The two James Tobins

Not many people know this, but there were actually two James Tobins.

The first James Tobin (pdf) said that there cannot be an excess supply of money, because if anyone did have an excess supply of money he would immediately run to the bank to get rid of it.

The second James Tobin said that people hold positive average stocks of money for the same reason anyone holds positive average stocks of any inventory: because it is too costly to keep running back and forth to the bank to get rid of inventory immediately and then get it back the moment before you need it again.

I might add that the bank is only one of a million places an individual could get rid of an excess supply of money. Any shop will take the stuff. Because it's money. An excess supply of money must mean an excess demand for something, but that something could be anything, and not necessarily anything the bank has for sale. But if you get rid of it anywhere other than at a bank, it just means someone else gets it. And if it keeps on circulating around the shops, rather than going back to the bank, that will change things like nominal income, and changing things like nominal income might also change people's minds about whether they really did want to get rid of it after all.

It all becomes a lot clearer when you realise there were two James Tobins, and they didn't agree. I think the second James Tobin was right.

(I sort of stole this from Laidler, but he's not responsible, of course.)

Comments

[JoeMac]

Shouldn't the second James Tobin's argument be less relevant now that we have e-banking? I think his argument should only work with currency, but currency is a tiny fraction of money held and spent.

[David Andolfatto]

Nick:

Not sure what point you are trying to make.

Money is an asset. At any point in time, *somebody* (or some agency) is in possession of it. As an asset, it must compete with other assets in the wealth porfolios of individuals. Because money facilitates trade, its market price (purchasing power) trades above its "fundamental" value (there is a liquidity premium associated with liquid assets). The rate of return on non-interest-bearing money (the inverse of the inflation rate) reflects the collective demand for money, vis-a-vis other goods and assets. Individuals are perfectly capable of "hoarding" (not spending) money; banks are not a prerequisite. Am not sure what anything of this has to do with "excess supply" or "excess demand." In the spirit of Laplace, we have no need for that hypothesis.

[Steve Roth]

Can you explain what it would mean for a person to hold a negative average stock of money? Or are we just talking about people not maintaining zero balances?

(Is it possible to hold a negative average stock of cows?)

And: I don't understand how the second Tobin "sed" contradicts or disproves the first.

I really don't understand how people holding two weeks' spending on hand instead of one for convenience implies any kind of hot-potato incentive.

> the bank is only one of a million places an individual could get rid of an excess supply of money

But the bank(s) that you (or the real sector) owe(s) money to is/are special, because when you give your money to them to pay off the loan, they burn it. (Along with your IOU).

If I pay off a loan to you I'm quite sure you won't burn the money. You'll just burn the IOU.

This raises an issue that troubles me: "bank liabilities as 'money.'" It's not the liabilities that are money, right? It's the credits that banks issue against the liabilities on their balance sheets. All money is credit, variously redeemable. Make a deposit at Target, you get a gift card/credit voucher. Make a deposit at your bank, you get an electronic credit voucher. If the bank offers to give you all their liabilities, would you think that they'd given you money?

Deposits may be counted/measured by looking at the liability side of the banks' balance sheets, but that doesn't mean that the liabilities are the money. ??

[Karsten Howes]

Hi Nick,

There's a big difference between exchanging money for goods and money for bonds. The first is an irreversible decision to exchange wealth for consumption. The second is a perfectly reversible asset allocation decision as it involves no change in wealth. If the Fed suddenly expands the quantity of fed funds by, say, 10%, consumption good prices, being sticky, will for the most part not jump at all. This therefore causes a large increase in the supply of the real balance, and so the nominal short rate, being the equilibrium opportunity cost of holding real balances, will fall. This market (fed funds, tbills) will equilibrate in seconds via a large quantity of reversible trades. Inflation expectations will likely move upwards resulting in a real rate drop that is at least as big as the nominal rate drop. Equilibrium in the rate market, resulting from tens or hundreds of billions of dollars of reversible trades, will occur before anybody has even had a chance to consume an extra hotdog. It is reversibility that guarantees that the money market (and all liquid capital markets) always equilibrates, and excess real balances are resolved via changes in the nominal rate before the goods market even transacts.

[Steve Roth]

Karsten: "There's a big difference between exchanging money for goods and money for bonds."

This is a point I've been wanting to raise here. Increased transaction volume in the market for real, newly produced goods inevitably results in increased production (and/or over time, price). In our 80% service economy (with many physical goods i.e. iphones produced on demand), if it isn't purchased it isn't produced. If it is purchased, it is.

Markets for financial assets are utterly different. Higher volume can be associated with either higher or lower prices, and increased production is incentivized not so much by volume (arguably the reverse is true), but by price (instantaneously per Karsten), in a rather loose causal relationship. And: there are almost zero inputs to production (relative to the "product's" value).

I just don't understand how supply/demand dynamics can be treated similarly for the cow and the bond markets. There's a reason that bond purchases aren't part of the equation of exchange. (Though I seem to remember Nick sort of sneaking them in there in the past...)

The also relates to stock supply and flow supply, which Clower once tried to theorize but nobody else really has that I know of. I've seen Nick toss those concepts out, with no theoretical underpinnings that I could find. Seems odd to me cause it seems crux-ial to understanding that central economic construct. Yes, "at the margin" maybe they're the same, but still...need some solid theory here.

Ah, here's Nick's stock-supply/flow-supply post:

http://worthwhile.typepad.com/worthwhile_canadian_initi/2011/07/the-peanut-theory-of-recessions.html

[Tom Brown]

Nick, sorry to slam you w/ all that, but my last one there: several of us want to know. Thanks!

http://pragcap.com/when-non-banking-experts-become-overnight-banking-experts/comment-page-1#comment-152897

[Tom Brown]

Shoot! ... above was for your previous post! Sorry. The link is still good though.

[Tom Brown]

Steve Roth, you write

"Deposits may be counted/measured by looking at the liability side of the banks' balance sheets, but that doesn't mean that the liabilities are the money. ??"

But does that matter too much? Isn't that really just two views of exactly the same thing. Now the loan the bank purchased with MOE... THAT is different than the MOE, and it too has two entities looking at it: to one it's an asset (the bank), and to the other it's a liability (the borrower). Same can be said of a reserve note: To the Fed that's a physical manifestation of its liability (unless they hold it), and to everyone else that holds it, it's a physical manifestation of money (an asset). Same dollar bill in both cases. Same can be said for an IOU I scribble out to you.

[Nick Edmonds]

Karsten: "There's a big difference between exchanging money for goods and money for bonds. The first is an irreversible decision to exchange wealth for consumption. The second is a perfectly reversible asset allocation decision as it involves no change in wealth."

Many of Nick's parables seem to rely rather critically on an assumption that expenditure on produced goods is a closer substitute for money holdings than are holdings of other forms of savings. I don't know if there any posts on here justifying that assumption - if there are and someone could point them out, I'd appreciate it.

[JKH]

The first James Tobin said that there cannot be an excess supply of money, because if anyone did have an excess supply of money he would immediately run to the bank to get rid of it.

The second James Tobin said that people hold positive average stocks of money for the same reason anyone holds positive average stocks of any inventory: because it is too costly to keep running back and forth to the bank to get rid of inventory immediately and then get it back the moment before you need it again.

.............................

Like others, I'm puzzled as to why you think Tobin 2 contradicts Tobin 1.

Also, can you provide the transaction you refer to in Tobin 1, and where in his money creation essay he refers to that. I've reread it a few times lately, and don't recall that sort of connection.

[Philippe]

The argument here seems to be that because it takes some time to go to the bank to deposit your savings, people will instead choose to spend their savings on consumption goods.

I have no idea why this argument is supposed to make sense.

[Philippe]

Steve Roth:

"that doesn't mean that the liabilities are the money. ??"

yeah the liabilities are the 'money'. The bank liability is a promise to pay base money on demand (or thereabouts, depending on the terms and conditions). Because that promise is backed up by bank capital, the central bank, and deposit insurance, it is usually a perfect substitute for base money, for non-banks.

[Matthew]

The excess demand for money both exists and does not exist, until you go as James Tobin and it becomes one or the other. Also, the bank is in Copenhagen.

[Alex Godofsky]

Philippe:

The argument here seems to be that because it takes some time to go to the bank to deposit your savings, people will instead choose to spend their savings on consumption goods.

I have no idea why this argument is supposed to make sense.

Not quite; it's that because it takes some time to go to the bank to withdraw money when you want to spend it, people will choose to carry some of it around with them.

In a modern context with electronic banking, I think "deposit money" = "buy financial assets".

[Nick Rowe]

David: naturally I am assuming that (at least some) prices are sticky, so we get excess demands and excess supplies of all sorts of goods. And a recession is when a lot of goods are in excess supply. Tobin (both Tobins) makes the same assumption, so that's not the issue.

The question is: can there be an excess supply of money?

The first Tobin says: yes, for central bank's fiat money, there can be an excess supply, if the central bank refuses to buy it back; but commercial banks promise to redeem their money for central bank money on demand, so there cannot be an excess supply of commercial bank money as long as they keep that promise, because if there were an excess supply people would immediately redeem it at the issuing bank.

But the second Tobin says that the "demand for money" means the average (over time) desired level of inventory of money, and in the Baumol-Tobin model, if people could adjust their inventory immediately and costlessly then the average desired inventory would be zero (it approaches zero in the limit as the fixed transactions costs of changing your stock approach zero). So if the assumptions required by the first Tobin were correct, the demand for money would be zero anyway, so the banks assumed by the first Tobin wouldn't exist..

Moreover, if there are n goods, including money, then there are n-1 markets in a monetary exchange economy, with each market accepting trades between money and one of the other n-1 goods. (Clower). If some of those n-1 markets are clearing, some are in excess demand, and some are in excess supply, what does an "excess demand/supply of money" really mean? In which market would we look to see if there was an excess demand/supply of money? Why does it have to be the market for that one good for which the issuing bank redeems its money? What about all the other n-2 markets? Especially since, as Tobin 2 notes, we need to assume there are transactions costs.

[Philippe]

Nick,

why would an 'excess supply of money' necessarily result in higher demand for goods and services? Couldn't it just result in higher asset prices?

[Nick Rowe]

Steve: "Can you explain what it would mean for a person to hold a negative average stock of money? Or are we just talking about people not maintaining zero balances?"

We are talking about people holding strictly positive rather than zero balances. (You can have a monetary system where people hold negative balances, but that's not the issue here).

"I really don't understand how people holding two weeks' spending on hand instead of one for convenience implies any kind of hot-potato incentive."

Suppose they want to keep one week's spending on hand, but they look in their pocket and discover they have two week's spending on hand. What happens next? Tobin1 says they always take the excess back to the issuing bank, which means it gets withdrawn from circulation. I say they might spend it. (See also my post on "What Steve Keen is maybe trying to say")

Karsten: "There's a big difference between exchanging money for goods and money for bonds. The first is an irreversible decision to exchange wealth for consumption. The second is a perfectly reversible asset allocation decision as it involves no change in wealth."

Fair point. But there are also real investment goods. And consumer durables are really investment goods. You could go to the bank and buy a term deposit, or you could go to the bike store and invest in a bike. Which gives the better return?

But in the rest of your post, you are assuming that an "excess supply of money" means the same thing as "an excess demand for bonds". That's what many textbooks say, but I don't see it at all. With n goods, including money, there are n-1 markets in which money is traded. Why should we pick on just one of those markets, the "bond" market, and say an excess supply of money = an excess demand for bonds? Why not an excess supply of money = an excess demand for bikes?

[Philippe]

"Suppose they want to keep one week's spending on hand, but they look in their pocket and discover they have two week's spending on hand. What happens next?... I say they might spend it."

They might. But most people decide how much they want to spend based on their average income or net wealth, taking into account their expenses and liabilities such as mortgages and other debts.

If you replace someone's assets with money, they won't necessarily just spend the money on goods. They will generally only increase their spending overall if they think their income or net wealth has increased, and so they can afford to spend more.

They might not even spend more, even if they think their income or net wealth has increased - they might just think "great, now I can save more for my children's education/retirement/healthcare!", etc.

[Ron Ronson]

"Tobin1 says they always take the excess back to the issuing bank, which means it gets withdrawn from circulation"

Doesn't Tobin1 say that they will take any excess cash back to the bank and if they are happy to keep it in a deposit account the bank may increase lending. The economy will tend towards an equilibrium where cash held, bank deposits, loans , and interests on loans and deposits will all be at stable levels.

Its not clear why Tobin2 contradicts this - He just provides some theory as to how the individual will decide whether he has too much cash on hand or not.

[Ritwik]

I suppose many have said this, but don't see how Tobin 2 contradicts Tobin 1 at all. Tobin 1 does not require the process of reflux to be costless, simply that at some point, the cost benefit calculation turns in the favour of reflux. What that point is, is given by Tobin 2.

And if you assume that the economy was in equilibrium at T0, everyone's Baumol-Tobin demand is met and any fresh money creation does not meet the inventory calculation and will be extinguished.

[Tom Brown]

1st Tobin: "because if anyone did have an excess supply of money he would immediately run to the bank to get rid of it."

I'm with JoeMac on this one. "get rid of it?" What? The money has not been gotten rid of. The implication here is that this person was holding physical cash. But when they exchange that cash for bank deposits they still have the same amount of money. None has been "gotten rid of." And presumable they have a bank card they can make purchases with?

Can we translate this into a world of bank cards please... ONLY bank cards. The assumption of cash existing runs throughout this piece. I can't believe you'd write this if it depended on the existence of physical cash, so I'm going to assume I'm missing something here. My bank deposit IS my money stock.

OK, so here it goes, a cash free world, let's translate: It sounds like you're talking about people wanting to trade their bank deposits for other stuff... all sorts of other stuff... and you're saying that in Tobin 1, "people run to the bank and get rid of these deposits" ... well the ONLY ways that can happen in our cashless society is:

1. They trade their deposits for Tsy bonds... in which case Tsy spends the proceeds and their deposits are inflated again. So that can't be it.

2. They buy assets from the CB. If the CB were selling (which the Fed is not). OK, that works until the CB has no more assets.

3. They repay bank loans (and their deposits are debited)

4. They buy other stuff the bank is selling: CDs, bonds, bank equity, etc. (and their deposits are debited)

I don't see where the aggregate non-bank private sector gets rid of its bank deposits otherwise in this cashless world.

So Tobin 2 says they won't get rid of it all by these methods and they'll keep a stock of it. OK, well, that's pretty clear. I guess I'd agree with that too.

Let's make some further assumptions to try and clarify a bit here:

A. Tsy spends every $ it gets immediately. That cuts off escape route 1. above.

B. CB has no assets (T-bonds, MBS, etc) for sale. Just its CB deposits (which are not available to the public). That cuts off escape route 2.

That leaves Tobin 1 saying the aggregate, cashless non-bank private sector always immediately repays its loans or buys other bank products. Yes, that sounds 100% wrong. So Tobin 2 does sound better if it says that the private sector does keep some deposits on hand.

Is that what you're saying translated into a cashless society? Or is this whole argument about cash? If it is about cash it makes no sense to me. I had a much longer post written out about why it makes no sense, but I thought I'd spare you that monstrosity. :) This one is plenty bad enough.

[Ron Ronson]

The way that I see money being extinguished in Tobin1 is not because people bring cash back to the bank but because if they do not do this (and hold the cash) then banks may have to stop renewing loans and the money supply will shrink (unless the CB increases available reserves like they would do under interest rate targeting).

So start from an equilibrium where people are happy to hold banks deposits that match bank loans. Something changes and they decide to hold cash instead. Bank deposits decline as people convert them into cash. To supply this cash banks will have to reduce reserves to get cash from CB. They then will have too few reserves to back their current level of loans so they will have to stop renewing loans and this will reduce the money supply. They will probably increase interest rates in this process. Eventually a new equilibrium will be reached where demand to hold cash, bank deposits and loans will be stable. In the long term prices will need to adjust.

[Philippe]

[Sorry Philippe, but I had to edit that out, because it might confuse our spam filter even further! NR.]

[Philippe]

Tom,

"I don't see where the aggregate non-bank private sector gets rid of its bank deposits otherwise in this cashless world"

Bank deposits aren't just checking accounts. They're also savings accounts, which falls into your 4: 'They buy other stuff the bank is selling'.

If everyone wants to buy either checking accounts or savings accounts, and there's not enough going out the other side into loans, then you have a problem. The problem is not that too many people end up holding money, because overall the saving impulse will lead to lower incomes and thus lower saving in aggregate. The problem is that too many people don't want to spend or invest, and instead want other, low-risk people to spend or invest on their behalf (so the savers can get a low-risk asset with some return).

[Tom Brown]

Philippe, good point about savings accounts. Buying them won't pay down existing loans, and in fact the existing loan base can expand simultaneously (imagine every ones takes out loans, buys stuff, and the sellers take the proceeds and buy savings deposits: all done w/in seconds on their smart phones of course :). In my world there's no M0 but there is M1 (checking), M2 (savings), MZM, M3, M4, M4- and L (using the US definitions):

http://en.wikipedia.org/wiki/Money_supply#United_States

So it depends where you want to draw the line? M2 and below is money? Then translating this cashless world into Tobin 1, then that's got to go!... some into M3 and above. Right? And the rest into bank equity and bonds. Tobin 1 says it can't exist, so it has to go somewhere... even though the stock of outstanding bank loans can be expanding.

[Tom Brown]

Ron Ronson, I still refuse to believe that Nick would write an article all about cash here and not tell us explicitly (thus my request he translate it into a cashless world, since it makes no sense to me otherwise... lots of implicit assumptions floating around that money is really only cash and some such non-sense, but we KNOW Nick doesn't believe that, right?).

You write: "The way that I see money being extinguished in Tobin1 is not because people bring cash back to the bank but because if they do not do this (and hold the cash) then banks may have to stop renewing loans and the money supply will shrink (unless the CB increases available reserves like they would do under interest rate targeting)."

So that's a direct contradiction of the way Nick describes Tobin 1. Are you saying instead they withdraw ALL their deposits as cash? Demand as well as time, or just demand? Yes, why not assume inflation targeting long term and overnight rate targeting between adjustments of the overnight rate. That is how the system has been working. Doesn't this just drive up the interest rates a bit? Instead of cheap deposits, the banks must now replace them with slightly more expensive reserve borrowings... but the banks in aggregate don't have to borrow more than the reserve requirement (from the CB). Nothing here inherently prevents them from continuing to make loans.

"So start from an equilibrium where people are happy to hold banks deposits that match bank loans. Something changes and they decide to hold cash instead. Bank deposits decline as people convert them into cash. To supply this cash banks will have to reduce reserves to get cash from CB. They then will have too few reserves to back their current level of loans so they will have to stop renewing loans and this will reduce the money supply."

First of all reserve requirements are on demand deposits, NOT loans, so if all demand deposits are withdrawn as cash, they don't have a formal reserve requirement anymore. But you're right! A lot of reserves will disappear when it goes out the door as cash and the banks will have to borrow that... and thus raise rates to preserve their spreads. But the CB, committed to inflation targeting steps in and provides all that. They may even adjust the overnight rate in light of the banks' rate changes.

[Tom Brown]

Ron, Shoot! Spam filter got me I think..., I'll make this brief. If your assumption is that Tobin 1 means people withdraw demand deposits as cash, then this doesn't stop loans, especially under inflation targeting like you assume: Banks end up replacing cheap demand deposit liabilities on their balance sheets with more expensive reserve borrowing from the CB, and thus may need to raise rates to preserve spreads, but the CB is inflation targeting long term (and overnight rate targeting short term) so they'll probably adjust overnight rates in response. None of that stops lending. Also, the banks don't actually need to keep many reserves since reserve requirements are requirements on demand deposits NOT loans, thus their requirement goes to $0 when the cash walks out the door.

[Tom Brown]

Nick, am I in the permanent spam filter now? )c:

[Max]

I don't see a contradiction in the two statements, provided the "excess supply" is expected to last long enough so that transaction costs are worth paying.

[Nick Rowe]

Max: but if there are *any* transactions costs of getting rid of it at the bank, mightn't it sometimes be cheaper to get rid of it at the shops instead? After all, absent those transactions costs, the non-money stuff the shops sell are equally valuable to you, on the margin, as the non-money stuff the bank sells.

Tom: I fished two of yours out the spam filter. Sorry about that. Nearly all mine go there.

Forget about the physical properties of money. A demand deposit at a bank is just the same as the bank issuing its own paper notes, with one exception: A paper note is just like a bearer bond, where physical possession is taken as prime facie evidence of ownership; a demand deposit is just like a bond where you have to notify the issuer when you sell it, so the issuer can record the change of ownership in a ledger.

[Nick Edmonds]

Nick,

"With n goods, including money, there are n-1 markets in which money is traded. Why should we pick on just one of those markets, the "bond" market, and say an excess supply of money = an excess demand for bonds? Why not an excess supply of money = an excess demand for bikes?"

It might depend on where you assume the excess supply of money comes from. If it comes from a big unfunded tax cut, then you might reasonably assume that both demand for bikes and demand for bonds will increase. If it comes from a shortage of bonds (say the government fixes the issue price of new debt and suddenly reduces new issuance of debt to replace maturing bonds), then you might be better equating it with an excess demand for bonds.

[Mike Sproul]

Think of paper bank notes (no cost of printing and handling) that promise 1 oz. of silver in 1 year. If R=5%, those notes might start the year worth .95 oz and rise to 1.00 oz. by year-end. On the other hand, if the monetary convenience yield of those notes is 5%/year, they might be worth 1.00 oz. all year long, thus paying no interest to the holder. Let's say we're in the situation where they are worth 1.00 oz. all year (not sure why, still thinking it through). Then the central bank multiplies the quantity of notes by 10 times. The lowest those notes can fall is to .95 oz, and at that point they bear 5% interest and people are willing to hold huge amounts of them as an investment. They don't have to reflux to the bank, but neither will they do the hot potato.

[Ron Ronson]

Tom,

I think the key to understanding how the banking system keeps supply and demand for money inline is that they operate on FRB principals.

so: assume a fixed base and no reserve requirements.

When customers wish to hold smaller balances in their bank accounts then banks will have to hold more reserves to meet clearing requirement for the greater number of transactions that take place as customer attempt to reduce their balances. Bigger reserves mean less lending and a smaller money supply. This is how in my view the suppy of money will change to meet demand (and minimize the need for price changes) under FRB and with a fixed base and no CB activity.

The role of cash v bank accts is relevant here though. The more money is held in cash rather than deposits the less is available for FRB lending and so the smaller the money supply. As the demand for cash increases I see no automatic adjustment mechanism in a world where the CB control the base and notes issue. In a free banking system where banks issue their own notes there would really be no difference between cash and deposits (from the bank lending POV).

[Tom Brown]

Ron, you write "assume a fixed base" ... ah... yeah, but originally your wrote to assume inflation targeting. Even though my cashless thing is a bit fanciful, I think the fact that it preserves inflation targeting (or something like it) and the facilitating role of the CB in a fiat system (to clear payments [via overdrafts if necessary] and to keep those ATMs from running dry ... and to act as lender of last resort!) is a LOT closer to the reality of our situation than a "fixed base" assumption. Fixed base means we lose ALL the benefits of having a fiat currency in the first place! FRB doesn't really apply when fiat is used like it's supposed to be used.

... but anyway, getting back to your point, assuming this "fixed base" now:

Actually if ALL deposits are withdrawn in cash there are NO more clearing requirements... no electronic CB deposits need to be credited or debited by the CB behind the scenes to clear payments any longer!

But yes, if you assume a fixed base... and not bank deposits (all withdrawn as cash), then this is a MAJOR problem for banks expanding the money stock via lending. Fortunately that's NOTHING like the system in countries with true fiat money.

Wait... I can't tell now in your final paragraph what you're saying. I thought you said in your original post that Tobin 1 is saying: "take all the deposits out in cash?"

[Ron Ronson]

Tom

Re: 'I can't tell now in your final paragraph what you're saying. I thought you said in your original post that Tobin 1 is saying: "take all the deposits out in cash?"'

I was trying to reconcile what Nick said about all excess reserves being returned to banks with my reading of the Tobin paper. At first I could only think of cash v deposits as causing the money supply to be elastic (with no CB involvement anyway), but then I thought about how FRB might change things and realized that FRB has an inherent mechanism for causing the money supply to change to meet demand.

The Tobin paper is not very clear on these issues so it is just speculation on my part what Tobin had in mind. But he clearly has the view that banks ability to create new money is determined by the public's demand to hold it.

[Tom Brown]

Ah, OK. Well I haven't read the Tobin paper, so perhaps I should do that! One thing you wrote though:

"all excess reserves being returned to banks"

I don't know about Canada, but in the US it's quite impossible for reserves to be anywhere except at banks. Reserves, excess or otherwise, are DEFINED as BANK HELD base money (vault cash and Fed deposits). So to speak of reserves leaving the banking system or returning to it makes no sense. Yes reserve levels can rise and fall, but reserves, by definition, can't be anywhere else. Base money can, but reserves can't. Sorry!... but that one always bothers me when I see it.

[Ron Ronson]

"all excess reserves being returned to banks"

Ugh - Typo: Should have been "all excess MONEY being returned to banks"

[Tom Brown]

Nick, O/T, but I asked Sumner this in his recent explanation of the HPE:

http://www.themoneyillusion.com/?p=23314#comment-272021

What's your take? Any expectations benefit to having the Fed request a bunch of paper reserve notes be printed up? What about if they increase the amount every month by 20% until they hit their NGDP target?

What about coins? Coins are legally (base!) money from the time they leave the Mint. They could Mint a special new $1000 coin (super cheaply), the Fed could buy it, and to ensure that the seignoirage didn't end up in Tsy's Fed deposit, perhaps congress could authorize a new Fed deposit be created ONLY for the Mint, with which it could ONLY use to pay for future raw materials and employee salaries, etc. So basically the Mint's Fed deposit could ONLY decline very very slowly.

So now you'd have all this new base money sitting at the Fed... all actual money, but would it cause a benefit? The Fed would announce all this and be MORE than willing to sell the new $1000 coins to the banks should there be a demand.

Now lets say that the Fed announces it will increase $1000 coin purchases by 20% each month until they hit their NGDP target. Any difference? (Say they start off with $85B a month).

Perhaps the Fed could even issue a new kind of note "backed" by these $1000 coins (redeemable in $1000 coins) in the Fed's vaults... these notes would be the usual denominations: $1, $5, $10, $20, etc, so there might actually be a demand for them. Any difference? How is this different that Scott's example where a new vast gold reserve is discovered, but it will take two years to dig it out of the ground, AND gold is the MOA? (That's his case 4 or 5 I think).

[Tom Brown]

... I think I answered my own last question above: My final example above would be similar (but not exactly analogous) to Scott's case 4 (gold = MOA, and a lot more gold is discovered) except that the central bank (CB) is the gold's discoverer instead of a "company" like Scott writes. In addition, the CB does not have to remit this gold to Tsy, but it does announce the discovery to the world and likewise puts it up for public sale. But what would people buy the gold with? More gold? Or could they use other MOE? Bank deposits? Tsy debt? MBS?

[Nathanael]

You left out the obvious -- we can put the money under our mattresses. There are things equivalent to that: things which effectively take money out of circulation.

Frankly, the rate at which people do this increases whenever there's instability and crime in the banking system. The rate at which *corporations* do this increases under the same circumstances. The rate at which BANKS do this increases under the same circumstances...

There are also some often-overlooked *distributional* effects which can take money out of circulation in the "real economy". If money keeps circling around a gambling game between the Forbes 400, it may be effectively gone from the real economy. I strongly suspect that this actually happened in the US.