the Bank of Montreal to reduce its IOUs in my account by $20 and increase yours by $20.

The Bank of Montreal can increase the supply of its money in exactly the same way as the Bank of Canada. If the Bank of Montreal "prints" a $20 Bank of Montreal monetary IOU, and buys something with it, the supply of Bank of Montreal money increases. The only difference is that Bank of Canada IOUs are written in ink on paper, and Bank of Montreal IOUs are written in electrons on a computer.

Ordinary people and firms bank at commercial banks, like the Bank of Montreal, and TD Bank. We can't open an account at the Bank of Canada. But we do use Bank of Canada banknotes as money. Which is the same thing, just with paper instead of electrons.

The commercial banks do have accounts at the Bank of Canada. If I buy something from you, and pay by a $100 cheque, and I have an account at the Bank of Montreal, and you have an account at TD, this is what happens. Your account at TD goes up by $100, the TD's account at the Bank of Canada goes up by $100, my account goes down by $100, and Bank of Montreal's account at the Bank of Canada goes down by $100. Or BMO might borrow $100 from TD so that both their accounts at the Bank of Canada go back to where they were.

The Bank of Canada is the central bank. Central banks are the bankers' bank.

But how is it that the Bank of Canada is the one that controls Canadian monetary policy? If both the Bank of Canada and the Bank of Montreal issue IOUs that people use as money, shouldn't we say that both partially control the supply of money in Canada?

There's one big difference.

Bank of Canada monetary IOUs aren't really IOUs. They aren't a promise to pay anything. The Bank of Canada might choose to target 2% CPI inflation, in which case they are indirectly convertible into the CPI basket of goods. But the Bank of Canada might change its mind.

Bank of Montreal monetary IOUs really are IOUs. They are promises to pay Bank of Canada IOUs on demand, at a fixed exchange rate. It's the Bank of Montreal's job to ensure that one Bank of Montreal Dollar = one Bank of Canada Dollar. It's not symmetric. The Bank of Canada can make its dollar worth whatever it wants, and the Bank of Montreal has to follow along and make its dollar worth the same amount.

That's what gives the Bank of Canada the power to control the total supply of money in Canada.

Central banks and commercial banks together.

Suppose the Bank of Canada wants to increase the supply of money in Canada. It buys something, and pays for it with $100 in Bank of Canada monetary IOUs. The stock of money has increased by $100.

But it doesn't stop there. If that $100 lands in my pocket, and if I decide I would rather have $100 in my Bank of Montreal chequing account, I give it to the Bank of Montreal in exchange for an electronic Bank of Montreal IOU.

If the Bank of Montreal had to keep 100% reserves of Bank of Canada monetary IOUs against Bank of Montreal monetary IOUs, it would stop there. The total money supply has only increased by $100. (We don't count the $100 in the Bank of Montreal's reserves plus the $100 in my chequing account, because that would be double-counting.)

But it doesn't stop there. Because the Bank of Montreal isn't required to keep 100% reserves, and isn't required by law to keep any reserves nowadays, so probably doesn't want to keep that Bank of Canada IOU either. It will want to buy something.

If the Bank of Montreal buys an IOU (makes a loan) from one of its clients, in exchange for a $100 Bank of Montreal monetary IOU, that increases the total supply of money by another $100. If that client spent the loan to buy a bike from someone who banks at TD, that $100 Bank of Canada IOU gets transferred from BMO to TD.

But it doesn't stop there either. Because TD probably doesn't want to keep that $100 Bank of Canada IOU either. So TD buys something with a $100 TD monetary IOU, which increases the total supply of money by another $100.

And so on.

What stops this cumulative expansion of the money supply?

There are three things that could bring this cumulative expansion of the money supply to an end.

1. Commercial banks choose (whether or not they are required by law) to keep some fraction of Bank of Canada money in reserve against their own money.

2. People choose to keep some ratio of Bank of Canada money to commercial bank money.

3. The Bank of Canada chooses to stop it, because it thinks that inflation will rise above 2% if it doesn't stop it. It stops the cumulative process by selling something.

The demand for money.

At this point I would normally start talking about people holding inventories of money, and how the stock of money in inventory rises and falls whenever they sell or buy something. And I would define the "demand for money" as the average desired stock of money in inventory. And I would talk about the demand for money depending on: the price level; real income; and the spread between the rate of interest on money and the rate of interest on other assets.

Then I would put the demand and supply of money together.

But I'm going to stop here. Because this post is already too long.

And because I am beginning to think that "the demand for money" is a misleading concept that should be abolished.

Because once you start talking about "the demand for money", in the same way you talk about "the demand for apples", you end up with (orthodox) nonsense like the idea that the stock of money is determined by the quantity demanded at the rate of interest set by the central bank. Which makes sense for a good like apples, which is  traded in only one market, and which people either buy or sell. But doesn't make sense for a good like money, which is a medium of exchange traded in n-1 markets, that everyone both buys and sells. The quantity of money will be determined by the demand for loans at that rate of interest, but the demand for money is not the same as the demand for loans. But that's a topic for another post.

This is far too complex and too abstract.

How is a first year student supposed to understand all that?

Comments

[Ritwik]

Nick

I'd say, go ahead. You'll be surprised by what 1st year students can understand, if they've not already internalised other ways of understanding.

My parents were taught set theory basics in undergraduate advanced math. I was taught set theory basics in the 9th grade. Nowadays kids learn the basics in the 3rd grade and seem to do just fine.

[Mike Sproul]

Nick:

Here I am, 3000 miles away from Quebec, and the first thing I tell my students about money is the story of the French playing card money issued in Quebec in 1685. This card money comes close to being the first government-issued paper money in the Western world.
There you are, a stone's throw from Quebec, and you apparently don't talk about it. For shame!

Here is the story:

The soldiers in Quebec were paid in silver coins called livres, which arrived on supply ships from France. In 1685 the ship was late. In those days, in that part of the world, if a ship didn't make it into the harbor by late summer, the whole harbor would freeze over and it would be 8 months before they could try again, so the payroll was delayed by 8 months. This created obvious problems for the soldiers, who couldn't buy stuff, but it also created problems for local merchants, who couldn't sell stuff. They were all stuck in a recession, which was caused by a shortage of money.
The intendant, Jacques Demuelles, finally had the idea of paying soldiers with paper IOU's denominated in livres, each of which would be redeemable for 1 livre in actual coin when the ship arrived. The only problem was that they were in the Canadian frontier in 1685, and they had no paper. The best Demuelles could do was to confiscate every deck of playing cards he could find. Even then there was far less paper than he would have needed, so he cut each card in 4 pieces and on one piece he wrote "1 livre", on another "5 livres", etc, thus economizing on paper. He paid them to the soldiers, who paid them to the shopkeepers, who paid them to the farmers, etc. As nearly as I can tell from the historical record, their severe recession ended in a matter of hours.
Now consider the T-account of the colony. Suppose that there were 30,000 lvs in coins on that boat, intended to pay 30 lvs wages to each of 1000 soldiers. The colony's T-account would show 30,000 lvs of coins on a boat as the asset, while the liability side of the T-account would show 30,000 lvs of wages payable. If the intendant cautiously issued 10 paper lvs to each soldier, then wages payable would fall by 10,000 lvs, while the 10,000 paper livres would be added on the liability side. Since the initial issuance of paper livres was a success, he would have gone on to issue another 20,000 livres in paper, which would reduce wages payable by 20,000, while the 20,000 paper livres would get added to the liability side.
At this point I ask the students a question: "The colony has just tripled the supply of paper money, from 10,000 to 30,000 lvs. What will this do to the price of groceries?" A few of them who remember their Freshman macroeconomics will answer that the price of groceries will triple. But then they think it through and realize that the 30,000 lvs of paper are adequately backed by 30,000 lvs of coins on a boat, and they all agree that the price of groceries will be unaffected by the tripling of the money supply. This is their introduction to the real bills doctrine, and a few of them are appalled to think how easily they had been taken in by the quantity theory in their macroeconomics class.
Toward the end of the lecture, I ask what would happen if a soldier came to the intendant and offered him a 100 livre French bond if the intendant will pay him 100 newly-printed paper livres. (Canada's first open-market operation!) By this time the students clearly see that this new money will cause no inflation, since the 100 new livres are adequately backed by the 100 livre bond. Then I ask what would happen if a farmer asked to borrow 200 newly-printed lives, while offering his 300 livre farm as collateral. Once again, they see that the 200 new livres are adequately backed by the 200 livre IOU, and would cause no inflation. They are even able, most of them, to describe the situation in a T-account.

[Joseph]

I've found most early undergrad economics to be kind of worthless in one sense. The way intro econ is taught always seem to be an over simplification by necessity. On the micro side you atleast get to see the basic practices which economist use and the way they frame problems but the theory is always simplified out of necessity. On the macro side the theory is extra simplified to the point of being worthless or even detrimental. In my upper level Econ classes like public financeI would have class mates making arguments like "gov'ts should redistribute because poor people have higher MPCs so that would raise the average MPC which would increase consumption which would increase GDP according what we learned in first year macro" which is obviously wrong but makes a fair bit of sense in the framework taught earlier.

All of which is to say that I think you should go ahead and give a more complex more accurate description. It's better if the students hear the accurate thing even if they don't can't quite package it into a wholly self contained schema yet.

[Lord]

Just when it was getting interesting. Demand for money tends to be lack of demand for non money, especially when prices are falling or expected to fall, lack of demand for loans when this occurs or a decrease in opportunities for an increase in investment prices, and demand for less indebtedness when expecting reduced future opportunities and incomes.

[Varun]

"It makes no difference what the Bank of Canada buys. If it prints money and uses it to buy a bicycle the seller of the bicycle now holds more Bank of Canada money. It's the same if it buys cars and computers. "

Doesn't it matter since in the first case the price of a bicycle increases, and in the second case the prices of cars and computers increase instead?

[Gene Callahan]

"If it raised the rate of interest, that would persuade people to buy back more IOUs (pay back loans)"

If I have taken out a fixed-rate mortgage at 4% and the central bank raises rates to 5%, I am less likely to pay back my mortgage.

Of course, if it is floating rate your point holds.

[Nick Rowe]

Ritwik: Thanks! you are giving me confidence in myself, and in my students.

On re-reading it though, there are some little things i could do to make it clearer. Some IOUs are used as money, and others aren't used as money. But I am using the same name "IOU" for both. I need two different names, when I am talking about banks.

Joseph: I know what you mean. The Keynesian Cross model in particular, which used to be the centrepiece of intro macro, can be very misleading, if that's the only model in your mind. You end up with exactly that nonsense about MPCs that you talk about. Which isn't to say it doesn't have some insights as a model. The money/banking multiplier model also has some insights, about the difference between 100% reserve banking and anything less than 100%, and about the cumulative process of money creation. But M=(1/rrr)H doesn't quite do it.

Lord: Hang on! I have just finished writing a short follow-up post, on just that topic. I will probably post it tomorrow morning, after this one has had some time in the sun.

Varun: You are right. But unless the BoC buys a big amount of bikes, relative to the total market for bikes, it doesn't make a very big difference. And the BoC's balance sheet is small, relative to the total number of non-monetary IOUs. The BoC's total assets are normally less than 10% of annual GDP, IIRC, and total financial assets are bigger than annual GDP.

And it doesn't make any difference to the fact that the money supply increases, which is what I was concentrating on here.

[Nick Rowe]

Gene: True. I am less likely to pay off my fixed rate mortgage. But I am also less likely to borrow to buy a car. And I am more likely to lend to the bank by buying a term deposit. So the *net* flow demand for loans falls, which is what matters.

[Ron Ronson]

'But it's still a puzzle how it all got started in the first place."

Historically it may be a puzzle - but in theory it is seems plausible to explain the growth of IOU-based money as an extension of the process that allows a trusted individuals IOUs to be used as money in a local community. A bank that gathered accurate information on individuals credit worthiness could buy these IOus in exchange for standardized notes that would have the advantage to the IOU-holder of being in more standardized units than the IOU and of having a circulation beyond the local area. In parallel banks could directly create loans to individuals and businesses whose credit they trusted in exchange for their IOus. Once the money-issuing bank had gathered market share it might try and make its money into the unit of account in the economy by saying something like "We guarantee that the value of our notes will remain stable over time measured against a bundle of goods that we specify". The bank then buys and sells IOUs on the open market in order to keep this commitment.

[Nick Rowe]

Gene; Also, Canadian mortgages are usually much shorter term than US mortgages. It's normal for people to roll them over every couple of years, at the new rate of interest. So my example will sound more normal to Canadian eyes.

[Nick Rowe]

Ron: what you say there sounds plausible. And part of it also sounds historically OK. But part of it isn't historically quite right, AFAIK. We went from Gold Standard to No Standard. And only later did we switch from No Standard to the CPI standard (inflation targeting).

[Scott Sumner]

Nick, Great post. I would prefer not to give up on the "demand for money," but rather divorce monetary economics from the financial system. Just put 1/P on the vertical axis, and you'll be fine.

[Ron Ronson]

"We went from Gold Standard to No Standard"

Was the "no standard" really an implicit CPI standard ? People had to trust that the monetary authorities would keep the money valued within an acceptable range, or they would not hold it. Historical hyper-inflations seem to have occurred when peopled stopped believing this and dumped the currency. Even when CBs had inflationary policies as during the 19070's as long as the inflation was within a forecast-able and manageable range people still were prepared to hold it as along as it was better than the alternatives.

[Peter N]

Barter is so inconvenient, that primitive people don't use it. The barter to money transitions seems to be a myth. Money serves other purposes. If I'm remembering right

1) As a medium of exchange for ritually important matters like bride prices and honor debts.

2) As portable store of value for traders - commodity money

3) To facilitate the operation of temples. They end up acting like banks. Often the king will coopt the system.

http://books.google.com/books?id=YCU1AQAAMAAJ&pg=PN0-IA28&lpg=PN0-IA28&dq=temple+receipts+money&source=bl&ots=5aOJ8WwgMK&sig=MlLIh3m-6mAivfNkKaPb3A_t8n0&hl=en&sa=X&ei=xygpUeC-GJLo9gSry4CQBw&ved=0CFQQ6AEwBg

4) To memorialize debt - IOUs and later bills of exchange. These can be endorsed and traded as money.

This is a more interesting story and traces of these ancient methods survive in our current system (except use 1), though the resemblances were stronger 100 years ago.

[K]

Nick,

I like it quite a lot. It's pretty mind blowing stuff for a thinking first year student and it's well presented.

Personally I would have framed it more around something like this...

The stuff most of you think of as money is only 3% (or whatever) of total bank accounts. Forget that stuff. If it disappeared, nothing would change. When you buy something you don't need any "money". All you need is for the seller to make a loan to the buyer in the amount of the price. To avoid everyone having loans to everyone else, it's good to have a central counterparty for all loans. If I buy something from you, you extend a loan to the central counterparty and the central counterparty extends a loan to me. That central counterparty is called the banking system.

Notice that each loan doesn't involve any movement of "money." The loan is created against a movement of a physical good or service. Once we have a central counterparty each new "loan" doesn't need to be created as a new legal instrument either. Is is simply a credit or debit from an existing loan, i.e. a deposit account or line of credit.

The total amount of outstanding loans is determined by the relative desires of different people (young/old rich/poor) to save or consume. It has no consequence for the price level. If we all had the same propensity to save and we all were at the same stage of our lives there would be essentially no outstanding loans.

The central bank controls the interest rate on risk-free loans.

The end.

[Frances Woolley]

Nick, Mike Sproul is caught in spam.

[kim mcphail]

Nick:

One might also ask how the roles/functions of money and importance of each of those – and for different types of money – are (have) changing (changed) in the 21st century. Also, whether there are new functions/roles for money that we didn't typically think about much in prior centuries, but which may have becoming more significant and relevant in the 21st century.

For example, one the features of some money (bank notes, IOUs of central banks) that received some prominence in the 20th century was that of anonymity. The anonymity of bank notes was often contrasted with private-sector electronic money (bank accounts and other forms of electronic money). Bank notes were said to be anomymous – because they didn't leave a paper trail – although, ironically, bank notes were made of paper. On the other hand, bank accounts and electronic money, said to leave a paper trail, are bits and bites in computer systems, and were said to be non-anonymous.

One reason routinely given in the 20th century for the elimination of some high-denomination bank notes in several countries was that some said they were often used for illegal or tax-evasion purposes, while private-sector bank money was typically believed to be less useful for such nefarious purposes. Is that true?

And what about the 21st century? I think a more critical role/function of money in the 21st century for consumers and merchants perhaps than anomymity was said by some to be in the 20th century – and one never discussed in anything I've ever read – may be accessibility.

Technology has changed to the point that individuals or merchants can easily be cut off from access to their bank deposits or other electronic money and, thus, to the essential transactions services of their bank money – selectively and with no notice or explanation, and little recourse other than public pressure.

We've already seen situations in Canada – just last year -- that made news headlines in Ottawa newspapers when some Canadians were denied access not only to bank services, but also to to their bank money, and to the essential transactions services of their money – and for reasons that were reported in news reports to be very dubious.

One of the differences between central bank IOUs in the form of bank notes and electronic money is that it's not easy to “cherry-pick” (or “lemon-pick”) or be “selective” in who will be barred access to their bank notes when a consumer and merchant do a legitimate transaction in a store, for example. But it's increasingly simple to “lemon-pick” or “cherry-pick” or be “selective” in who will be barred access to their bank deposits or other electronic money and the essential transactions services associated with those (either a consumer or a merchant) and be barred from the same legitimate consumer/merchant transactions.

And then, of course, there are things like power outages, wars, and stuff like that to worry about, where people may find their access to electronic money and its critical medium of exchange function – often necessary for survival of firms and individuals, whether there is any war declared or not – suddenly and inexplicably interrupted.

With bank notes, it's “all or nothing” in the sense that if any government tries to repudiate in any way, say a $5 bill, it must repudiate all $5 bills. But in some states of the world, one can imagine that people with a particular ethnicity, or religion, or country of origin, or colour of skin, or political views, or suspected to be/have any of those or other things, or whatever, are barred from accessing even $5 of their electronic money at a bank or elsewhere. It's technologically a lot easier to bar access to a means of payment/exchange based on electronic money than for bank notes.

It's also easier to do that in a quiet, surreptitious fashion with electronic money – so that it takes time for people to figure out that it's even happening – and especially people not directly affected. That's quite different than the situation with bank notes.

This is something that, in Canada anyways, a lot of people never thought about in the 20th century. But there is already evidence in Canada that it's something that people should be thinking about in the 21st century – with regards to their money.

There are other examples that aren't about money but which, nevertheless, raise reasons to be concerned about individuals' and merchants' access to electronic money and payments services these days. I recall reading about a large and reputable electronic bookseller, who just zapped up – removed – without any notice or explanation – all copies of particular books that clients had legitimately purchased, paid for, and legally downloaded to their computers/handheld computer device/bookreader.

Does anybody worry about that kind of thing for electronic money – whether held in the form of banks accounts at reputable banks, or any other forms of electronic money on computers or handheld devices?

You can't do that with $5 bills, or $20 bills, or $50s or $100s. But it's easy to do, technologically, with electronic money.

Nor do introductory courses on economics/money – or even advanced graduate economics courses on money – typically discuss payment systems, which is where almost all the money in the planet churns around every day at the speed of light – if it moves at all. The amounts involved are truly quite astonishing to people who have never thought about that. I think they should – and even in a first-year introductory economics course that talks about money. What is a payments system? Why are they used? How do they work? What is their role/function. How do they work in Canada?

This is linked to what I mentioned about anonymity (in the 20th century) versus accessibility (in the 21st century) in some ways.

What's the difference between “finality” of a transaction settled with bank notes and one settled in an electronic payment system? What's the chain of events required to settle a transaction in a payment system between a consumer and merchant? And settled for who? The financial intermediary which accesses a (wholesale) payments system directly? What about the individual or merchant (who accesses the electronic payment system through a financial intermediary)? What are the benefits/costs to the different parties to such an intermediated transaction? What are the risks to each? What are the types of risk? How is the importance of each of those risks changing in the 21st century? How might those costs/benefits/risks differ in war zones? Which risks are becoming more important in the 21st century compared to the 20th century? From the perspective of banks in the payments system? What about from the perspective of consumers/individuals/merchants? Sometimes it can take the legal systems a decade or so to figure some of that stuff out.

I realize these issues don't fit into the typical introductory economics course's discussions about money (or even graduate economics courses on monetary theory) where one often begins by talking about why centuries ago societies centuries ago used seashells or other objects as “money.”

I think these issues would be good for economics courses that discuss money in the 21st century because they're very important to peoples' lives – both consumers and merchants.

Osprey

[Bill Kelly]

Nick, possibly somewhat off topic, here is a link (courtesy Mark Thoma) about Roman finance that may interest you.

http://libertystreeteconomics.newyorkfed.org/2013/02/historical-echoes-cash-or-credit-payments-and-finance-in-ancient-rome.html#.USlNj2Iw7K4.mailto

[primedprimate]

Joseph and Nick, why is the MPC story 'obviously' wrong?

[Alejandro Villagomez]

Great!!!
}
I´ve been teaching Money for many years.... I will use your ideas for my next course!!!!

[Nick Rowe]

Mike Sproul: Sorry you got stuck in spam. Thanks to Frances spotting you there I fished you out.

I actually live in Quebec, though I work in Ontario!

The Quantity Theory says a *permanent* doubling of the money supply will, *in the long run*, double the price level. Everyone knew that the playing card money would be withdrawn soon. And prices are sticky in the short run.

But this is what I don't understand about your theoretical approach. You acknowledge that the shortage of money caused a recession. Presumably it caused a shortage of demand for goods, so there was excess supply of goods? Did that recession, and excess supply of goods, put downward pressure on prices? And did the increase in the supply of playing card money not end that recession and end the excess supply of goods and end that downward pressure on prices? And if prices had been perfectly flexible, wouldn't prices have fallen in the recession, and risen back up again when the playing cards were introduced?

I'm trying to make sense of your model's AD curve, in {P,Y} space. Your model seems to have a real bills theory of P, but a monetarist theory of Y. A monetarist would say that the increase in playing card money supply, even though it was known to be temporary, shifted the AD curve to the right to some extent. But because prices were sticky, the main effect was an increase in Y, with little effect on P (though P would eventually have fallen if the playing card money hadn't been introduced).

[Nick Rowe]

Scott: Thanks! See what you make of my new post, on the demand for money. I agree that there is no *necessary* theoretical connection between money and finance. But we live in a world where money and IOUs are linked in practice.

Ron: "Was the "no standard" really an implicit CPI standard ?"

Hmm. OK. Yes, it sort of was. Governments were still concerned about inflation, even though there was no explicit numerical target.

Peter N.: "Barter is so inconvenient, that primitive people don't use it. The barter to money transitions seems to be a myth."

Barter is indeed very inconvenient. And that is indeed why we so rarely observe it. Jonathan Graeber et al have got it totally the wrong way round. If he could show lots of times and places where people *did* continue to use barter, *then* I would begin to wonder what was wrong with the Mengerian story about why people don't use barter. If he can't find such times and places, that *confirms* the Mengerian story about barter being unstable. Just like we don't observe eggs standing on end.

K: Thanks! Suppose everybody buys and sells everything on credit. The buyer gives the seller a personal IOU. There needs to be a central clearing house where circles of offsetting IOUs are cancelled out. Credits and debits at that central clearing house are then "money". The IOUs of the BMO and TD are only money because there is such a central clearing house where offsetting cheques from BMO clients to TD clients and vice versa are cancelled, so only the net balance is owed by BMO to TD.

Hmmm. Maybe I should have said something about that.

Frances: thanks for letting me know. I fished him out.

Kim: I really worry about that sort of thing. Remember that bit in Margaret Atwood's Handmaid's Tale, where they cancel all the women's credit cards (or something)?

I said there is no difference between IOUs written in ink on paper and IOUs written in electrons on computers. But there is a difference, if one can be transferred from buyer to seller without anyone else knowing, while the second is a matter of semi-public record.

But I don't think I would teach it in Intro.

Bill: thanks. That's close enough to on-topic.

primed: Because:1. a shortage of AD is usually not the problem, Y is usually limited by AS, not AD. 2. Even if AD is a problem, there's monetary policy. 3. Even if (for God knows what reason) monetary policy can't be used, there's a lot more to fiscal policy than making transfers from low MPC to high MPC people. 4. It's not MPC that matters. It's Marginal Propensity to not hold money. For example, if the people who had a low MPC had a high Marginal Propensity to Invest, it wouldn't work.

Alejandro: Thanks!

[Nick Rowe]

Damn TypePad put *me* in spam!!! Doesn't it know who I am??

[Mike Sproul]

Nick:

Thanks for rescuing me from the spam folder. I felt better after I saw that it happened to you too.

The playing card money was in more or less continuous use until 1751, when it was forcibly suppressed. Every time they had a crisis they would issue it again, and every time, the French King would try to suppress it. They had extended periods of stable prices, and periods of inflation. The data are too noisy to give much support to your explanation (that they expected the money to be withdrawn) or to mine (that the money was adequately backed).

When I say that a money shortage causes a recession, I am thinking that the lack of money just makes it more difficult to trade. But an apple still trades for 2 oranges, 2 oranges still trade for 3 bananas, which trade for 1 oz of silver, which is still the unit of money. I view a money shortage as having an effect similar to the sudden destruction of our roads or telephones. The relative value of silver (i.e., money) would be unaffected, but trade would still slow to a crawl.

There are no AD curves in my world. I'm fine with drawing a demand curve with apples on one axis and the price of apples on the other, but it makes no sense to put "all goods" on one axis, and "price of all goods" on the other.

[Nick Rowe]

Mike: If there were a (stock) demand for silver to use as jewelry, the equilibrium prices of oranges and bananas in terms of silver would presumably depend on the demand and supply of silver to use as jewelry. And the supply of close substitutes for silver jewelry, made (say) of pretty playing cards, would also affect the equilibrium prices of oranges and bananas in terms of silver.

Similarly, if there were a (stock) demand for silver to use as money, the equilibrium prices of oranges and bananas in terms of silver would presumably depend on the demand and supply of silver to use as money. And the supply of close substitutes for silver money, made (say) of pretty playing cards, would also affect the equilibrium prices of oranges and bananas in terms of silver.

[primedprimate]

Thanks - I thought I was missing something more fundamental, but the reasons you list are indeed familiar to me, and I guess, by extension, obvious.

Great post, by the way!

[Mike Sproul]

Nick: Money shortages are strange and confusing things, which is why the card money makes such a good teaching example. Let's say that the tardy boat reduced their supply of silver coins by half. What is the result? First, and most obviously, trade becomes more difficult, especially for soldiers who have nothing to barter. Second, silver coins will start to sell at a premium. How much of a premium? At one extreme, people might claim that the halving of the quantity of silver coins would double their value. This would mean that the silver price of groceries falls by half, so real output effects are minimal. That's clearly not what happened. At the other extreme, people might claim that the silver coins will not gain any premium at all, that the price of groceries will not change, and that there will be a shortage of coins, complete with bad effects on real output. That might or might not be what really happened, but we'll probably never find data to answer the question.

My best guess is that the halving of the supply of silver coins caused the coins to trade for a 10% premium. (I remember a couple of times when my local grocery store offered people $1.05 for 100 pennies. That's the kind of shortage and premium I'm thinking of.) So you still get a big recession along with a little deflation. Of course, any paper IOU's that are a (adequately-backed) claim to 1 silver coin would still be worth 1 silver coin, but like the silver coin, they would buy 10% more groceries. That is, unless plenty of those IOU's were issued, in which case the coin shortage would be relieved, the recession would end, and prices would go back to normal.

Shortages of coins, by the way, are much easier to understand when you look at the money supply of the English colonies, starting with Massachusetts in 1690. Those colonies rated their coins at a price of (say) 5 shillings=1 oz of silver. As the coins wore out, the market rate wanted to go to 6 shillings=1 oz., but the legal rate would be left at 5s/oz, and people would write about "the last wagon load of silver leaving the colony". Those money shortages were also alleviated by issuing paper IOU's, and their recessions ended just as suddenly as in Quebec.

[Joseph]

Primed pirate: I was a little pithy when I said obvious but like nick said if you work through a slightly more complicated model then it doesn't work out.

Nick: I agree that those models have useful insights. I think at the undergrad level students are often given the obligatory models are not the world but are useful lecture the first day of intro and then it's never mentioned again and it's easy for students to stop making that distinction 3 lectures into the class. Most textbooks will note the limits and assumptions in the models studied but it's like a side note and I've never heard of an exam question bassed off of identifying the limits of the models. ( not that this is limited to Econ. Same problem pops up in physics courses too.) more emphasis on the limits of the models would probably be useful to most students to give them a much fuller understanding. I suspect this may be linked to the thought in a lot of lay people's minds that they can "disprove" modern economics by taking a simple model and reducing it ad absurdum.

[JP Koning]

Good post. Definitely a lot of meat here.

One addition that might help students is the dynamic between the unit of account and how that unit is defined: the medium-of-account. I've found that the concept of MOA has helped me abstract from all the various standards we've been through as a society, from gold to bimetallic to currency pegs and gold pegs, and finally CPI. Describing these as mere changes in the MOA might save the teacher from from getting bogged down in the institutional details of each regime.

[VN]

Nick,

Thank you for the post. I’m a huge fan. I have an off-topic pop-sociological rant.

When I took Macro 101 twenty years ago, the story of money I was told was very similar to yours, down to barter and the money multiplier. Maybe that’s what you have to teach because that’s what Macro 101 students are supposed to learn. I went on to take Macro 102, was bored with it, and decided to study something else.

I now find myself wondering if teaching our brightest minds the money multiplier is doing us a disservice. As you alluded to in your response to Joseph, we grown-ups don’t actually believe in the money multiplier, especially not now. I’m sure David Colander, who taught me Macro 102, made it abundantly clear that the money multiplier was just an implausible model. But that’s not what I remember – I remember that I got an A and thought I kinda understood macroeconomics. Until the fall of 2008.

I worry about the hordes of Macro 101 students who learn about the money multiplier and go on to govern us and shriek about the imminent dangers of the Fed’s balance sheet. Have they knowingly been taught something false, because that’s what they were supposed to learn?

I also worry about those few Macro 101 students who learn about the money multiplier and think, “what a boring and settled field this is – I better go study something else.” Could you talk about Graeber instead of barter? I don’t know, MMTers? That’s the opportunity cost of the money multiplier. I have since learned that David Colander is a well-regarded economic historian; there is so much I could have learned from him. But I learned the money multiplier.

Of course this rant is not about the money multiplier, but about Macro 101 in general. Please, you CAN do better!

[Frank Restly]

Nick,

"Because once you start talking about "the demand for money", in the same way you talk about the demand for apples, you end up with (orthodox) nonsense like the idea that the stock of money is determined by the quantity demanded at the rate of interest set by the central bank."

Simple model of credit money

D * V = P * Q = Income * ( 1 - Liquidity Preference ) + New debt origination

D * V = ( D * V + INT * D ) * (1 - LP) + dD/dt
D = exp ( f(t) )
dD/dt = f'(t) * D

V = ( V + INT ) * ( 1 - LP ) + f'(t)
V * LP = INT * ( 1 - LP ) + f'(t)
V = [ INT + f'(t) ] / LP + INT

Solving for liquidity preference (LP) we rewrite:

LP = ( INT + f'(t) ) / ( V - INT )

The demand for money is really just liquidity preference. It is affected by the change in demand for loans ( f'(t) ), the interest rate (INT), and the velocity of money through an economy.

And so in explaining the "demand for money", it would be better to call it the demand for the most liquid asset (currency or currency substitutes).

[Frank Restly]

Got a sign wrong;

V * LP = INT * ( 1 - LP ) + f'(t)
V = [ INT + f'(t) ] / LP - INT

Solving for liquidity preference (LP) we rewrite:

LP = ( INT + f'(t) ) / ( INT + V )

[James Oswald]

"Intrinsically worthless money"
You don't need a historical conversion story where people forget that you can't convert paper anymore. The opportunity cost of any action is what you have to give up to do it. The opportunity cost of using apples for currency is eating those apples, even temporarily. The opportunity cost for using an intrinsically worthless item for money is nothing. It's clear to see which one of those two options is superior. The more useless an item is, the better suited it is to being used as money. Convertability is a bug, not a feature. It forces someone to maintain a store of the conversion material and refrain from doing useful things with it. Gold in Fort Knox can't be used for jewelry.

[Too Much Fed]

So medium of account (MOA) = medium of exchange (MOE) = currency plus demand deposits, right?

[Frank Restly]

Too Much Fed,

Currency denominations (dollars, pesos, euro's, etc.) are units of account. Paper and coin currency are media of exchange.

The use of paper and coin currency as a media of exchange allow the relative prices of two goods to be established absent a direct bartering system. The currency itself does not maintain a record (account) of prices that are paid. For that, either a receipt (for goods already purchased) or a bill (for goods already received) is required.

[Too Much Fed]

Frank Restly, I agree with Nick here:

http://worthwhile.typepad.com/worthwhile_canadian_initi/2012/10/medium-of-account-vs-medium-of-exchange.html

"[Update: just to clarify terminology: in my model, gold is the medium of account; and (say) an ounce of gold is the unit of account.]"

Replace gold with dollars. Replace dollars with paper currency, coin currency, and demand deposits.

As long as there is 1 to 1 convertibility and entities accept demand deposits just like they accept currency, I see MOA = MOE = currency plus demand deposits.

[Too Much Fed]

"Suppose a bank wanted to increase the supply of its money. It needs to buy something to do this, and it normally buys more IOUs. If it lowered the rate of interest, that would persuade people to want to sell it more IOUs (borrow more), so it would be able to buy more IOUs, and increase the supply of its money."

So lower interest rates are about more debt then?

[Frank Restly]

Too much Fed,

"If it lowered the rate of interest, that would persuade people to want to sell it more IOUs (borrow more), so it would be able to buy more IOUs, and increase the supply of its money. So lower interest rates are about more debt then?"

First, the Fed can set interest rates, but it cannot force people to borrow at those interest rates nor can it force banks to lend at those interest rates. Second, the Fed sets nominal interest rates and so even low nominal interest rates can be punative in a deflationary environment.

[Too Much Fed]

"it cannot force people to borrow at those interest rates nor can it force banks to lend at those interest rates."

OK, but do lower interest rates persuade/encourage "people" to borrow under certain assumptions? If the fed can't get "people" to borrow, will it get the gov't to borrow for them?

[Too Much Fed]

There are only two ways: 1. print money and give it away; 2. print money and buy something with it. Either way you increase the stock of Bank of Canada money in circulation, because someone gets the money the Bank of Canada gave away or used to buy something.

The Bank of Canada never gives money away, though it could in principle. (But it can and does buy IOUs from the Canadian government, and it gives its profits to the government, and the government can and does give money away, and since the Bank of Canada is owned by the government, it amounts to the same thing.)

The Bank of Canada uses the second method: it increases the supply of money by buying things.

It makes no difference what the Bank of Canada buys. If it prints money and uses it to buy a bicycle the seller of the bicycle now holds more Bank of Canada money. It's the same if it buys cars and computers. But the Bank of Canada doesn't buy many bikes, cars, or computers. It does buy a lot of IOUs. Mostly government IOUs."

I believe the three(3) methods have differences.

More MOA and MOE given away is different than more MOA and MOE used to purchase bonds/loans is different than more MOA and MOE used to purchase bikes.

[Too Much Fed]

"What is money?"

"People use money as:

A medium of exchange. They sell everything else for money, and then use that money to buy everything else.

A unit of account. They talk about prices in terms of money.

Nearly always, those two defining functions go together. It's easy to understand why. It's more convenient. If the supermarket wants you to buy food with Canadian Dollars, they will put Canadian Dollars on the price tags."

Could you define "money" with these terms or combination(s) of them?

1) currency

2) central bank reserves (the demand deposits of the central bank)

3) demand deposits

[Saturos]

I can't imagine how you would make this simpler whilst still telling people everything they need to know. Don't underestimate your pedagogic abilities, Nick! This is way better than existing textbooks. Maybe restructuring?

I've always been disturbed by the standard teaching of the money multiplier, which ignores the distinction between central bank money and commercial bank "money", and never explains why the various banks accept each other's fiat liabilities as money ("just treat them all the same"). The value of redeemability is now non-obvious, since households don't "have" to hold cash - indeed we can imagine no one holding cash. Base money is real "money".

"If that client spent the loan to buy a bike from someone who banks at TD, that $100 Bank of Canada IOU gets transferred from BMO to TD."

Many people don't actually understand this.

Ritwik, is that in India? Just got a new anecdote...