Taxes and the value of paper money

Does the value of an intrinsically worthless (paper) money depend on the government's power to tax? I am going to answer this question from a quantity-theory perspective. The short answer is "yes". But the full answer is different from some other theories that also answer "yes".

Those of you who already understand the modern quantity theory of money will find nothing new in Part 1 of this post. I'm just trying to explain the standard story as simply as possible, without any math or jargon (I have even carefully avoided using the words "real" and "nominal"). You may find Part 2 more interesting.

Part 1.

The quantity theory of money says that the value of money depends on the quantity of money. If you double the stock of money, other things equal, the value of money will halve (i.e. the price of goods in terms of money will double).

Most quantity theorists recognise that prices of most goods don't adjust to equilibrium instantly. So it may take time for a doubling of the quantity of money to double prices.

(Strictly, the quantity theory only works for something like a paper money that is useless except as money. Because the monetary units don't really matter. If you add a zero to every $1 bill, so it becomes a $10 bill, and add a zero to all the prices, so $1 becomes $10, etc., then nothing has really changed in the economy. If you were in equilibrium before the change, you are still in equilibrium after the change. So you increased the stock of money tenfold, and each unit of money is worth one tenth its original value, and the equilibrium price level has increased tenfold. If money were gold, and dentists could use it, things wouldn't be exactly the same if you doubled the stock of gold and doubled all prices. Because dentists would use more gold to fix people's teeth, and the cost of getting your teeth fixed wouldn't exactly double.)

But the quantity theory also says a lot more than that. The current stock of money (the money supply) is not the only thing that affects the value of money. The value of money also depends on the demand for that stock of money (money demand). The demand for money is like the demand for an inventory. Money flows into and out of our pockets, so the amount we hold fluctuates hour by hour, but we demand to hold a certain average stock of money, because it makes shopping easier.

A doubling of the supply of money, if demand is initially unchanged, means that people are holding more money than they want to. They will try to spend or lend the excess. But if the total supply of money stays the same, even though each individual can spend or lend his money and pass it onto someone else, it isn't possible for all people to do get rid of their money this way. But their attempts to get rid of it by spending or lending are what causes the demand for goods to rise, and the prices of goods to rise. And when the price of everything doubles and people's money incomes and expenditures double, they will want to hold exactly double the amount of money for shopping, So prices adjust to make the demand for money double to match the doubled supply.

And that demand for money will depend on many things, including the cost of holding money. Because demand curves slope down, and the higher the cost of holding money, the less money you want to hold. You might spend the same amount of money per year, and earn the same money income per year, but you want your inventory of money to turn over more quickly, so you want hold a smaller stock of money on average. And the cost of holding money depends on how quickly that money is losing its value. (It depends on the expected rate of inflation, in other words).

And, just as the level of prices depends on the level of the stock of money, so the rate of change of prices over time will depend on the rate of change of the stock of money over time. The faster the stock of money is growing over time, other things equal, the higher will be the rate of inflation. So, the value of money today depends not just on the quantity of money today, but also on how quickly that quantity of money is growing over time.

Take two otherwise identical economies, with the exact same stock of money today, but the first economy has a growing stock of money and the second economy has a constant stock of money. The first economy will have a higher price level today, even though it has the same quantity of money today.

When governments spend money they must either get that money from taxes, or else borrow it, or else print it and increase the stock of money in circulation. Taxes reduce the stock of money in our pockets. For a given amount of government spending, and a given amount of government borrowing, the higher are taxes the smaller will be the amount of money printed and the slower will be the growth rate of the stock of money. And the slower the stock of money is growing over time, the higher will be the value of money today.

Take two otherwise identical economies, with the same quantity of money today, and with governments spending the same percentage of national income and neither country borrowing. But the first economy has taxes as a lower percentage of national income than the second. The first economy will have a higher price level than the second economy, even though they are otherwise identical, and have the same quantity of money today. Holding government spending and borrowing constant, higher taxes increase the value of money, because they reduce the growth rate of the money supply, reduce the rate of inflation, reduce the cost of holding money, increase the demand to hold money, and so increase the value of money.

There's a second way that taxes may affect the value of money, that is associated more with a Keynesian approach, but can nevertheless be incorporated into the quantity theory of money.

For a given level of government spending, and a given growth rate in the money supply, a reduction in taxes means the government has to borrow more. It has to issue more bonds. That increase in government borrowing may increase the equilibrium rate of interest on government bonds. If money pays no interest, but bonds do, then the foregone interest on bonds is the opportunity cost of holding money rather than bonds. So a reduction in taxes will cause an increase in the opportunity cost of holding money, reduce the demand for money, and so reduce the value of money.

Take two identical economies, with the same stock of money, and the same growth rate of money, and the same government spending as a percentage of income. If the first has lower taxes as a percentage of income than the second, the first will have a higher price level.

Part 2.

But this still leaves a puzzle. Paper money is only useful for shopping if it has value. If it has value, people will find it useful for shopping, and so there will be a demand to hold an inventory of money for shopping, and that demand (and supply) will determine its value. But if it didn't have any value at all, it wouldn't be useful for shopping, so there will be no demand for it, and so it won't have value.

There are two equilibria. In one equilibrium people leave the worthless bits of paper lying on the ground. In the second equilibrium people use those same bits of paper for shopping and they are valuable. Why are we in the second equilibrium, rather than the first?

Or, if there are red bits of paper and green bits of paper, what determines whether people will use the red, or the green, as money?

One possible answer to this puzzle is that the government requires people to pay the government 10 bits of green paper per year, as taxes. Does this answer work? Maybe, it depends.

If the government is spending 10 bits of green paper per person per year to buy goods, and collecting 10 bits of green paper per person per year in taxes, that means the government is using that green paper to do its shopping. But that does not eliminate the equilibrium in which green bits of paper are worthless. The government spends 10 bits of paper at the beginning of each year, gets nothing in return because the paper is worthless, and then at the end of the year people simply give the green bits of paper back to the government.

Hell, even I could do that. "Here are 10 bits of green paper per person. Unless I get all 10 back at the end of the year I will beat you up. How much will you give me for them?". "Nothing" is one possible answer, even if my threat is credible.

Let's try Plan B. "Here are 9 bits of green paper per person. Unless I get 10 back at the end of the year I will beat you up. How much will you give me for them?"

Plan B works. I have created an excess demand for the bits of green paper. 10% of the population will get beaten up, whatever happens. But each individual can avoid a beating by outbidding others for those bits of paper. The value of each bit of paper will be one tenth of the value of avoiding a beating for the person at the tenth percentile of the distribution of willingness to pay to avoid a beating.

And once the green bits of paper have value, people might also use them to do their own shopping, if they are more convenient to use than other media of exchange.

But Plan B only works because the government is trying to reduce the stock of money in circulation, by trying to collect more in taxes than it spends. Governments don't usually do this. When governments issue money, they normally do so because they want to spend more than they collect in taxes.

And Plan B only works because the intrinsically worthless paper money is now no longer intrinsically worthless. It is a convertible currency, at the margin. 10 bits are convertible into one avoided beating, and an avoided beating is a valuable good.

Plan B is therefore just a specific example where the puzzle is solved by supposing than an intrinsically worthless paper currency was initially, when first introduced, convertible into something intrinsically valuable. And it doesn't have to be the government that issues the money, or makes it convertible. And though people are likely to choose to use the same money that the government uses, it's also true that governments are likely to choose to use the same money their people use. The government is a big shopper, and one that has more powers than most. But the money people use isn't always the one issued by their own government, and in which they pay taxes.

Suppose that paper money is initially convertible into gold, or some other good that has at least some positive value independent of its use as a medium of exchange. Then people get used to using paper money as money. Even if convertibility is eventually suspended, and so the equilibrium where paper money is worthless becomes one possibility, people will continue in the second equilibrium, where the paper is valuable, simply because that's the equilibrium they are already in. It's a network effect, like speaking a particular language, or using a particular word processing program. I use Canadian dollars for shopping because everybody I shop with uses them too. And because we use them, they are valuable. And because they are valuable, we use them.

Plan B made me think of this:

 

Comments

[Kevin]

Economics = Attempt to mathematically model human psychology.

[Nick Rowe]

Kevin: and this post = an attempt to mathematically model without math ;-)

[Shangwen]

What if paper money also has value in terms of traceability of transactions (a negative transaction cost)? Right now in my neighborhood, people are having decks built and landscaping put in by fallow school teachers and off-duty firemen, and they are being paid under the table in cash. They are buying their building supplies in cash, and I doubt they have a business or GST number. Then, faced with the option of getting landscaping from a legal business that would charge for the work at a factor of 1+T, my neighbor is buying the good with a value X(1+T) for only X or even less. Even if my neighbor pays for it with cash taken from his HELOC, T>k so the interest charge is still worth it. And of course there is the personal satisfaction of helping others avoid taxes (not that I would ever do such a thing). I know this is not the core of the economy and that this is impossible for many transactions, but it contributes to the demand for paper.

[Scott Sumner]

Nick, I missed something. Are you saying that money has value because of the power to tax, or that the power to tax is one factor than could give money value? It seems to me you've shown the latter, but not the former.

Suppose a society of people stranded on a Pacific island agree to use Monopoly money for transactions. There is no government. Would that money have any value? This is not a rhetorical question, as I actually don't know the answer.

My instincts say "yes," but I am not certain.

[Nick Rowe]

Shangwen: the fact that currency may make it easier to evade taxes will increase the demand for currency relative to money in the form of chequing accounts. Chequing accounts may (or may not, it depends) pay interest, and are sometimes more convenient than currency (sometimes less), and are harder for thieves to steal. So there are different balancing motives for preferring currency vs preferring chequing accounts, and most people hold and use both.

But, from the perspective of this post, a chequing account is just a promise to pay currency. The $100 in your chequing account is valuable because it is convertible into currency that's valuable. But why is currency valuable, and what determines that value? It's easier to start out imagining an economy where the only money is paper currency. Which is what I'm doing. We can bring chequing accounts in later.

Scott:

I'm saying that taxes affect the value of money. That's Part 1.

In part 2 I explore the idea that paper money has value because of the power to tax. But I reject it. Instead, I propose that paper money has value initially because it is initially convertible. The government *might* be the one making it convertible, say by letting people pay 10 bits of paper in lieu of one hours labour (or to avoid one beating). But it could also be anyone else making it convertible.

"Suppose a society of people stranded on a Pacific island agree to use Monopoly money for transactions. There is no government. Would that money have any value? This is not a rhetorical question, as I actually don't know the answer."

I was thinking through that when you wrote that post a couple of days back. That's partly what lead me to write this one. It's a question I've thought a fair bit about on and off over the years.

Yes, if they all get together in a meeting and say "Lets all use this stuff as money" it would work. Contractarians would say that meeting *is* the government. But I would say that what that meeting does is pick out a focal point, out of many possible equilibria, so they can coordinate on that particular equilibrium. It doesn't need enforcement. It's self-enforcing. It's like tossing a coin in public to decide whether to drive on the left or right. It's a sunspot equilibrium.

But, historically, I don't see that happening often. The most powerful focal point is what we did in the past. Custom. I see the current focal point as having evolved out of a past in which paper money was initially convertible, then the promise of convertibility slowly faded away but paper money remained. Menger, then Von Mises Regression Theory of Money, is basically my approach.

[Lee Kelly]

Selgin has done some work on banknotes in the pre-Civil War United States. The government had to actively handicap and then suppress private banknotes to get the public to use the government's own currency.

[Nick Rowe]

Lee: in that case though, the both private and government banknotes were presumably convertible into gold? So the government was attempting to suppress demand for private banknotes to increase demand for its own. The examples that really contradict tax theories are where an irredeemable paper money is not used for taxes. The old Iragi Dinar is said by some to be an example. The use of US dollars in Cuba, even when illegal, is perhaps close, though Cubans could (illegally) use them in the tourist stores, or use them if they emigrated. Bitcoins, if they survive, will be a third.

[Nick Rowe]

Lee: on second thoughts, I see your point now. Absent prohibitions, people wouldn't have used government banknotes as money even though they were still used for taxes.

[Kevin]

I'll give you +1 for a cheeky retort Nick.

But in essence that's all economics is.

I certainly don't mean for it to sound like that is something negative.

[Min]

"The quantity theory of money says that the value of money depends on the quantity of money. If you double the stock of money, other things equal, the value of money will halve (i.e. the price of goods in terms of money will double)."

Err, that's not other things equal, is it? I mean, the nominal monetary values in contracts has to double, as well, right?

[Min]

"Take two otherwise identical economies, with the exact same stock of money today, but the first economy has a growing stock of money and the second economy has a constant stock of money. The first economy will have a higher price level today, even though it has the same quantity of money today."

This seems to be a contradictory state of affairs. If the first economy has a higher price level, then how can it be otherwise identical? Isn't there going to be some other difference besides the growth rate of money?

[Nick Rowe]

Min: That's what the QTM says. If you are arguing about whether the QTM is exactly true or not, then yes, you have to consider a lot of things like whether nominal debt contracts are also redenominated, because otherwise the distribution of wealth will change, and that may or may not have an aggregate effect, and may mean the price level will either more or less than double. There's also the question of whether it's just the total stock of money that doubles, or whether each individual's stock of money is doubled.

All theories are false. Some are useful.

[Don Lloyd]

See Mises' regression theory for the origin of the value of money (gold in particular). Yesterday gold was used exclusively for cups, let's say, and not for money. Today somebody had the idea of holding on to surplus gold cups for trading purposes, and it catches on. But we have a pre-existing value from the gold cup when none of them were money. In the future there is a combined supply and demand for both money and non-money gold and money is so useful that the monetary value eventually predominates.

The same thing applies to cigarettes in a prison camp. The starting point for the value of a cigarette as money is also its value before any of them become money. The key point is that an arbitrary value cannot be assigned for a new money.

Regards, Don

[Nick Rowe]

Don: I agree. I really like von Mises Regression Theory. I was hinting at it in my part 2. See my reply to Scott above.

[Nick Rowe]

Min: "If the first economy has a higher price level, then how can it be otherwise identical? Isn't there going to be some other difference besides the growth rate of money?"

If you are going to tackle that question rigourously, you have to build a model and decide what's going to be exogenous (so can't be different by assumption) and what's endogenous, and then figure out which of the endogenous things will change too. I prefer to work backwards, informally. The first has a higher inflation rate, and a lower M/P (real stock of money) according to the QTM. OK, so what would those things also cause to change? Well, if M/P is lower, but people had the same real income and expenditure, they must be rolling over their inventory of money more quickly. So they have more hassle in shopping. Their shoe leather wears out more quickly. Restaurants are having to change their menu prices more frequently, so costs of printing menus are higher. Etc.

[Sergei]

Most agents in the economy do not spend government money. They spend bank deposits.

"10% of the population will get beaten up, whatever happens"

Yes, it is called unemployment.

[Nick Rowe]

Segei: but bank deposits are (usually) convertible into government money. So their value is determined by that of government money. (Actually, there's more simultaneity than that, of course, because the availability of bank deposits as a substitute affects the demand for government money.)

Unfortunately, you can't avoid unemployment just by paying your taxes.

[Sergei]

"But if the total supply of money stays the same, even though each individual can spend or lend his money and pass it onto someone else, it isn't possible for all people to do get rid of their money this way"

No, it is possible. Higher spending means higher taxes.

[Sergei]

"bank deposits are (usually) convertible into government money"

Yes, this how we have setup our financial system. There is no law which says it can only be this way. And even this system fails once in a while.

Besides this, government creates very little of bank deposits. Banks create the majority of them.

"Unfortunately, you can't avoid unemployment just by paying your taxes"

It is the basic claim of MMT that if government does not spend enough, then unemployment will result. In your words "10% of the population will be beaten up, whatever happens".

[Nick Rowe]

Sergei: but then the supply of money is not staying the same, and I just assumed it was. For a simple example, assume that taxes are 40% of income, and government spending is the same 40% of income, and there's no government borrowing. Then as people's spending rises, both taxes and government spending rise by the same amounts, so the money the government is taking in rises, but the money the government spends out rises, so the total amount of money stays the same.

[Nick Rowe]

Sergei: "Yes, this how we have setup our financial system. There is no law which says it can only be this way. And even this system fails once in a while.

Besides this, government creates very little of bank deposits. Banks create the majority of them."

Agreed.

"It is the basic claim of MMT that if government does not spend enough, then unemployment will result."

Agreed; MMT does say that, AFAIK.

[Sergei]

Nick: "but then the supply of money is not staying the same"

Exactly! The act of spending reduces the supply of money due to taxes. You can not separate one from another. So by spending people in aggregate can reduce the supply of money.

"so the total amount of money stays the same"

Sounds like equilibrium to me. Why should prices increase if everything stays the same?

[rogue]

“Take two otherwise identical economies, with the exact same stock of money today, but the first economy has a growing stock of money and the second economy has a constant stock of money. The first economy will have a higher price level today, even though it has the same quantity of money today.”

Nick, questions regarding your assumptions:
1. What is causing the growing stock of money in the first? People spending more? More people starting new businesses? Government spending more? Or does the money stock just automatically rise?
2. If money stock is automatically rising, what is causing it? Helicopter money? Banks automatically lending out new reserves from the central bank?
3. If the two economies have the same quantity of money today, why does the first already have a higher price level today? With what are the people paying for the higher prices today?

[Nick Rowe]

Sergei: but I *assumed* the supply of money is staying the same. And in my example in my comment, where both taxes and government spending are the same 40% of total spending, the total supply of money is indeed staying the same, just as I assumed. If the government does not borrow, then government taxes reduce the stock of money, and government spending increases the stock of money, and if government spending and taxes are equal, the net effect of the flow out and the flow in on the stock of money is zero.

[Nick Rowe]

rogue:

1 and 2. Why is the stock of money growing over time? Either the government is spending more than it takes in through taxes, or else the government is printing money to pay down its debt (or the banking system is expanding if we are talking about bank money, or counterfeiting ;) ).

3. The supply of money today is the same in the two economies. But the demand for money is lower in the first economy than in the second. That's because people in the first economy know that the stock of money is growing over time, and so prices are rising over time (inflation), i.e. money is depreciating over time, so they don't want to hold as much.

How can they buy the same amount of goods at higher prices with the same amount of money (all compared to the second economy)? That same stock of money has to flow more quickly. It has to change hands faster than in the second economy. The velocity of circulation is higher. If I hold one months income and spending in my pocket, my money has to circulate 12 times faster than if I hold one year's income and spending in my pocket.

[Leigh Caldwell]

From Part 2: "Hell, even I could do that. "Here are 10 bits of green paper per person. Unless I get all 10 back at the end of the year I will beat you up. How much will you give me for them?". "Nothing" is one possible answer, even if my threat is credible."

I don't think "Nothing" is a realistic answer. I'll offer you something for your 10 bits of paper (and for 100 more of them) because I know I can sell them to other people who will need to give them back to you at the end of the year. The scarcity, combined with their practical application, creates value.

I don't think this means that taxes are in themselves necessary to create a value in currency - taxes are just one of many ways to establish the network effect which you have described.

And in Part 1, I don't believe taxes intrinsically give the currency value either. In this model, taxes are simply a way to reduce government borrowing - which reduces the broad money supply. But any other reduction in borrowing would have the same effect; if companies decided to fund more of their investment from retained earnings and less from borrowing; or if people bought houses with cash instead of mortgages.

The value of the currency arises (partly) from the desire to use cash instead of debt - government is just the biggest of many agents who can independently make the decision to do that. Taxes aren't special.

[Leigh Caldwell]

Ah - reading your reply to Scott, I think we are both saying the same thing.

[Nick Rowe]

Leigh: if the supply is 10, and the demand (to pay taxes) is also 10, they aren't scarce, unless some people want to hold onto some of them to do their shopping instead of paying taxes, but they won't do this if they think they are worthless.

But yes, one individual might buy more than 10, and act as a monopolist just before tax time. To get around your response I would have to make the supply 11, the demand for taxes 10, and assume lots of small people, so none could corner the market.

[TC]

You're forgetting about the desire for savings. People might be scared they cannot pay the 10 piece of paper per year tax. That's why governments don't have to tax more than they spend, and why plan B works in the real world.

I'll admit it is a bit funny watching Scott Sumner go through MMT 101. Taxes provide the initial demand for currencies, but not the final demand for currencies. It forces people to use the currency as a medium of exchange. After that, demand for savings in the medium of exchange is more important - this is the demand for money that we see in the real world.

Plus, wait until they figure out that Private Savings = Government deficit, and that it's also equal to the money supply. Mindblowing.

[Alex Godofsky]

Isn't the more important story here the fact that real taxes are typically levied as a proportion of the value of a transaction, rather than per capita? This forces the use of the currency as the medium of account, and you get out of the zero-value equilibrium through the same historical process as you described in part 2. As someone observed above, the medium of exchange is usually bank deposits, not dollars. Is it really ready convertibility that's important, here, or just the fact that deposits are denominated in dollars?

[Don Lloyd]

Please identify an 'official economic' version of the following :

The generally accepted concept of the supply of money is fundamentally flawed.

Money is not consumed by its involvement in an exchange. It merely is instantaneously transferred from one owner to another. A single piece of money can only have one owner at a time.

This means that money is not a consumption good, but, for lack of a better term, is a 'rental' good. This means that the real scarcity of money comes from its total time of possession, not its total dollar amount alone. With a fixed amount of money, there is no limit in theory on the number and size of exchanges that are supported, given fast enough computers with flexible scheduling. To spend $1000 I need to have it first, but only for a nanosecond before I spend it. If I hold it longer, 1 day or 30 days, more and more other people are prevented from holding and spending. The key is that the supply of money must have dimensional units like dollar-days per month. The total money supply will be DOLLARSx30 days per month, where DOLLARS is the conventional supply of money in dollars.

Let's say you have a fleet of 1000 rental cars. Your maximum capacity of providing rental services is 30,000 rental-days per month. This is a much more important number than 1000 rental cars. Because of timing conflicts you can't ever get to the maximum ( you might have 1000 people wanting the same rental day), but it's the same for money.

Regards, Don

[Nick Rowe]

TC: "Plus, wait until they figure out that Private Savings = Government deficit, and that it's also equal to the money supply. Mindblowing."

I-S+G-T+X-M=0

where I=investment, S=savings, G=government spending, T=taxes, X=exports, M=imports.

Rearrange, assume a closed economy, and S-I=G-T.

That accounting identity is what you (incorrectly) describe as "Private Savings = Government deficit"

I LEARNED THAT AT AGE 16 IN HIGH SCHOOL!

Sorry for shouting, but you really need to understand that this is NOT NEWS to ANY economist. I have taught this same accounting identity at least once a year for the last 33 years, mostly to first year students, though I remind the second year students about it, in case they have forgotten.

I am also sure that Scott Sumner learned this same accounting identity many decades ago, and has almost certainly taught it a dozen times since.

I, and Scott, also understand the difference between an accounting identity and an equilibrium condition, and that stating an accounting identity says absolutely nothing about what variables determine what other variables.

"...and that it's also equal to the money supply."

No it isn't. (And Scott understands that it isn't either). The government deficit (in nominal terms) is equal, not to the money supply, but to the annual growth in the stock of monetary base plus the annual growth in the stock of bonds in public hands.

And we teach this in second year too (and sometimes in first year).

Look TC, sorry to get over-excited by this, but *many* (not all) of you MMTers really don't have a clue. You think you have discovered something that nobody else knows. So you come across as a cult, saying "the number of children = the number of sons + the number of daugters, and this equation is what determines the number of children people have, and all those stupid demographers don't get it!"

(The top, serious, MMTers do, I am pretty sure, do understand that we all know this. It's mostly the camp-followers who don't get it, and find it a stunning new insight.)

I will calm down in a few minutes and maybe respond to the rest of your comment.

[rogue]

Nick: "That's because people in the first economy know that the stock of money is growing over time, and so prices are rising over time (inflation), i.e. money is depreciating over time."

So is MV=PQ a snapshot of price level at a certain point in time, or a description of the movement in price level over time? I ask because it's hard to get your mind around "Take two otherwise identical economies, with the exact same stock of money today, but the first economy has a growing stock of money and the second economy has a constant stock of money." How can you say both economies have the same stock of money when one has a growing stock?

Take for example: "How can they buy the same amount of goods at higher prices with the same amount of money (all compared to the second economy)? That same stock of money has to flow more quickly."

If we assume a point in time when both economies' stock of money is at the same level, then price levels haven't risen yet. Once it has risen, it would be at a point when the amount of money is already higher. The same stock of money flowing more quickly should not in itself lead to higher price levels unless people are bidding up the prices while getting rid of money. And people can only bid higher if they already have more money to pay with. Am I missing something here?

[Nick Rowe]

Alex: "Isn't the more important story here the fact that real taxes are typically levied as a proportion of the value of a transaction, rather than per capita? This forces the use of the currency as the medium of account, and you get out of the zero-value equilibrium through the same historical process as you described in part 2."

Hmmm. Interesting theory. If, every time you buy something, you have to pay government money equal to (say) 10% as sales tax on the transaction, you might as well use government money for the whole transaction, so it becomes the medium of exchange.

Logically coherent theory, I think. But I don't believe it! Aren't broad-based sales taxes fairly recent?

"Is it really ready convertibility that's important, here, or just the fact that deposits are denominated in dollars?"

Well, the value of what's in my chequing account is determined by the fact that I believe the bank's promise to redeem (convert) it at a fixed rate into central bank currency.

But if I have a chequable account in a stock mutual fund it would be different. There, the value of my account would be determined by the real value of the company shares I own.

[Lee Kelly]

On part 2: I think you're right. I had just assumed this was the standard explanation.

Just like the government can increase the demand for health insurance by mandating that everyone buys health insurance, they can also increase the demand for bits of green paper buy mandating that taxes be paid in bits of green paper. That said, I do not think that explains why the equivalent of green bits of paper are the media of exchange today, it merely provides another reason why fiat monies do not immediately collapse. I think the main reason is that modern fiat monies benefit from the residual networks effects of commodity-backed ancestors; this is what I think you claim too.

However, perhaps the tax effect on money demand explains why national borders are also monetary borders -- as soon as you cross an arbitrary line, something else is the medium of exchange. This strikes me as inexplicable unless taxes and other regulations are exerting their influence.

[Nick Rowe]

Don: I'm not sure. But isn't everything you say there consistent with MV=PT ? You are saying that "money" should be measured as a flow, which means MV ?

rogue: "I ask because it's hard to get your mind around "Take two otherwise identical economies, with the exact same stock of money today, but the first economy has a growing stock of money and the second economy has a constant stock of money." How can you say both economies have the same stock of money when one has a growing stock?"

This is easier explained with a graph:

Put time t on the horizontal axis and the stock of money M on the vertical. Draw a red horizontal line about halfway up. Draw an upward-sloping blue line that crosses the red line about in the middle of the graph. At time t*, which is "today", the blue and red lines cross. Red and blue economies have the same stock of money today, but the blue economy has a growing stock of money and the red economy doesn't.

Now draw a second graph directly below the first. Put time on the horizontal axis again, and the price level P on the vertical. The red line will be horizontal, and the blue line will slope up, as before. But at t*, today, the blue line's P will be above the red line's P.

What's happening is that V is higher in the blue economy than the red economy, even though Y and M are the same, at t*.

[Nick Rowe]

Lee: I don't think there is a standard explanation for my Part 2 question. (Maybe many economists don't even see there is a question, since it doesn't get talked about much.)

"I think the main reason is that modern fiat monies benefit from the residual networks effects of commodity-backed ancestors; this is what I think you claim too."

Yes, that's my main preferred explanation. Though I wouldn't rule out other possibilities, historically.

"However, perhaps the tax effect on money demand explains why national borders are also monetary borders -- as soon as you cross an arbitrary line, something else is the medium of exchange. This strikes me as inexplicable unless taxes and other regulations are exerting their influence."

Something that's always puzzled me too. There *is* currency substitution, internationally, especially near "borders". But there's less than one would expect, prime facie. Compare it to language shifts, near borders, since languages also have network effects. Does money have a sharper cutoff than languages, at national borders? The tax and regulation explanation seems roughly plausible.

[Lee Kelly]

Nick,

It's a question which has occurred to me before, and I am not an economist. I had just assumed there would be some standard answer and this was it.

In any case, how do you explain, say, how France changed over from francs to euros? The euro was conceived a fiat money; it never had a commodity-backed ancestor, right? Was it just because everyone knew that the French Government would soon begin collecting taxes in euros?

[Nick Rowe]

Lee: suppose (as in Canada) they withdraw the $1 note and replace it with the $1 coin. The old note was convertible into the new coin. And 5 new $1 coins are convertible into a $5 note. People think they will have exactly the same value, so they do, unless something else changes. It's not so different with a currency reform. Or with the introduction of the Euro, followed by a long period of fixed exchange rates. "The King is dead, long live the King!" The focal point of who we obey just follows the line of succession naturally. Unless there are two competing claimants for the throne, and nobody knows who everyone else will obey, so nobody knows who to obey.

[Don Lloyd]

Nick : "Don: I'm not sure. But isn't everything you say there consistent with MV=PT ? You are saying that "money" should be measured as a flow, which means MV ?"

I can't see any way in which what I am talking about can be described as a flow. If there is $100 of money in a 4 person economy, then that $100 is fully distributed among the 4 every day of the month. This is a static concept for each day.

The re-distribution of the money overnight (for simplicity only) is only limited by the fact that the total ending distribution must remain at $100 total.

If one person, for some reason, decides to hold $97 dollars each of the 30 days of the month, then there are only $3 available for the other 3 people to hold in total for each of the 30 days of the month.

Let's say Congress, in its infinite wisdom, decides to pass a new law (separate versions for the House and the Senate). The House mandates that everyone paying either a mortgage or rent must deposit their payment in a holding account exactly 1 day before payment is otherwise due. The Senate version is similar, except that the holding account must be filled 21 days in advance.

If the Senate version is passed, as opposed to the House version, is it not clear that the demand to 'rent/hold' money will be much greater and the resulting unit value of money will likely be much higher? Remember that the passing of the new law or not, and which version, have NO effect on the mortgage/rent exchanges.

Regards, Don

[Lee Kelly]

Nick, so how does this prove that MMTers are wrong?

[Nick Rowe]

Don: I'm still not sure I'm following you. I'm going to make another stab in the dark. This may be tangential, or may be what you are getting at.

Suppose each individual can buy and sell whenever he likes. So each individual has a desired pattern of daily ups and downs in his desired money balances. What, if anything, is there to ensure that when we add up those daily ups and downs of each individual, the aggregate comes out nice and smooth over the month?

A small number of monetary economists, I believe, look into questions like this. I'm not one of them. I think there are four answers:

1. By sheer fluke, with randomness, and enough people, it may come out smooth in aggregate. Unlikely.

2. If prices were perfectly flexible, on a daily basis, prices could adjust daily so that individuals desired ups and downs cleared all markets and so money supply equals money demand at both the individual and aggregate level on a daily basis. Also unlikely.

3. The banking system allows enough flexibility in the aggregate money supply to let individuals' plans add up. This certainly does happen on a seasonal basis, ever since central banks decided to allow it. And IIRC this eliminated seasonal fluctuations in some asset prices/interest rates that previously followed the agricultural cycle. There's a literature on this, but I'm not familiar with it.

4. The very nature of an inventory is to act as a buffer stock to absorb unexpected shocks to flows in and flows out. If I unexpectedly arrive at a car lot, and buy a car, the dealer's car inventory is lower than he planned, and his money inventory is higher than he planned. He is willing to accept that extra money temporarily, even though he doesn't wish to hold it. It's a hot potato, that he gets rid of as soon as it's convenient for him to buy what he wants.

In general, if something coordinated all our plans so that each of us got paid just before all our bills were due, we would all demand to hold a smaller average stock of money than we do now. But in a decentralised economy, and ones where people's plans are always changing, and where it is costly for me to visit the supermarket at exactly the time the supermarket wants me to shop there, it is very hard to do this. And that's why we all hold buffer stocks of money, and the supermarket holds buffer stocks of milk, and I hold buffer stocks of milk in the fridge, etc. Money, like other inventories, saves us the costs of having to coordinate our plans.

That's my take anyway.

But I may have missed your point, again.

[K]

Here are a few random thoughts: if the central bank is independent, then government bonds are no different from corporates as assets of the CB. Both are just a risky nominal asset backed by the revenues (taxes in the case of government) of an unrelated third party. And if the cb has some equity, the valuation of its liabilities become a very standard capital structure valuation problem, and there is nothing particularly Ill determined about it. But if it's not independent from government and the balance sheet is backed by tbills, then it's backed by taxation just like those tbills on its balance sheet.

Another random thought: The proper way to run a cb is as an independent organization whose assets form as broad a cross section as possible of the investable capital assets of the economy. When you did you blue sky post I argued that all money (including bank deposits) should be replaced with money created this way. These latest, really excellent, posts of yours have only convinced me more.

[Nick Rowe]

Lee. I'm not really trying to prove the MMTers, or the Fiscal Theory of the Price Level, or Real Bills theorists, or New Keynesians, or anyone else, are wrong in this post. I'm just trying to lay out the QTM.

I'm not 100% clear on how MMT exactly relates taxes to the value of money. That may be because they aren't exactly clear, or it may be because I haven't read them carefully or fully enough. I thought it would be more productive if I tried to lay out the QTM version of the relation between taxes and the value of money as clearly as I could, because that's something I'm better at. I wanted to go further than just say something like "Paper money has value because the state has the power to tax".

I could put forward my own opinion on exactly what the MMT view is on this question, and why MMT, as I interpret it, is wrong. But I wouldn't trust my interpretation. And the whole debate would just dissolve into a hopeless mess anyway, over terminology, stocks and flows, real vs nominal variables, the shape of the AS curve, accounting identities vs equilibrium conditions, etc.

This stuff is hard. Rogue, for example, is making a good-faith attempt to really get his head around some of the things I'm saying, and is not finding it easy. If it were a debate, with people committed to their own views, it would be even harder.

[Nick Rowe]

K:

Your first paragraph has two possibilities: the first describes the "Real Bills Doctrine" (as interpreted by Mike Sproul for example -- see my last post for the link to Mike); and the second describes the Fiscal Theory of the Price Level. Great random thoughts! Both give very different theories than the QT.

Second paragraph. Sometimes I think that's a good idea too. Monetise everything! And thanks!

[K]

It's actually the same idea as SDRs. We replace the USD as the international currency with units which we create by exchanging national currencies ("capital assets") for units of a fund (the CB) in whatever quantity is demanded at the market price of those currencies. The fact that units are created and destroyed on the demand of anybody with currency is what would make the real bills view incontrovertible.

[Giang Le]

Look at Bitcoin, it came to existence and acceptance without any tax/beating up threat or convertibility in the first place. It is the convenience (e.g. buying illegal drugs etc.) and low transaction costs (and thereafter the network effect) that make Bitcoin a medium of exchange. If governments let it go, Bitcoin or similar protocols may well become a true international currency one day. I think a big shortcoming of monetary economics is the disregard of the technological aspect.

[Nick Rowe]

Giang: yes, if Bitcoin survives (and if I understand it correctly) it would be one of the purest examples of a quantity-theoretic money. It has value because people want to hold it because it makes their shopping (for certain goods) easier. No backing, no taxes, no convertibility. Period.

[Min]

Nick Rowe: "Well, if M/P is lower, but people had the same real income and expenditure, they must be rolling over their inventory of money more quickly."

Thanks, Nick. That's pretty much what I guessed. :)

Nick Rowe: "All theories are false. Some are useful."

You can prove anything from a falsehood. :) That's why, IMO, you need to talk about the limits of validity, about context and errors.

Nick Rowe (@ TC): "Look TC, sorry to get over-excited by this, but *many* (not all) of you MMTers really don't have a clue. You think you have discovered something that nobody else knows."

I think that that phenomenon reflects the (IMO) parlous state of much public economic discourse in English. Economists is the past year have said that the US is going bankrupt. Legally, that is incorrect. However, they may be speaking rhetorically. Still, their lay audience on the whole thinks that they mean it legally, and that scares them.

Last year, when the debt/deficit hawks began their onslaught, a few economists called them on their BS, based upon the identity that you learned when you were 16. But their voices were not prominent. Most of the economic profession remained mum. What evidence would a layman have that the economists who remained mum were aware of that identity?

BTW, this is related to theory. An identity is like a conservation law in physics. To me, that makes the lack of a general response by economists against the falsehoods of the debt/deficit fear mongering last year somewhat surprising. (Not that there are no real concerns, but they are different from the false fears raised in the initial rhetorical onslaught.)

[Mike Sproul]

Nick:

Your ability to beat people up is your asset, and the money you issue is your liability. If the present value of all your future beatings is 1000 oz. of silver, then you could issue as much as $1000 of paper money, and each paper dollar will be worth 1 oz. If you issue $2000, each dollar will be worth .5 oz. If you lose 40% of your assets, each dollar will drop in value by 40%. Your dollars are valued just like any ordinary liability, and you don't need weird quantity theory concepts like money demand, velocity, etc. to explain their value.

[Greg Ransom]

These economies will NOT be identical -- relative price relations will be systematically shifted/adjusted by a growing stock of money flowing into particular places in the economy, systematically shifting/adjusting price relations across the time structure of coordinated production and consumption plans, CHANGING output plans, e.g. changing preferences for output from production processes that promise greater output as production time is lengthened.

Hence, the important difference from the point of view of causation and patter change is systematic and structural: change in "price level" is of relatively superficial significance.

"Take two otherwise identical economies, with the exact same stock of money today, but the first economy has a growing stock of money and the second economy has a constant stock of money. The first economy will have a higher price level today, even though it has the same quantity of money today."

[Nick Rowe]

Min: "That's why, IMO, you need to talk about the limits of validity, about context and errors."

Yes, maybe I should have done, in this post. But it would have been a lot longer post, and not as clear. I just wanted to lay out the QT, not give it a critique. And I might have needed to lay out a number of alternative theories too, because all theories have errors, some the same, and some different, and we can only really judge a theory by seeing if and when it does better than alternatives. And that would have been a whole book, and one I'm not up to writing.

"Economists is the past year have said that the US is going bankrupt. Legally, that is incorrect. However, they may be speaking rhetorically. Still, their lay audience on the whole thinks that they mean it legally, and that scares them."

In one sense, the US, as a country that can print the money it owes, cannot be forced into bankruptcy. "$1zillion Federal debt? Just print $1zillion banknotes, and pay it off." But, *under some circumstances*, that would be hyperinflationary, and in economic terms equivalent to bankruptcy, whatever the lawyers might say. The us would pay back the debt, but using paper that is worth on 1 cent on the dollar in real terms. And that distinction between real and nominal magnitudes is one that *many* MMTers (and others, perhaps) seem to me to slide over.

A year or so back someone at the Economist (RA?) put forward a "Ketchup Theory" (more of a metaphor really). You can't get the ketchup out of the bottle, so you keep hitting it harder and harder, and then all of a sudden it all comes pouring out in a rush. The US keeps on adding debt, and while the economy is in its current state it seems to do nothing whatsoever. People are willing to hold very large amounts of debt and money at very low interest rates. But there might be a "tipping point", either endogenous or exogenous, and all of a sudden people want to start spending again, and don't want to hold all that money and debt. And the US would need to redeem money for debt and raise interest rates to persuade people to hold that debt rather than spend, and so prevent inflation. And at that higher equilibrium interest rate, the US might face an ugly choice between massive sudden fiscal austerity, or legal default, or economic default via hyperinflation.

Now, I certainly do not have any fully worked-out theory of this tipping point, but I also recognise that things do change. If something, anything, does cause some people to start spending again, there is a more-or-less well-understood Old Keynesian multiplier and accelerator effect that would cause others to increase desired spending too, and it could well snowball.

And MMTers do not seem even to recognise this possibility, because: they are very sloppy on the distinction between real and nominal magnitudes; do not recognise the standard role of real interest rates via a substitution effect on desired saving and investment. So they think that the US can set any real interest rate it wants, even in the long run. Their IS curve is vertical, while everyone else's IS curve (or equivalent) slopes down.

I cannot rule out the possibility that those who fear default, in economic terms, might be right. There is something to be scared about.

Someone like (say) Paul Krugman would understand what I'm talking about, but think the risks of expanding the deficit to get us back to normal quickly would be less than the risks of reducing the deficit. He might be right. But I would prefer using monetary policy to try to reduce both risks. So would Scott Sumner. So, I think, would Paul Krugman, if he were more optimistic about what monetary policy could do. But I don't see any sign that Functional Finance people in general, and MMTers in particular, have any grasp of these issues at all.

"An identity is like a conservation law in physics."

Not really, unless I have a different understanding of conservation laws in physics. Those conservation laws in physics are normally understood (I think) to be telling us something about the real world. They are telling us something that needs to be tested, and that might be wrong. Identities are simply mixes of definitions about how we use words, plus some math to rearrange those definitions.

[Nick Rowe]

Mike: but how does that handle the objection I put forward in Part 2? Governments typically issue money when they want to spend more than they collect in taxes. So I give out 11 bits of paper per person and only make them pay back 10 to avoid a beating. So nobody gets beaten up in equilibrium. So even if my conditional threat is credible, those conditions are never met. They accept the 11 bits of paper, give 10 back, and leave 1 lying on the sidewalk. The paper isn't scarce.

Greg: In a steady, fully-anticipated, inflation? That is ABCT, and I have never understood it. Yes, if the government spends the seigniorage revenue/inflation tax/central bank profits on extra tanks, the relative price of tanks is bid up, and there are more tanks and fewer other things produced. If the government gives that seigniorage revenue back as lump-sum transfers, it's a wash.

[Nick Rowe]

Plus, that seigniorage revenue is peanuts anyway, as a percentage of GDP, unless we get serious about creating inflation. The Bank of Canada's profits are around $2B per year, in normal times, at 2% inflation and 5% NGDP growth.

[Min]

Moi: "That's why, IMO, you need to talk about the limits of validity, about context and errors."

Nick Rowe: "Yes, maybe I should have done, in this post."

Sorry, Nick, I did not mean you personally, I meant the generic you. :)

Nick Rowe: "In one sense, the US, as a country that can print the money it owes, cannot be forced into bankruptcy. "$1zillion Federal debt? Just print $1zillion banknotes, and pay it off." But, *under some circumstances*, that would be hyperinflationary, and in economic terms equivalent to bankruptcy, whatever the lawyers might say."

The point is not so much what the lawyers say, as how lay people interpret the claims of imminent bankruptcy. A quick look in the dictionary shows the legal definition and the metaphorical one, but not the equivalency to hyperinflation. I also checked The New Palgrave Dictionary of Economics Onlne ( http://www.dictionaryofeconomics.com/dictionary ), the Babylon Economics Dictionary ( http://www.babylon.com/define/35/Economics-Dictionary.html ), and Economypedia ( http://www.economypedia.com/ ). Many ordinary citizens came to believe that the US was running out of money, that it was going broke. The fear mongers had a great rhetorical success, imposing a pall over public policy discussions that greatly influenced last year's elections and still exists. In the current circumstances, a number of prominent economists have spoken up, but the best time to do that was over a year ago. when the false claims were first trumpeted.

Moi: "An identity is like a conservation law in physics."

Nick Rowe: "Not really, unless I have a different understanding of conservation laws in physics. Those conservation laws in physics are normally understood (I think) to be telling us something about the real world. They are telling us something that needs to be tested, and that might be wrong. Identities are simply mixes of definitions about how we use words, plus some math to rearrange those definitions."

The fact that formal identities are tautological truths is not a mark against them. Failure to be consistent with a conservation law may be because the law is disproven, failure to be consistent with a formal identity is simply a mistake. Scientists should denounce such errors when they occur in public debate and not gloss over them.

One thing I wanted to indicate is that conservation laws, like identities, may not suffice for prediction. For instance, the click of billiard balls colliding is only partially explained by the conservation of energy and the conservation of momentum. Those two laws tell us that the energy of the collision is typically greater than the kinetic energy of the balls' motion. Some of the excess energy is dissipated as sound waves. Hence the click. But in a vacuum no such sound waves would be generated. And back in the days of celluloid billiard balls, sometimes the energy would be dissipated in an explosion. ;)

When politicians say that both the public and the gov't should tighten their belts while maintaining a strong currency, don't you wonder if they know what they are talking about?

[Winslow R.]

"I-S+G-T+X-M=0

where I=investment, S=savings, G=government spending, T=taxes, X=exports, M=imports.

Rearrange, assume a closed economy, and S-I=G-T.

That accounting identity is what you (incorrectly) describe as "Private Savings = Government deficit"

Sorry, short of time so I just scanned comments, at the risk of Rowe wrath....

1) It looks like your QTM model is also based on a closed economy. China etc, is able to credibly force its citizens to hold U.S. dollars. In other words, the value of the dollar is externally determined as well as internally determined.

2) Why do economists include spending on capital goods in their definition of 'savings' so that it needs to be subtracted by including investment (S-I)? Why bring the real economy and real resources to an equation that essentially describes the financial economy and financial resources and in particular vertical money? Does keeping both S and I in the equation really provide more insight than confusion?

[wh10]

Nick, what if it was stated as savings of net financial assets? I think, but don't quote me, this is what MMT really says (RE: your ire of the camp who ostensibly misstate accounting identities already known to all economic schools of thought).

If you think of the Govt sector, the domestic private sector, and the foreign sector in a situation where the govt deficit spends $100 (which by accounting identity is $100 more in NFA), and the private sector net sends out $30 to the foreign sector, then the govt sector has a net deficit of -$100 in NFA, the private sector has net savings of NFA of $70, and the foreign sector has net savings of NFA of $30. If it was simplified to closed economy, then the govt deficit of $100 = private sector savings of NFA of $100.

[Scott Sumner]

Nick, If cash has value because it was once backed by gold (but never will be again), does gold have value because it once backed cash (but never will again?) My hunch is that the answer is that about 1/2 to 2/3 of the value of gold derives from its history, the fact that it once backed cash.

But there might be a good counterargument---platinum---so I'd be interested in your view.

[Patrick]

The government is the network effect? The gov't doesn't just use its "or I'll beat you up" power to take away (and often in return provide what you wouldn't be able to get), it also provides cover for private transactions to be completed, even when parties don't know or trust each other all that well. Sure, in most places they can agree ahead of time to settle the transaction however they please, but if they don't bother to do that, and a debt is incurred then the courts will only use their "or I'll beat you up" power to extract payment in legal tender. In a dispute, the court won't AFAIK, take your car (for example) and give it to the other party and call it even. They'll tell him to give you money (which may for him to sell his car).

It's like we all agree that we'll have one enforcer for debt payments, and he will only agree not to beat you up if you pay your debts with legal tender (in the absence of other prior arrangements), and even if he does beat you up he'll make you bleed legal tender. Since it's cheaper and easier to just use the shared enforcer and the medium of exchange he recognizes, most people do just that.

This probably has nothing to do with QTM, but it seems to me to be a big part of why people use (or don't) a nations currency within its borders.

[Nick Rowe]

Min: on the fearmongers. If they said "the US will face a choice between default or a hyperinflation that would destroy the monetary system", it's not that much less scary that their saying "the US will default". (Whether they are right is, of course, a separate question, and whether their reasons for saying that are right, is a third question.)

I wish I could say their conclusions are clearly BS, but I can't.

"the number of my children = the number of my sons + the number of my daughters" is to demographic theory what S-I=G-T (closed economy) is to economic theory. It can sometimes help us make sure we are adding things up right, but it tells us nothing about what determines what. That identity is true in every logically consistent theory of economics which includes S,I,G, and T. It is, for example, consistent with a theory that says increases in G cause equal decreases in I, and cuts in T cause equal increases in S, so increases in G or cuts in T don't increase AD and Y.

[Nick Rowe]

Winslow:

Yeh. I'm a bit sorry I lost it with TC. He's a good guy. But oh man. Someone like Scott, or me, has spent most of our time for the last 40 years reading or listening or thinking about economics. Doesn't mean we're right. But it does mean we've learned lots of ways to suck eggs.

1. The closed economy thing is no biggie. I just mentioned it for completeness. It's OK to simplify. Does the QT assume a closed economy? Not really. It should generally work about as well in open or closed. Sure, there may be cases where an open economy throws an additional spanner (wrench) in the works.

2. Defining "S'" the way every other group of economists defines "S-I" *is* a problem. Sure you can define words how you like. And different definitions may sometimes work better for different theories or purposes. But it sure makes communication with others hard. And arguing with others very hard. I have already seen one major case of needless confusion caused by this peculiar MMT definition of S'. Both MMTers and all other economists sometimes see differences where there really are none. In this case, everything in MMT could easily be re-stated in standard economic definitions that every economics student learns in first year.

The original reasoning behind the standard definition of S is that S is that part of people's disposable income which is not consumed, and so can be added to our wealth. This is a useful definition if an individual faces a choice between consuming today, or saving and consuming tomorrow. We can then ask what determines the desired mix between C and S for an individual. And we may need to know the answer to that question, if we want to understand AD.

Also, the Keynesian vision was that households choose how much to save, and firms choose how much to invest. In a very simple model, closed economy, no government, that leads us to the question: "what, if anything, ensures that households' desired savings equal firms' desired investment?". And the whole of Keynesian (for example) macroeconomics was built around this question. Is it income that adjusts to equalise desired savings and desired investment, or is it the rate of interest? An awful lot of ink was spilled in answering that question. And we don't want to have to re-open all those debates simply because someone wants to re-define S' as S-I. It's hard even to restate the old problem in that new S' language.

Yes, we really do need to keep S and I separate.

(And notice, by the way, that I was very careful to say *desired* S and I. An awful lot more ink was also spilled on understanding the difference between S=I as an accounting identity and as an equilibrium condition.)

[Nick Rowe]

wh10: "Nick, what if it was stated as savings of net financial assets?"

Yep. Maybe not an ideal choice of words? But I get your drift.

But it does throw together two very different sorts of decision-making entities, households and firms. (OK, even though some investment is done by households). Plus, even if it's a household doing the investing, the reasons behind the savings decision are very different to the reasons behind the investment decision. There are two questions:

1. How much do I want to consume today vs save and consume tomorrow?

2. Do I want to save in financial assets or buy real assets like machinery?

There's also a theorem ("The savings-investment separation theorem"??, my memory has gone) that says under certain conditions those two decisions are totally independent. Paradoxically, how much I want to invest is totally independent of how much I want to save.

So, S-I bundles together two very different choices.

[Nick Rowe]

Scott: my logical brain says "no", because gold isn't used as medium of exchange, so there's no demand for gold based on its history as money.

But my gut says "yes". Gold, because of its history, is something more than an industrial commodity. It's got some sort of symbolic value. As jewelry, obviously. But maybe jewelry is just a symptom of something more.

I would trust my gut more in this case.

[Nick Rowe]

Old masters would be my metaphor for gold. Nobody buys them because they are beautiful. A good reproduction is just as good. We buy them for their history. And because everyone else wants them, so it's another network effect.

Patrick: maybe. It could be hard for governments to enforce transactions in strange monies. (And, IIRC, in some forwards/futures/whatever markets governments allow settlement in cash rather than goods, despite what the contract says???). You could also think of the national currency as the "default option" in contracts, which simplifies things all round.

[Jacques René Giguère]

Historically, the chines gov't made peasants use paper money in the 10th-11th centuries by telling them that the paper soldiers used for paying food would be accepted for tax payment and did so.

For those who fear hyperinflation:
HI never happened because some CB suddenly woke up some morning and multiplied monetary base or any other forms of currency in circulation. It always come from a sudden real shock. It may be war(Central Europe after the collapse of the Austro-Hungarian Empire), foreign occupation followed by general strike (Ruhr 1923), tin miners strikes collapsing the exchange rate ( Bolivia any time you can think of) or toal collapse (reckless voluntary destruction ot the agricultural sector in Zimbabwe).
If there isn't any goods left, the price will jump to infinity whatever happens on the monetary front.
In 1996 in Sept-Îles. the heavy storm known as the "déluge du Saguenay"

http://fr.wikipedia.org/wiki/D%C3%A9luge_du_Saguenay

http://www.google.ca/search?q=d%C3%A9luge+du+saguenay&hl=fr&rls=com.microsoft:fr-ca:IE-SearchBox&rlz=1I7GGLD&prmd=ivns&tbm=isch&tbo=u&source=univ&sa=X&ei=AeoxTrXdK8Th0QGqnuXaCw&ved=0CC8QsAQ&biw=847&bih=485

destroyed the road leading into the town. Within 3 days, the price of milk had decupled on the de facto black market... Soon enough people would have exchanged their 8-year-old daughter for a can of beans. When the first relief ship came in, prices crashed in a few minutes. No CB was involved and yet all the signs of HI were there.

[rsj]

Is there are separate demand for paper money rather than electronic money -- i.e. deposits? Can $1 of deposits be worth more or less than $1 of paper money?

But households, in aggregate, can easily convert their paper money into electronic money (e.g. deposits) by depositing cash. What fool would bid up the price of a good to get rid of his paper money when banks are offering to pay interest to take it? Similarly, if households want to get rid of electronic money, they can easily do so by purchasing bank equity. That converts a deposit claim on a bank into an equity claim on the bank.

They can also convert to other claims on banks -- e.g. bond claims, commercial paper, repos -- there are all sorts of ways that households can lend money to banks.

Again, what fool would bid up the price of goods in an effort to get rid of his deposit when he could obtain an interest bearing bank bond instead?

Not a single good needs to be purchased for households to cut their money holdings in half, or by 2/3, or whatever they want.

[Max]

You are confusing the question of what powers the government needs to have with what the government needs to actually do given its powers.

And "the government is spending its money, and the money is worthless" is self-contradictory. Either the government is spending its money or it is worthless, can't be both.

[Nick Rowe]

Jacques: "Historically, the chines gov't made peasants use paper money in the 10th-11th centuries by telling them that the paper soldiers used for paying food would be accepted for tax payment and did so."

Did the govt also say that each bit of paper would be equivalent to a fixed amount of gold (or silver?) in payment of taxes? If so, that would make it a convertible currency, because you can, in effect, convert it into gold at tax time. And there's no problem in understanding why a convertible paper money has value. Canadian Tire money has value too, because I can pay my CT bill with it, and so it's, in effect, convertible at a fixed price into Loonies, just at CT rather than Revenue Canada.

Zimbabwean prices increased a zillion-fold (whatever). The money supply increased a qillion-fold too, and the rate of increase of the money supply increased a pillionfold. What happened to food production? Did it halve? Peanuts, compared to what was happening to M, Mdot, and P. Food production was a rounding error, in comparison.

Saguenay, I can see as a real shock. What happened to food supply? Probably dropped 90%? More?

Again, the QT does not say that M and Mdot are the only things that matter.

[Min]

Nick Rowe: "on the fearmongers. If they said "the US will face a choice between default or a hyperinflation that would destroy the monetary system", it's not that much less scary that their saying "the US will default". (Whether they are right is, of course, a separate question, and whether their reasons for saying that are right, is a third question.)

I wish I could say their conclusions are clearly BS, but I can't."

I happened to follow the rhetorical assault of the fearmongers pretty closely last year. Their early claims (in the spring) were obvious BS. The early onslaught had the effect of turning the public discussion away from jobs and recovery to fears of the long term debt. The early claims had not been debunked, except for a few voices, but were no longer being made (by June or July). Instead, less unreasonable claims and concerns were being discussed. But the climate of fear remained. I think that it is these later claims that you are referring to.

To be sure, hyperinflation is scary, but I think that it is less plausible to TCMITS (The Common Man In The Street) than the gov't running out of money and being unable to pay its bills. Had the initial rhetorical attack said that we are in danger of hyperinflation, I doubt if it would have achieved its effect.

Nick: "the number of my children = the number of my sons + the number of my daughters" is to demographic theory what S-I=G-T (closed economy) is to economic theory. It can sometimes help us make sure we are adding things up right, but it tells us nothing about what determines what. That identity is true in every logically consistent theory of economics which includes S,I,G, and T. It is, for example, consistent with a theory that says increases in G cause equal decreases in I, and cuts in T cause equal increases in S, so increases in G or cuts in T don't increase AD and Y."

The same goes for the law of conservation of energy. As we know it can sometimes be violated, but when it is not, it does not really say anything about causality. If I pull back on a bow string, I may shoot an arrow, or I may break the bow. Energy is conserved either way. :) Conservation laws tell us about constraints on the system. They reduce its degrees of freedom.

[Min]

Not to beat a dead horse, but the identity S - I = G - T has four variables, but there are only three degrees of freedom. That's how it acts like a conservation law.

[Nick Rowe]

rsj: "Is there are separate demand for paper money rather than electronic money -- i.e. deposits? Can $1 of deposits be worth more or less than $1 of paper money?"

There is a separate demand, but because deposits are convertible on demand *at a fixed price* they stay priced at $1 (as long as people believe they are convertible, and if not, they trade at a discount, from historical experience, I think).

I did say that when there's an excess supply of money, people will try to spend *or lend* the excess. You are assuming they lend the lot. OK, but if aggregate Ms is fixed (as I assumed) that will create an excess supply of loans, force down r, and that will lead people to try to spend. So prices get bid up.

Max: "You are confusing the question of what powers the government needs to have with what the government needs to actually do given its powers."

You lost me there, sorry.

"Either the government is spending its money or it is worthless, can't be both."

OK. Let me say the same thing in a different way. "The government is offering its money in exchange for goods, but the money is worthless, so it gives up its money and gets no goods in return".

[Jacques René Giguère]

Nick: The Chinese gunmint used Gresham's Law but in reverse. The peasants, fearing paper would turned worthless,used the new currency and hoarded silver. But the gunmint kept accepting paper till the peasants were convinced of its value and safety. After all, unlike Europe , where metallic coins were routinely altered, the Chinese gunmint was efficient and honest.
The gunmint then began to buy back silver. In part, because before the Great Isolation, the Emperor was sending huge fleets abroad (Asia, Africa, probably America). Silver, unlike paper, could serve as foreign currency (border effect).

Zimbabwe is a low-income low-GDP country. Most of GDP and consumption is food. Price elasticity is low and there may be a Giffen effect. Even a slightly lower production ( and reduction seems to be rather high by anecdotal evidence, no data can be trusted) will send prices skyrocket. Most exports were agricultural, so they crashed, sending the exchange rate down ( bolivian tin miner strike effect) and so every price go up. Taxable income disappear,tax collection not being either efficient or honest to begin with, the CB prints money and off you go. Obama may be half-kenyan but do anyone outside the Tea party believe 75% of GDP and 99% of export will magically disappear next Tuesday?

[Gizzard]

I really cant believe that serious, thinking people can even consider that the US could possibly go down the road of ZImbabwe.

ZImbabwe had what micro percentage of world GDP? How many people on their best days had ever transacted or were saving in their currency? Have they ever had a bond market to speak of?

The US is currently debating over what level of govt spending/borrowing (Ill use the term even though the US NEVER borrows to spend) to continue with. Cuts of 10-20% in some programs like SS and Medicare are being considered amongst other things, even though those two programs are virtually fully funded by special taxes. So in an economy of 15+ trillion annually we are talking about less then 10% of current GDP that is the ballpark range of considered spending adjustments. This is NOT hyperinflation numbers boys.

And will Nick or Scott please tell me exactly how these bond traders will reject the US dollar and why? When you get rid of something there has to be a market for it, so by definition if these guys are getting OUT of dollars, someone else is getting INTO dollars. They wont just burn their bonds, not these money grubbers. Zimbabwe and Weimar Germany didnt have a bond market, didnt have a CB that was coordinated with the rest of the world , they were island economies essentially isolated from the rest of civilization.. Comparing the US to either of those is absurd beyond belief. You should both lose your doctorates for even bringing them up in serious discussion about the US fiscal situation.

[rsj]

Nick,

But I am arguing that aggregate M_s *cannot* be fixed if bank deposit liabilities are accepted as payment. And that creates a bit of a funny situation because nominal interest rates are not going to fall -- as they reflect expectations of future short rates. (nominal) income, rather than rates, is going to decline.

Another way you can see that directly is that if I use my income to purchase consumption, then I am supplying income to someone else. But if I use my own income to lend to someone else, then that is a capital transaction, and the rest of the economy's income declines by the amount that I lend.

And in this sense, there is no difference between purchasing a deposit or purchasing a bond with my savings -- the fact that I have refrained from purchasing consumption means that the rest of the economy's (nominal) income has declined by the amount that I have saved.

Similarly, there is no difference between borrowing to purchase consumption or selling an existing asset to purchase consumption. In both cases, the (nominal) income of everyone else increases to due to my purchase.

If you believe that changes in nominal income affect prices of consumption goods, then as many people attempt to lend rather than purchase consumption, the resulting decline in (nominal) income should result in a decline in prices, whereas your mechanism of getting rid of excess money balances by purchasing goods results in an increase in prices. Both methods result in a clearing of the excess money balances -- in once case, deflation in consumption prices raises the return for holding money, and in the other case, inflation lowers the real quantity of money. For some reason, you are only looking at the latter possibility.

[Max]

"Max: "You are confusing the question of what powers the government needs to have with what the government needs to actually do given its powers."

You lost me there, sorry."

I just mean that the fact that the government needs to have the power to demand more currency than it puts into circulation doesn't prove that it ever needs to do so, not even when starting a fiat currency from scratch. If the government is credible then people will begin saving immediately.

[Nick Rowe]

Gizzard:

"And will Nick or Scott please tell me exactly how these bond traders will reject the US dollar and why? When you get rid of something there has to be a market for it, so by definition if these guys are getting OUT of dollars, someone else is getting INTO dollars. They wont just burn their bonds, not these money grubbers."

This is a good place for me to explain to you the distinction between an equilibrium condition and an accounting identity.

1. "Acres of land sold = acres of land bought" That's an accounting identity. It's true by definition.

2. "Acres of land supplied = acres of land demanded". That's an equilibrium condition. It is not true by definition. Something (like the price of land most likely) has to adjust to make it true. It means "The acres of land that people *want to* sell = the acres of land that people *want to* buy.

There cannot be more land sold than land bought. That would violate the accounting identity 1.

But there can be more land supplied than land demanded. That would violate the equilibrium condition 2. What happens next? Most probably, the price of land will fall. And it will fall until the acres supplied decreases enough and/or the acres demanded increases enough that the two are equal at the new, lower, equilibrium price.

An excess supply of dollars means people *want to* sell more dollars than people *want to* buy. So the price of the dollar falls. I.e. the price of other things rises, in terms of dollars.

[Nick Rowe]

rsj: I'm getting a bit tired. This may not be a full answer.

I think you misunderstand me.

Simplest thought-experiment. Start in equilibrium. Then double everyone's money by helicopter. People don't want to hold that extra money. They want to spend that extra money and/or lend that extra money. Suppose they want to lend it. That does not mean their desired spending (on consumption) declines. It means they want to go on spending as before, *and* lend more than before. They want to lend out that stock of excess money. But they can't, of course, because everyone else is trying to do the same. So there's an excess supply of loans, so r drops, so now desired C and I increase relative to the old equilibrium levels.

[rsj]

OK, I understand, but a helicopter drop of money is fiscal policy.

What I mean by that, is that a helicopter drop increases nominal savings. When you pick up $100 bill that you find lying on the ground, your nominal savings just went up by $100. When you sell a $100 bond that you previously owned for $100 in deposits, your money holdings increase, but your nominal savings are unchanged. There is a difference.

So yes, if nominal savings are too high, I agree that everyone starts bidding up the prices of consumption goods. But they do so because their nominal savings are greater than the amount demanded, not because their money holdings is greater than the amount demanded. The quantity of money is just irrelevant in a modern economy.

However, if nominal savings are too low, and money holdings are too high, then people lend their excess money and do not consume. Still, they succeed in eliminating excess money holdings, but in this case prices fall, they don't increase.

It seems that you are either implicitly assuming that money = nominal savings, or that whenever money holdings are more than the level demanded, that savings must be more than the level demanded as well. But the two are completely different. I think you are omitting the case of households being satisfied with their money holdings, but not being satisfied with their nominal savings.

[K]

Scott, Nick: here's how I see the price of gold:

Above ground gold is currently worth about $7.5Tn. The global money supply is about $50Tn. The ratio of those values, 15%, is the expected value of future gold backed currency. So it's a market assessment of the probability that the world will revert to the gold standard. It's kind of backward looking since we would be unlikely to expect the gold standard if we hadn't had it before. But the value is really just a forward looking probability.

[wh10]

"Simplest thought-experiment. Start in equilibrium. Then double everyone's money by helicopter. People don't want to hold that extra money. They want to spend that extra money and/or lend that extra money. Suppose they want to lend it. That does not mean their desired spending (on consumption) declines. It means they want to go on spending as before, *and* lend more than before. They want to lend out that stock of excess money. But they can't, of course, because everyone else is trying to do the same. So there's an excess supply of loans, so r drops, so now desired C and I increase relative to the old equilibrium levels."

Nick, I can't give you a sophisticated argument, but some people with a sophisticated understanding of financial markets would say treasuries are very valuable as collateral, incredibly liquid, and result in credit creation several times over. The result is that they are effectively equivalent to currency, such that the economy is not meaningfully inhibited or slowed down in spending by holding treasuries. In other words, holding treasuries or not has little impact on desire to spend in the form of C or I. The Fed deliberately has 20 Primary Dealers making a deep, liquid market for treasuries, and they purchase them using repo markets, such that it's effectively not a zero-sum game.

Some of this logic and more is here at point 7. I recommend you read it (point 7), if you haven't already, and would request commentary if you care and/or have time :)- http://neweconomicperspectives.blogspot.com/2011/07/qe3-treasury-stylego-around-not-over.html

Maybe we could create a hypothetical scenario where treasuries aren't as equivalent to deposits, and "swapping them" would lead to meaningful price changes, holding all other variables constant, but my feeling is it would have to be quite extreme relative to most real world situations.

[wh10]

Wow- Sorry Nick. That is only tangentially relevant to the quote. You may ignore.

[Jacques René Giguère]

Nick: in Sept-Îles 1996, given the turnover in inventory now common, food publicly supplied ( food already at home is hoarded) may have dropped by 90% over 3 days. Fresf food,such as meat and vegetables was gone almsot within the first day.
Today, arrangement are in place (new pier in various cities along the North Shore and priorities to food and other essentials on the ferries from the South shore)to mitigate the problem. When Katrina hit in 2005 ( it wreaked havoc all along the East Coast), we got road cuts once again. Within a day , ships were in. Life went on without any oppotunity for controlled experiment in monetary theory.

As Mike Davis said in his "Late Victorian holocausts"

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

http://www.amazon.com/Late-Victorian-Holocausts-Famines-Making/dp/1859843824

A drought is a meteorological event. A famine is a political one.

[RebelEconomist]

Sorry to be snarky, but you don't need all this angels-on-pinheads economic theory to realise that the answer to the question in your first sentence is "no".

Assume that the central bank has monetary policy independence (which MMTers may deny but is the political reality in most serious countries at present), and further, can call on the government for more capital if necessary (albeit unlikely, as I shall explain). Then, whatever happens, the central bank can hold the value of its currency as required (ie typically in terms of a basket of consumption goods) by trading backing assets, as Mike Sproul says. If the government stopped taking the domestic central bank's money in tax payments, the demand for, and hence the value of, the currency may decline, but the central bank can fix that by selling assets (including not rolling over loans). Normally, a central bank will have slightly more assets than liabilities (not least because central banks tend to hold earning assets against zero or low-yielding liabilities), so the central bank should not need recapitalisation, but if it can call for more capital it can maintain the value of its currency in practically all circumstances. In fact, it is not hard to imagine circumstances in which the central bank can drive the value of its money to infinity, regardless of whether or not the government accepts its money in payment of taxes.

How about that for an atheoretical argument?

[Nick Rowe]

wh10. Yes, it's slightly OT, but the "are T-bills money?" question is also an important and interesting one, that I have tried to wrap my head around, and failed. Sometimes I do "blue sky" thinking, and imagine a world we are maybe trending towards, in which *all* of our wealth, from houses to cows to labour, is monetised, in some literal sense. "Money=wealth" is, generally, totally false. But people's desire to make their wealth as liquid as possible, and "Finance's" attempt to satisfy that desire, seems to create a trend in that direction over time. A chequing account at a bank effectively monetises the bank's assets. But why should Finance stop there? And 2008 was just a rather large hiccup in the road towards the eventual destiny of the monetisation of everything. ?

rsj: "What I mean by that, is that a helicopter drop increases nominal savings."

Re-phrase it. An *ongoing stream* of helicopters, dropping a *flow* of money, would be creating a *flow* of transfer payments (negative taxes), which, if people did nothing with, and just stuck under the mattress, would mean that the flow of saving had increased. S=Y-T-C, and if T falls and Y and C stay the same, means S increases.

But my thought-experiment is different. A single one-time flight of helicopters drops a *stock* of money at a point in time. Thereafter, until people start trying to get rid of that extra stock of money, S,C,Y, and T, are all the same as before.

K. Wow! Makes good logical sense. But I don't believe it!

Rebel: neat argument! I like it! A determined and independent central bank, if it has enough assets, can always control the value of its money to whatever it wants, by adjusting the quantity. So unless that central bank *wants* taxes to affect the value of money (and whyever should it want that?), they won't!

[Nick Rowe]

Rebel: I'm surprised Scott Sumner didn't make the same argument you just made. It's right up his street. It's a variant of an argument he has made before, I now realise.

[K]

Nick: "Wow! Makes good logical sense. But I don't believe it!"

That's cause you don't believe in the EMH!  Even when it provides a perfectly consistent explanation for a phenomenon, you choose to believe in irrational gold fetish or nostalgia. Anyways, it's a falsifiable theory: If the value of gold ever rises above the value of all money, I'll be wrong. Otherwise, you'll be wrong!

[Gizzard]

Nick

Thanks for the response.

I actually think your answer supports my general point. I understand that all that needs to happen is for there to be an overwhelming preponderance of bond sellers and not actual buyers for the price to plummet. The bond traders know this well too. Explain to me why they WOULD in fact act so irrationally (a huge run on the sell side) when in fact what they want are MORE dollars. I say this because everyone is calling for interest rates to rise in response to the bond vigilantes. A rise in interest rates means bond traders would get MORE dollars not less. This shows a demand for MORE dollars not less. The bond traders are not nor will they run FROM dollars, they want things that will give them more.

Guys who have hundreds of millions in bond wealth will not act to destroy that wealth, simply because their govt is quibbling over whether or not to pay granny a couple hundred more bucks a month. The numbers being talked about in the debt ceiling/budget debates are peanuts relative to the total wealth/output of the US econopmy.

Hyperinflation talk is absurdly irresponsible and you and any economist with ANY sense of decency should squelch it.

[Nick Rowe]

Gizzard: I don't fully understand your comment. But I'm going to make two stabs at it that might (or might not) be helpful.

There are two paradoxes:

1. Right now, a bit of fear of inflation, *by the people and not by the policymakers*, is exactly what the doctor would order. Keynesians like Paul Krugman understand this, and so do Monetarists. If people thought there would be more inflation, they would try to get rid of money and/or bonds, and try to buy more goods. This would increase the demand for output, and that would increase the demand for labour, so output and employment would rise. Which is what is so desperately needed right now, in the US especially. But if the policymakers fear inflation, and the people don't think there's going to be any, then the policymakers will tighten monetary and/or fiscal policy, which is exactly contrary to the doctor's orders. (Keynesians and Monetarists might disagree on whether it's fiscal or monetary policy that really matters most, but both would agree on this.)

It's weird. We want the people to fear inflation but we don't want the policymakers to fear inflation.

2. If *everybody* *tries* to get rid of land, or bonds, by selling it, they must fail. They can't find buyers, because everybody wants to sell, and nobody wants to buy. So the only thing that happens (roughly) is that the price of land, or bonds, drops.

But money is not like land, or bonds. Money is really weird. At least at times like now, when there's lots of output and labour that firms and unemployed workers want to sell but can't, because they can't find buyers.

*Even if* the amount of money that exists is fixed, like land, so people can't in total get rid of it (except by burning it, which they won't), if everybody *tries* to sell more money they will succeed in doing so! They can't, in total, get rid of money by selling it, if the total amount is fixed, because it just goes from one person's pocket to another person's pocket. But they *can* still sell more, even if everyone else is trying to sell more. They buy output and labour from the unemployed, who are only too happy to "buy" that money in exchange for their labour. But then those newly-employed workers turn right around and "sell" that same money again for all the goods and labour they want to buy, and so more unemployed get hired, and so on.

Money is weird. It's not like land or bonds or any other good. And understanding why it is weird is what could help get us out of this mess. And that's why "quasi-monetarists" like me are so fixated on money. And why we get so totally picky about understanding stuff like this properly.

Now do you see where I'm coming from?

[Jacques René Giguère]

Nick: As a quasi-monetarist, are you that different from the "eclectic Keynesians" that Pierre Fortin tried (and succeeded) to turn young uns like us into back in the 70's?

[Nick Rowe]

Jacques: I'm not totally familiar with Pierre Fortin's perspective. But in the big scheme of things, I would say I'm not that different. We are in the same ballpark. I am probably much more fixated on money as the medium of exchange than he is.

At grad skool the two profs that had the biggest influence on my macro perspective were David Laidler and Peter Howitt (though I had other good teachers like Michael Parkin and Joel Fried, and also Don Patinkin, briefly).

David is thought of as a monetarist, and he is. But he's a British monetarist. American monetarists used to jibe that British monetarists were really just sensible Keynesians. And I could never figure out if Peter Howitt (like Robert Clower) was really monetarist, Keynesian, or neither. It didn't seem to matter much.

I played with Austrian ideas by myself, was initially enamoured by the New Classical thinking of Lucas et al, then got very involved in the beginnings of New Keynesian thinking in the mid 1980's. Then veered back closer to my monetarist and Howitt/Clower and Yeager roots over the last couple of years. Bit of a mess, really.

Maybe, like Pierre Fortin, I'm eclectic enough to try to grab ideas from anywhere, and try to meld them into a vaguely coherent whole. Nearly every approach has *something* interesting to say. But fitting it all together into a coherent big picture, which is the very essence of macro, is the hardest part.

Doing macro is like looking at one of those Escher drawings of a staircase. Each flight of stairs may make sense. But does the whole picture add up?

[Min]

Nick Rowe: "if everybody *tries* to sell more money they will succeed in doing so! They can't, in total, get rid of money by selling it, if the total amount is fixed, because it just goes from one person's pocket to another person's pocket. But they *can* still sell more, even if everyone else is trying to sell more. They buy output and labour from the unemployed, who are only too happy to "buy" that money in exchange for their labour."

Nick, I am not picking on you, but from what you say, it seems that in your thinking **everybody** does not include the unemployed. From what I have seen, that viewpoint does not seem to be uncommon among economists.

[Nick Rowe]

Min: good point. You could correctly say that the unemployed are trying to "buy" money. But, and this is important, because it is why money is different, that does not mean the unemployed want to *hold* more money. They don't want to *hold* money. They want to buy money, then immediately sell it again, to buy food, clothes, furniture, etc.

[Lee Kelly]

Nick,

That is what I have been trying to explain to MMTers over on Sumner's blog for the last two days. Money is weird. They don't get it.

[Nick Rowe]

Lee: I don't blame them for not getting it. It's hard, even in a really simple pretend world. (And they don't like those simple pretend worlds). I keep thinking I've got it. Then I screw up, and make a mistake like the one Min spotted just above.

[rogue]

Nick: "If everybody *tries* to sell more money they will succeed in doing so! They can't, in total, get rid of money by selling it, if the total amount is fixed, because it just goes from one person's pocket to another person's pocket. But they *can* still sell more, even if everyone else is trying to sell more. They buy output and labour from the unemployed, who are only too happy to "buy" that money in exchange for their labour."

How about we turn your statements the whole way around.

How about: If everybody *tries* to BUY more money they won't succeed in doing so! They can't, in total, get rid of LABOUR AND OUTPUT by BUYING MONEY, even if money's total amount is increasing, because money just stays in a few people's pocket. But they *can* still BUY more labour and output, even if everyone else is trying to BUY more of it. They SELL MONEY TO the unemployed, who are only too happy to "SELL" LABOUR AND OUTPUT in exchange for their MONEY."

or when you say this:
"But, and this is important, because it is why money is different, that does not mean the unemployed want to *hold* more money. They don't want to *hold* money. They want to buy money, then immediately sell it again, to buy food, clothes, furniture, etc."

How about: Because it is why money is different, that does not mean the unemployed want to *hold* more money. They don't want to *hold* money. They want to SELL LABOUR AND OUTPUT, then immediately sell it again, to buy food, clothes, furniture, etc."

You get my point? I just turned your statements around, to look at it from the other side - the side not of people trying to get rid of money, but of people trying to hoard goods and services. I did not change the implications of your statement, but I changed the viewpoint. . But now it makes sense to think of increasing money as not necessarily resulting in inflation. It may just end up lowering people's value for goods and services, labour and output. Hence, it can lead to a deflation. It can lead to less transactions. It can lead to less velocity. Less people trying to get rid of goods and services, labour and output.

[Lee Kelly]

Nick,

You're right, of course. It's still tiresome.

[Min]

Nick Rowe: " this is important, because it is why money is different, that does not mean the unemployed want to *hold* more money. They don't want to *hold* money."

Thanks, Nick. :) That makes things clearer.

[Nick Rowe]

rogue: I'm not quite getting it.

You are trying to see what happens in reverse, when instead of trying to get rid of money everybody wants to get more money?

Try this:

If everybody wants to hold more money, they first try Plan A. In Plan A, everybody tries to buy more money. They all try to sell more labour and output, to buy more money. But since nobody wants to buy more labour and output, they fail. Plan A fails. So now they switch to Plan B. In Plan B, everyone tries to sell less money. They all try to buy less goods and labour. Plan B is a total success at the individual level, and a total failure at the aggregate level. Everybody succeeds in buying less goods. Nobody can stop you from buying less goods, if that's what you want to do. But at the same time as people are selling less money, they find they cannot buy as much money. Because everybody else is selling them less money than before.

We get a recession, in other words.

[rogue]

Nick, yes that's the situation I was getting at, without trying to squeeze the point into your earlier statements. Essentially, more money may not necessarily lead to inflation as an end. The future expectation of inflation could lead to less activity.

Or as we saw in 2008, actual increasing inflation (due to greater demand, money stock was increasing greatly too in 2008 because of the savings glut, the excess credit creation, and the derivatives growth) eventually led to a demand collapse, particularly in oil-intensive industries.

[rogue]

Actually, that's two different kinds of recession due to inflation - the demand collapse of 2008, where buyers stopped buying.

And the situation where sellers stop selling, due to future expectation of inflation. Perhaps we can lump gold owners as well as Vancouver homeowners in this.