Banks are special because the medium of exchange is special

If we used cows as media of exchange (if we bought and sold everything else in exchange for cows), would you say that dairy farming is a special industry that is macroeconomically important?

I would. Because if we used cows as media of exchange, then what happened on dairy farms would affect the supply of media of exchange, and I think the supply of media of exchange is macroeconomically important in explaining short-run recessions.

And if you wouldn't say that, why would you say that banks are special and macroeconomically important in the real world?

Banks can create the medium of exchange "at the stroke of a pen", and lend it out for a monthly fee. But a talented artist can create valuable drawings at the stroke of a pen, and lend those drawings out for a monthly fee to those who want to hang them on their walls. If the medium of exchange is not special, or macroeconomically important, then banks are no more special or macroeconomically important than that talented artist.

Ask yourself these two questions:

1. Is the medium of exchange a special good, that is unlike all other goods, in a way that is macroeconomically important?

2. Are commercial banks special firms, that are unlike other firms, and unlike other financial intermediaries, in a way that is macroeconomically important?

I think those two questions are very closely related. It would be hard to answer "no" to the first and "yes" to the second. That's because:

Banks are financial intermediaries whose liabilities are used as media of exchange. If the medium of exchange is not a special good, and not a macroeconomically important good, then banks cannot be special or macroeconomically important either. Banks would be no more special or macroeconomically important than mutual funds.

My own view:

In long run equilibrium, banks are no different from other financial intermediaries. They borrow and lend. They make life easier for ultimate borrowers and ultimate lenders to get together, just like shopkeepers. And the main task of monetary policy is to keep the economy in that long run equilibrium by preventing monetary instability by preventing banks lending either more or less money for some people to invest or consume than other people want to save.

But in the short run disequilibrium, the quantity of medium of exchange matters a lot:

1. Every market in a monetary exchange economy is a market for the medium of exchange. This means that the medium of exchange is special because there are two ways an individual can get more medium of exchange: he can buy more medium of exchange (by selling more of other goods); or he can sell less medium of exchange (by buying less of other goods).

2. An excess demand for the medium of exchange prevents people making the mutually advantageous exchanges that would be possible if barter were easy so we did not need monetary exchange. That's what we call a "recession". If I could sell my goods, I would want to buy your goods; and if you could sell your goods, you would want to buy my goods; but neither of us will part with our money because we don't know whether that money will return to us, and it's too difficult to organise a  swap.

3. Unlike other goods, people will accept more medium of exchange even if they do not plan to hold more medium of exchange and even if they think nobody else plans to hold more medium of exchange. We accept money in exchange for other goods only because we know that others in turn will accept it from us. The quantity of medium of exchange is supply-determined in a way that is not true of other goods. The talented artist can create drawings at the stroke of a pen, but cannot rent them out unless he can find someone who wants to hang that drawing on his wall. An excess supply of drawings immediately refluxes to the artist. The talented artist cannot create an excess supply of his product; banks (in aggregate) can. I borrow the drawings because I want to hold the drawings; I borrow the money because I want to buy something else.

So I think that commercial banks are macroeconomically important because, and only because, they influence the supply of media of exchange. Otherwise, if that were not important, they would be just like any other financial intermediary.

But tell me: is the medium of exchange a special good, and a macroeconomically important good, and a good whose supply matters, in your macroeconomic view of the world? If not, what's the big deal about banks?

And if we did use cows as money, would you forget about financial intermediaries, and start writing about dairy farms instead? Or if we used the drawings of talented artists as money, would you forget about financial intermediaries and start writing about artists instead?

For Steve Roth.

Comments

[Vladimir]

Nick--from a regulatory standpoint doesn't this tend to reinforce the view that banks are like the utilities and should be regulated in a similar fashion ? A second question: Has anyone tried to apply principal-agent theory to central banking? As per your post, I think it's correct to say that banks are the means through which monetary policy is conducted.

[Nick Rowe]

Vladimir:

On the regulation question: hmmm, dunno. I think it does mean that the theory of the regulation of banks should be different from other industries. And central banks already "regulate" the banking system just by being the central bank. Commercial banks promise to make their money redeemable on demand into central bank money at a fixed exchange rate, and not vice versa, which is what gives the central bank its leverage over commercial banks.

Nothing immediately comes to mind on applying principal-agent theory to central banking, but that probably just means my mind is failing again. You can think of inflation targeting (or NGDP targeting) as the government, as principal, telling its agent the central bank what to target, and holding its agent accountable for hitting that target.

Commercial banks are *one* of the channels through which monetary policy can be conducted. (Some of us old guys still use central bank currency!)

[Frank Restly]

Nick,

"And the main task of monetary policy is to keep the economy in that long run equilibrium by preventing monetary instability by preventing banks lending either more or less money for some people to invest or consume than other people want to save."

Except for that to happen, you would need to identify the saver before you identify the investor / consumer. With banking, the reverse happens - a bank finds a willing borrower, makes a loan, and then tries to find a buyer for the loan (saver).

[Lee Kelly]

This post--it's like you read my mind.

[Nick Rowe]

Frank: "Except for that to happen, you would need to identify the saver before you identify the investor / consumer."

You would need to measure *both* desired saving and desired investment across the population in real time. Which can't be done, of course, which is why macroeconomists instead recommend policies like inflation targeting or NGDP targeting instead.

And BTW, any individual commercial bank that created media of exchange and lent it out before figuring out what would happen when that loan gets spent and the cheque gets deposited at another bank would be a foolish bank. Likewise, an individual commercial bank that accepted a deposit before figuring out if it had potential customers for loans wouldn't be very wise. My bank both advertises for new borrowers and advertises for new depositors. Simultaneity. Commercial banks care about the spread, which means *both* their lending and borrowing rates and opportunities. (Central banks aren't like that.)

[Phil Koop]

OK - I have read this post, and clicked over to read the Steve Roth post too, but I am overlooking some critical subtext. Everything you say here seems completely uncontroversial. I remain convinced that the MOE is special and that banks are, in some sense, special as a consequence.

I also disagree with Steve Roth: I don't think that a detailed understanding of the mechanics of banking is required for the study of macroeconomics (though I personally find those details quite interesting.) But I don't understand why Steve Roth should change his mind because of this post. How does it bear upon his claim? Why wouldn't he just say that in a world in which cows are money, the incentives of dairy farmers are "orthogonal" to those of cow borrowers? Whatever that means.

[Phil Koop]

But wait ... maybe cows really are money. They certainly seem to be special:

"We examine the returns from owning cows and buffaloes in rural India. We estimate that when valuing labor at market wages, households earn large, negative average returns from holding cows and buffaloes, at negative 64% and negative 39% respectively."

[Nick Rowe]

Lee: thanks! Yep, because you already believe the MOE is special and important.

Phil: The subtext: some bloggers criticise Market Monetarists (and people like Paul Krugman who aren't MMs) for "ignoring" banks, or for not understanding that banks are "special" and macroeconomically important. In this post, I am turning that accusation around: are those who make that accusation ignoring the supply of the MOE, or failing to understand why the MOE is "special" and macroeconomically important?

I kept that subtext implicit in this post, because I am more interested in trying to present my own views clearly (hard enough to do) than in getting into arguments about who said what or didn't say something else.

(There's a side-argument going on as well, because not all MMs think alike on this question. Scott Sumner for example emphasises the MOA aspect of money, and it is central bank money that is the MOA. I put more emphasis on MOE, and both central and commercial bank moneys are MOE. I ignored that side-argument in this post.)

I don't find Steve Roth's story of why banks are special very convincing (though I'm not sure I understand what he is trying to say). And even if it were convincing, and I understood it, it doesn't go to the root of the question of what it is that gives banks those properties he says they have. My story goes to the root of it, so he might find my story to be the underlying explanation of his story. (Of course, a still deeper explanation would explain why people use money instead of barter, etc., and why they use bank money instead of cows. But we have to start somewhere, or any blog post would be very long.)

"Everything you say here seems completely uncontroversial."

It seems that way to me too. But I think I am contradicting the Tobin view of banks. The idea that the banking system in aggregate can create an excess supply of money is highly controversial nowadays. And the role of the supply and demand for MOE in understanding recessions is controversial. My guess is that some will find it controversial.

[W. Peden]

Curse of spellcheck: I assume you mean the dairy industry, not the diary industry.

(If we uses diaries as the medium of exchange, this would be good for the diary industry and terrible for privacy, unless we left them blank or were untruthful.)

[Frank Restly]

Nick,

"And BTW, any individual commercial bank that created media of exchange and lent it out before figuring out what would happen when that loan gets spent and the cheque gets deposited at another bank would be a foolish bank."

"Likewise, an individual commercial bank that accepted a deposit before figuring out if it had potential customers for loans wouldn't be very wise."

Banking system = no fools allowed? Wasn't that what quantitative easing was all about - federal reserve forces banks to accept deposits without figuring out if those banks had potential customers for loans?

Here are the results of the U. S. federal reserve's quantitative easing attempts:

http://www.federalreserve.gov/releases/z1/Current/z1.pdf

Page 73 - Private Depository Institutions

Assets - $15.24 Trillion
Liabilities - $15.52 Trillion

"Commercial banks care about the spread, which means *both* their lending and borrowing rates and opportunities."

The spread is relevant for cash flow analysis of a bank. The spread does not factor in balance sheet analysis of a bank where deposits are treated as liabilities of that bank. Obviously things like deposit insurance mitigate the potential for bank runs when the solvency of a bank is in question.

[Tom Brown]

Nick, I take you may have read Krugman and Cullen on this?:

http://pragcap.com/banks-are-special-the-ins-outs-of-money

(Krugman link at the top)

[Nick Rowe]

Phil @12.42: Nah, maybe cows are just Veblenesque status-goods, like having fancy artwork on your walls!

W Peden: Curses! Fixed. Thanks.

[W. Peden]

"The idea that the banking system in aggregate can create an excess supply of money is highly controversial nowadays."

Is that supposed to be true of the commercial banking system AND the central bank, or just the commercial banks?

[Nick Rowe]

Tom: No I hadn't, but I have now. My own opposition to the Tobinesque view has been influenced more by monetarists like David Laidler and Leland Yeager (and I disagree with other MM's like David Glasner who sides with Tobin against Yeager, and sometimes I agree with Paul Krugman and sometimes I disagree). It would take too long to go fisking through these posts and explain precisely where I agree and disagree with everyone, which depends so much on how you interpret them. I think it's more productive if I just try to explain my own views clearly.

[Nick Rowe]

W Peden: Some (e.g. the Neo-Wicksellian orthodoxy) say neither central nor commercial banks can create an excess supply of money. Others say central banks can and commercial banks can't. I say that no *individual* commercial bank can create an excess supply of *its* money, but that both the central bank and the commercial banks (in aggregate) can. This, to my mind, is the old textbook story, as defended by Yeager against Tobin. The central bank can create a hot potato of central bank money among commercial banks, which in turn creates a hot potato of commercial bank money among the public. At one point Tobin seemed to argue for the Law of Reflux even for central banks, then he relaxed that and said the Law of Reflux applied to commercial bank money only.

[Tom Brown]

Nick, nice article. This really gets me thinking. On first read, I'm inclined to agree with you that the MOE aspect of money is the important part here... and thus this DOES make banks special.

I look at it this way: banks in aggregate buy EVERYTHING they obtain from the non-bank private sector by creating MOE and crediting bank deposits with this MOE: time/work from employees (salaries), shareholder dividends, assets (2ndary market T-bonds, etc), electric bill, interest to depositors, and loan agreements from borrowers (i.e. they loan money to borrowers). Likewise, all payments made to banks (interest, points, fines, fees, service charges) by the non-bank private sector are accomplished by allowing banks to erase (destroy) this liability the banks previously created (i.e. they debit bank deposits). Well, I lied of course, because these things can be accomplished with paper reserve notes and coins too... but I doubt that figures in prominently: if it's not costing us to keep our deposits at banks, then why hold any more cash than is convenient? It's entirely up to us.

Fed deposits are used by the aggregated banks only in transactions w/ other Fed deposit holders (e.g. Tsy) or the Fed itself. The Fed itself behaves exactly as the aggregated banks do, only at the next layer out: buying all by crediting Fed deposits and selling all by erasing them: only the Fed doesn't have to worry about solvency as much (but then again, neither do the aggregated banks... except in very unusual circumstances... like when the whole industry is in trouble)

[Peter N]

Cows are indeed money among the Maasai in Africa.

It seems to me that there are a number of issues involved with the role of banks. It may be better to deal with each separately. They are

The extent of net credit in the economy and its effects.

The role of banks in creating money, both on their own and in concert non-bank financial companies.

The model of the traditional bank used by most economists. Open at 10, close at three and profit from the spread between interest paid and interest received. No funny stuff. Glass-Steagall rules.

How much it matters the the bank of Krugman bears very little resemblance to an integrated financial institution like Bank of America. Morgan Stanley has retail banking as a vestigial organ -an appendix to the giant colon of their wealth management division. There's no way the First Bank of Krugmanland could lose $3 billion in a few weeks speculating in derivatives.

The financial sector has doubled in size since 2000. The purpose of a money center bank is to make money for its executives and stockholders (arguably in that order). Such a bank benefits from enormous information asymmetries (like manipulating the Libor rate) and views financial complications as opportunities to exact rents.

Morgan Stanley is currently defendant in 42 major law suits all with claims in excess of $100 million. It has also been notified of intent by customers to audit real estate securitization deals with a total value of (last I looked) $58 billion. Among other things, the GSEs want to be made whole for losses from loans not being of the quality required by the loan agreements.

It would seem to me that ignorable intermediation requires an alignment of interest between clients and the bank. Otherwise the invisible hand becomes the thumb on the scale.

If you want to see how far out of alignment interests can get and what the results of this are, take a look at Spain.

[W. Peden]

Nick,

That's very helpful.

"I say that no *individual* commercial bank can create an excess supply of *its* money, but that both the central bank and the commercial banks (in aggregate) can."

If there were no commercial banks (but still OMOs with primary dealers) then presumably you could ALSO possibly have an excess supply of money, i.e. the central bank could exchange more cash for securities than the non-bank public wanted to hold. Of course, there may be some uninteresting institutional issues with this (I don't know how non-bank primary dealers sell and buy from central banks and I know of no reason to care) but in principle, once the central bank is involved in an economy, there can be an excess of central bank money, even without commercial banks. So the central bank is special.

I think I'm (a) starting to understand what a lot of your posts over the past few years were saying and (b) starting to see why free banking of the sort that George Selgin and Lawrence White propose has such stabilising properties.

[Nick Rowe]

Tom: I *think* I agree with that.

When the Bank of Montreal buys a computer, it creates money. When it sells that same computer again it destroys money. But the BMO buys and sells far more IOUs than it buys and sells computers. And the same is true if I replace "Bank of Montreal" in that sentence with "Bank of Canada". A successful counterfeiter who printed money to finance his computer habit is as special and macroeconomically important as a bank.

Commercial banks, both individually and in aggregate, *do* have to worry about solvency. That's because they promise to redeem their liabilities in central bank money at a fixed exchange rate, and if their assets go bad they won't be able to do that. A central bank that promises to redeem its money for gold or foreign currency at a fixed exchange rate might also have to worry about solvency. A central bank that targeted CPI inflation (which means indirectly promising to redeem its money at a sliding exchange rate against the CPI basket of goods) might also worry about being able to keep that promise, if its assets went bad and there were a decline in demand for its money. But central banks, unlike commercial banks, earn monopoly profits, and the present value of those monopoly profits is a far more important asset than any asset that appears on their books.

[Nick Rowe]

W Peden: "If there were no commercial banks (but still OMOs with primary dealers) then presumably you could ALSO possibly have an excess supply of money, i.e. the central bank could exchange more cash for securities than the non-bank public wanted to hold."

Yes. It certainly could create an excess supply for money (or an excess demand). Just because the market for securities clears doesn't mean there isn't an excess supply or excess demand for money. The "money market" is not the same as "the securities market". The "money market" is every single market in the whole economy (except barter markets).

[Mike Sproul]

There have been lots of times and places where people did use cows for money. We have an empirical opportunity here. Did cow shortages actually cause recessions? Did cow surpluses cause booms?

I have not come across and historical reports that would support this view. What I do find, in abundance, is reports of coin shortages causing recessions, and of coin influxes causing booms. (But not a word about cows!)

The problems with coins seem to center around Gresham's law. New coins are minted that weigh 16 oz. Wear and clipping reduces the weight to 15 oz. This reduces real money balances and causes a recession. The mint tries to issue new (16 oz.) coins, but they are hoarded. So the mint devalues and issues new coins that weigh 15 oz. These circulate and the recession ends. But the coins wear down to 14 oz. The process repeats and new coins are devalued to 14 oz. Seven hundred years go by and the coin that originally weighed 16 oz weighs just 1 oz., and we have a 700 year history of periodic coin shortages.

[Nick Rowe]

And I forgot to say: Clower has been a big influence on my thinking too.

Mike: Dunno. Were prices in terms of cows sticky? Because you need that too. But why should the physical form of the medium of exchange matter? (Except for the obvious practical reasons). If a shortage of gold coins causes a diminution in trade, and the failure of people to make mutually advantageous exchanges, why shouldn't a shortage of cows?

[JP Koning]

"... because not all MMs think alike on this question"

Yes, I was just going to comment on that. Cue another big intra-market monetarist battle. Scott Sumner and David Glasner vs Bill Woolsey and Nick Rowe -- a big repeat of last year's big MOE vs MOA debate.

[Nick Rowe]

JP: Yep! Internecine warfare is much more fun! What's sort of funny is that disagreements on this question cut right across some other party lines. And I'm not sure that all people have internally consistent views. For example, do some post-keynesians argue both that the supply of money is unimportant because the quantity of money is demand-determined, *and* that banks are really special and important because they create money? (Maybe I'm just imagining that, or misinterpreting them.)

[Tom Brown]

Nick: point taken about solvency. I agree... both the CB and the banks in aggregate (but especially separately) have to worry about solvency. It's perhaps less of a concern for the CB though? And the banks in aggregate can go down or get close to it, as we've seen.... but again, for the aggregate institution it's less likely than for any one bank.

And yes, I agree that far more MOE is used to buy IOUs (loans) than computers.

[Tom Brown]

... well I guess I did say "...only the Fed doesn't have to worry about solvency as much" originally, so I wasn't ignoring solvency altogether (the "as much" part) the 1st time... but yes, what I said gave the impression of "not at all." Solvency is definitely an issue for all.

[Mike Sproul]

Nick:
The best historical episode for the study of money is the American colonial period. The subject of sticky prices never came up. Their biggest problems were (1) coin shortages and (2) people offering to pay with underweight coins, low quality tobacco, etc.

I'm not sure why a shortage of cows wouldn't cause a recession. All I know is that they never mentioned it.

[Too Much Fed]

"Banks are financial intermediaries whose liabilities are used as media of exchange."

I'm going to skip the financial intermediaries part for now. So you are saying demand deposits are MOE? Are they also medium of account (MOA)?

[Too Much Fed]

"In long run equilibrium, banks are no different from other financial intermediaries. They borrow and lend. They make life easier for ultimate borrowers and ultimate lenders to get together, just like shopkeepers."

That sounds like loanable funds?

So for you a 20% capital requirement and a 50% capital requirement would not make any difference?

[Nick Rowe]

Mike: In a sense, Gresham's law is an example of price stickiness. If bad money traded at a proper discount to good money, then bad money would not drive out good. If exchange rates are fixed, an excess supply or demand for money will look like a shortage or surplus of foreign exchange.

TMF: "So you are saying demand deposits are MOE? Are they also medium of account (MOA)?"

I would say that central bank money is the MOA. The definition of a MOA is that prices for other goods are expressed in terms of that MOA. Commercial banks promise to redeem their money on demand for central bank money at a fixed exchange rate, which means they are quoting the price of their money in terms of central bank money, which means the central bank money and not commercial bank money is the MOA. Both are MOE.

"That sounds like loanable funds?"

Yep. Loanable funds in the long run equilibrium, a mixture of loanable funds and liquidity preference in short run disequilibrium. Or, to say it another way, it's the job of the central bank to try to make the loanable funds theory true at all times.

An increase in the legislated capital requirement would: increase the equilibrium spread between banks' lending and borrowing rates; reduce the total quantity of commercial bank money; reduce the risk of the central bank having to act as LOLR or bail out insolvent banks. I think. But it wouldn't make a qualitative difference to what I say here.

[Steve Roth]

Thanks, Nick!

I think I've got the "money is special, n-1 markets" thing you keep prattlling on about (grin) about 85% internalized. I really need to get it fully compiled into my brain because the client-side interpreter/emulator I'm using over here is both slow and buggy. (And it's not like I can think about it all day every day...) Working on it, a couple of readings of this got me closer.

Could use some help on this:

"there are two ways an individual can get more medium of exchange: he can buy more medium of exchange (by selling more of other goods); or he can sell less medium of exchange (by buying less of other goods)."

3. He can borrow (or more accurately retain?) more medium of exchange. Use his credit card instead of his debit card. If it's a business, tap that line of credit. (I'm wildly curious why, if demand for money is so high [and the cost of borrowing is so low] people aren't credit spending like crazy.)

This back to the reflux discussion. Now you can say he's "selling" future spending/consumption/borrowing to "buy" current spending/consumption/borrowing (buying current borrowing by selling future borrowing??), but I just don't think that's a coherent or usefully explicatory analogue to (not) buying/consuming real, newly produced goods today.

But back to the whole post, and my interpreter problem: I'm kind of flummoxed about how this replies to my post, which was all about incentives, reaction functions, and intertemporal optimization of consumption preferences -- none of which is mentioned here. There is a lot of *implicit* stuff embedded herein about those subjects, but much assembly (and deduction) required.

A somewhat weak analogy: I'm trying to understand relativity theory by thinking about a person in an enclosed capsule, who can't tell the difference between that capsule's acceleration and the capsule being held in place in a gravitational field--a warp in the space-time continuum. You say, "Think about two people's subjective lifespans, one at light speed and the other at rest..."

[Too Much Fed]

"I would say that central bank money is the MOA. The definition of a MOA is that prices for other goods are expressed in terms of that MOA."

I'm going to disagree there. When I go to buy something at a store, it seems to me the price is quoted in terms of currency and/or demand deposits. I can buy something for all demand deposits. Let's even say the store and I share the same bank. I write a check for an item. There is a swap of my demand deposit(s) into the store's checking account, while I get the item. No central bank "money" (currency, central bank reserves) involved. I'm assuming 1 to 1 convertibility of demand deposits and currency.

Also, I believe Canada has a 0% reserve requirement. Let's say people don't want currency any more. They turn it in for demand deposits at commercial banks. The commercial banks swap the currency for central bank reserves. The commercial banks decide they don't want the central bank reserves. The commercial banks swap them for treasuries. Now central bank money (currency, central bank reserves) equals zero (0). Now there is an all demand deposit economy. It seems to me the MOA (and MOE) in that economy are demand deposits.

"An increase in the legislated capital requirement would: increase the equilibrium spread between banks' lending and borrowing rates; reduce the total quantity of commercial bank money; reduce the risk of the central bank having to act as LOLR or bail out insolvent banks. I think. But it wouldn't make a qualitative difference to what I say here."

If demand deposits (commercial bank money) plus currency is the MOA and MOE, then I believe the capital requirement would make a difference.

I don't know if I can find it, but have you seen where Miles Kimball says banks allow borrowers to increase their purchasing power more than the purchasing power of savers decrease?

[Frank Restly]

Too much Fed,

"I don't know if I can find it, but have you seen where Miles Kimball says banks allow borrowers to increase their purchasing power more than the purchasing power of savers decrease?"

Part of the reason is that long term borrowing by private individuals is amortization borrowing. As the borrower pays back the loan, interest is only applied to the outstanding loan balance. Contrast that with debt issued by the government (coupon or accrual bonds) - interest is paid on the principle of the bond until the bond reaches maturity at which time principle is returned in full. Tax policy also comes into play - many forms of private interest payments (mortgage, student loan, corporate) are tax deductible.

This is why you see that individuals (even at today's low rates) are net interest recipients. It also explains that absence a change in credit demand, liquidity preference, or productivity nominal interest rates and inflation are positively correlated.

[Nick Rowe]

Steve: I know the feeling. It took me hours to write this pitifully short post. And I'm doing almost nothing else at the moment.

"3. He can borrow (or more accurately retain?) more medium of exchange. Use his credit card instead of his debit card. If it's a business, tap that line of credit."

"Borrowing" is just another word for "selling an IOU". It counts as an example of "selling more other (non-money) stuff".

"But back to the whole post, and my interpreter problem: I'm kind of flummoxed about how this replies to my post, which was all about incentives, reaction functions, and intertemporal optimization of consumption preferences -- none of which is mentioned here."

You are right. My post here doesn't directly address your story of why banks are special. What I am saying here is that creating MOE is the only underlying reason why banks are special, so if what you say is right, it must be a consequence of creating MOE. And if what you say doesn't follow as a consequence from creating MOE, what you say must be wrong.

"You guys" accuse "us guys" of not understanding banks. I'm flipping the accusation back: if you guys don't understand money, you can't understand banks!

TMF: Assuming commercial bank dollars are accepted at par is begging the question. If the BMO became insolvent, and my deposits at the BMO were only worth 50 cents on the dollar, the store would double the prices it charged me in BMO dollars, if it accepted them at all. Some Canadian stores will accept USD, but prices are quoted in Canadian dollars.

[edeast]

Hey,here is a selfish comment.

This was the best post on banking. http://www.freebanking.org/2013/07/11/misunderstanding-financial-history/

Here, https://archive.org/stream/banksbankingbank00macluoft/banksbankingbank00macluoft_djvu.txt
is B.E. Walker's essay on free banking in canada. The Introduction.

The B.E. Walker is a must-read imo.

Lee Smolin's Time Regained, oddly is helping my mind deal with the number markets, moe, relational metaphysics. Whitehead, which is what I was pursuing in undergrad.

The cryptocoins, are commodity coins + distributed ledger. Waiting for banks to develop on top of them, then the cryptocoin Bank-notes will be moe. I was at a music festival in the summer, and I was able to transfer my primecoin over to bitcoin, then to cash at a booth.

Armen Alchian's University Economics, is the best text.

Also sticky prices don't make sense. The mal-allocative kind. I think market participants are right and the explanation is wrong. Still think my edgeworthian explanation of infrequently traded items having stable prices works.

[edeast]

Good post btw.

[edeast]

Wrong forum, I need to get my own blog. Stop spamming people.

[JP Koning]

"For example, do some post-keynesians argue both that the supply of money is unimportant because the quantity of money is demand-determined, *and* that banks are really special and important because they create money?"

Yes, there are a few examples of this. See Cullen, for instance, although I could be wrong. Maybe other commenters can chime in on this? Is there an equivalent PK rift?

[Steve Roth]

Nick:

>"Borrowing" is just another word for "selling an IOU". It counts as an example of "selling more other (non-money) stuff".

I would say, "selling an IOU" is a particular way of describing "borrowing."

Another way of describing borrowing: "selling future borrowing."

I really question the conceptual utility of either construction in thinking about "spending." Of course the distinct possibility exists that I just don't understand the utility.

[Too Much Fed]

Nick's post said: "TMF: Assuming commercial bank dollars are accepted at par is begging the question. If the BMO became insolvent, and my deposits at the BMO were only worth 50 cents on the dollar, the store would double the prices it charged me in BMO dollars, if it accepted them at all. Some Canadian stores will accept USD, but prices are quoted in Canadian dollars."

Let's assume it is a one time event. I also like to talk about relative pricing. If it is a one time event, $1 of BMO demand deposits will still buy $1 of goods/services and $1 of Bank of Canada currency will still buy $1 of goods/services. The difference is the person holding $1 of Bank of Canada currency still has $1, while the person holding $1 of BMO demand deposits now has only $.50 (fifty "cents") of BMO demand deposits. Currency is not directly "defaultable", while demand deposits are directly "defaultable".

For example, assume all currency and demand deposits are circulating. $800 billion of currency and $6.2 trillion of demand deposits get converted to $7.0 trillion in demand deposits with a velocity of 2. Next, a bank/banks become insolvent, and depositors get affected. Now there is only $6 trillion in demand deposits with a velocity of 2. Everyone panics even though it is only a one time event and turns in their demand deposits for currency. Now there is $6.0 trillion in currency, still a shortage of MOA/MOE. If commercial banks (or even hedge funds) are the ones that produce new demand deposits (MOA and MOE) and don't use leverage/use margin, how is the amount of MOA/MOE going to expand?

[Tom Brown]

JP Koning, Cullen supposes (and so do I) that Nick is actually mostly agreeing with Cullen here:

http://pragcap.com/banks-are-special-the-ins-outs-of-money/comment-page-1#comment-152817

http://pragcap.com/banks-are-special-the-ins-outs-of-money/comment-page-1#comment-152835

As you can see, I was 67% sure of it. :^)

BTW, Nick, didn't this make your "flesh creep" to type out? "...supply of money is unimportant because the quantity of money is demand-determined"

I can't speak for Cullen, but what about this view: MOE is special, banks (uniquely in the private sector) create MOE, and thus they're special. In fact the MOE that banks create is so special that in some ways it's MORE important than the MOE that the Fed creates (I'm pretty sure that's where Nick get's off this bus). In other words CB created MOE is one of two things:

1. physical cash
2. Fed deposits

2. is mostly for facilitating bank MOE (which is more special)
1. is mostly for convenience of the public, and the public freely swaps their bank MOE for cash whenever it's most convenient for them to do so... and as long as they don't have to pay the banks to hold deposits, there's no special reason to exchange more bank MOE for more cash, even if there's mountains of CB MOE piling up at the banks.

So I'm going to unceremoniously dump 1. here. Unless we have to pay for the convenience of holding bank deposits (which I think would cross some sort of psychological threshold for people), it's a function of convenience.

As for 2., I like the facilitating argument, however, I freely admit the following:

non-bank private sector's (the "public's") stock of money = L + B + F
= stock of bank deposits + cash in circulation with public

to a 0th order approximation, with the following definitions:

L = banks loans
B = Bank held Tsy debt (a small fraction in the US... on the order of $200M)
F = Fed held Tsy debt

independent of "C" where C = stock of cash in circulation.

So, sure, F is substantial and contributes directly to the public's money stock, and for that reason it's important. MOE, after all, is "special" ... we do agree about that. I have more here:

http://brown-blog-5.blogspot.com/2013/08/banking-example-11-all-possible-balance.html

Sure I'm just an amateur at this stuff, and I'm POSITIVE I didn't come up with that, but I like it nonetheless. Hopefully one of you will take a look and tell me where I've gone wrong so I can learn something. ;^)

I introduce mortgage debt and intra-gov debt holdings (e.g. SS) in the next two examples (11.1 and 11.2) but it doesn't substantially change the approximation made here in #11 (as long as we realize that T = total of Fed, banks, and public held Tsy debt in #11, and excludes intra-gov ... and I *think* maybe foreign too... but not sure yet about that).

If you do go, check out my fancy interactive charts/spreadsheet! MS actually made something for free better than Google there!

So the special thing about CB MOE piling up in banks as excess reserves is not the reserves themselves but the corresponding F contribution to the public's MOE.

That's what differentiates ER from the following: Say that the Fed calls up the BEP and tells them that for psychological reasons (read "expectations") they're going to need $10T more in paper reserve notes printed up, which they're going to pay the cost of production for (say $1000) and then announce to the world that they're going to stockpile in case there's a rash of cash advances. This is a way to get the psychological effects of expectations, but for a lot less BS growth from the Fed's point of view ($1k instead of $3T). The thing that would be missing here is F, which is still $0: so no growth of bank created MOE. Now Sumner says that at "zero rates" the HPE is "very weak" and it's mostly an expectations game (I'll show you the quotes if you don't believe me), so if that's true, then why not just have the BEP print up a bunch of notes ready to go?

I'm kind of being silly here, but it's a serious question: in terms of expectations (expectations of future NGDP growth, or inflation or HPE) ONLY what's the difference?

[Too Much Fed]

Mix of posts:

'"In long run equilibrium, banks are no different from other financial intermediaries. They borrow and lend. They make life easier for ultimate borrowers and ultimate lenders to get together, just like shopkeepers."

That sounds like loanable funds?

So for you a 20% capital requirement and a 50% capital requirement would not make any difference?"'

'"That sounds like loanable funds?"

Yep. Loanable funds in the long run equilibrium, a mixture of loanable funds and liquidity preference in short run disequilibrium. Or, to say it another way, it's the job of the central bank to try to make the loanable funds theory true at all times.

An increase in the legislated capital requirement would: increase the equilibrium spread between banks' lending and borrowing rates; reduce the total quantity of commercial bank money; reduce the risk of the central bank having to act as LOLR or bail out insolvent banks. I think. But it wouldn't make a qualitative difference to what I say here."'

I hope to come back to this.

[Too Much Fed]

Tom Brown said: "In other words CB created MOE is one of two things:

1. physical cash
2. Fed deposits"

I don't consider Fed deposits (the demand deposits of the central bank) to be MOE in the real economy. I also don't believe they are MOA in the real economy.

[Tom Brown]

TMF, well consider them MOE in the "fake" economy then: i.e. banks use those exclusively when buying and selling to other Fed deposit holders (foreign CBs, the Tsy, etc) or the Fed itself. They are also used in this manner of course when banks act as intermediaries between other Fed deposit holders or the Fed and the non-bank private economy (your "real" economy).

[Tom Brown]

Nick, question about this part of your post:

"The talented artist can create drawings at the stroke of a pen, but cannot rent them out unless he can find someone who wants to hang that drawing on his wall. An excess supply of drawings immediately refluxes to the artist. The talented artist cannot create an excess supply of his product; banks (in aggregate) can. I borrow the drawings because I want to hold the drawings; I borrow the money because I want to buy something else."

Is there a typo in there... or two, because I don't follow. Let me re-phrase and please tell me where I go wrong:

1. Talented Artist (TA) creates goods, but can't rent them out unless demand exits.
2. If TA creates excess supply of goods it "refluxes" to him.
3. The TA CANNOT create excess supply. Huh? Number 2. just described what happens when he DOES create an excess supply! How do we now conclude that CANNOT be done?
4. Banks, however, CAN create an excess supply of their product (MOE). What?? Again, this is a surprise.
5. ...some other stuff that I'll skip

Typo(s) or no? Should the "cannot" be a "can" and the "can" be a "cannot?"

[Tom Brown]

Nick, another question. You write:

"1. ... This means that the medium of exchange is special because there are two ways an individual can get more medium of exchange: he can buy more medium of exchange (by selling more of other goods); or he can sell less medium of exchange (by buying less of other goods).

2. An excess demand for the medium of exchange prevents people making the mutually advantageous exchanges that would be possible if barter were easy so we did not need monetary exchange. That's what we call a "recession". ..."

So why then, if a recession is an "excess demand for MOE" isn't the reaction from "people" just as likely to be "buy more medium of exchange (by selling more of other goods) [e.g. sell IOUs to banks, i.e. take out loans]" as it is to be "sell less medium of exchange (by buying less of other goods)"

You propose two responses to an "excess demand for MOE" in 1., but focus on just one of them in 2. Why? Why does a recession (which you say is equivalent to "excess demand for MOE") lead to less buying and not more borrowing, if both buying and borrowing are the two ways people can respond to "excess demand for MOE?" Shouldn't you at least mention why you choose to focus on "less buying" rather than "more borrowing?" Is there something that's preventing the "more borrowing" route?

[Ralph Musgrave]

If we did use just cows as money (or gold coins), commercial banks would immediately start counterfeiting cows or gold coins. That is, if the going rate of interest charged by those depositing cows/coins in banks for extended periods, so that others could borrow was 5%, then commercial banks would soon discover they could simply credit cows/coins to borrowers accounts without actually needing to breed cows or dig up gold, and charge 4%.

Counterfeiting should be illegal. That is, the only form of money should be monetary base (either cows, coins or a fiat base).

Moooo.

[Ritwik]

Nick

"The idea that the banking system in aggregate can create an excess supply of money is highly controversial nowadays"

Change "banking system" to "financial intermediaries". Change "an excess supply of money" to "a lower price of risk".

And then, it would be perfect. :)

[Ritwik]

As in, the idea used to be controversial (perhaps no longer is), and that's what's really going on.

[W. Peden]

Nick Rowe,

"Some Canadian stores will accept USD, but prices are quoted in Canadian dollars."

That's a very elegant way of explaining the MOA/MOE distinction. In Northern Ireland, most shops will price in pounds but take euros. I suppose, when they give prices in both, then both the pound and the euro are the MOA.

[Vaidas]

Nick:
"Commercial banks care about the spread, which means *both* their lending and borrowing rates and opportunities. (Central banks aren't like that.)"

Good central banks are like that. Here is a recent BoE's version of Bagehot's dictum: "lend early and freely (ie without limit), to solvent firms, against good collateral, and at ‘high rates’".

[Nick Rowe]

Tom: "Is there a typo in there... or two, because I don't follow. Let me re-phrase and please tell me where I go wrong:"

There are no typos there.

Let me re-phrase:

The level of water on the Canadian side of Lake Ontario cannot be higher than the level of water on the US side. Because if (hypothetically) it *were* higher, water would *immediately* flow downhill from the Canadian side to the US side, making the levels the same.

The talented artist who rents out paintings cannot create an excess supply of his paintings in public hands. Because *even if hypothetically* he did create an excess supply, people would immediately return the unwanted paintings back to him, eliminating that excess supply.

"Why does a recession (which you say is equivalent to "excess demand for MOE") lead to less buying and not more borrowing,"

If there is an excess demand for MOE, individuals can respond by either: 1. buying less other stuff (including buying less IOUs); or 2. Selling more other stuff (including selling more IOUs). But if everyone else is also trying to sell more other stuff (including selling more IOUs) you won't actually be able to sell more other stuff (including selling more IOUs), because you can't find people who want to buy more other stuff (including buying more IOUs) from you. So 2 won't work in aggregate. But 1 always works, because nobody can stop you simply buying less other stuff (including buying less IOUs).

Gotta go to work. Back later.

[Andy Harless]

Nick,

Your view of the medium of exchange seems to be that it is special because (1) people hold inventories of it and (2) it is highly liquid on the sell side. But aren't these statements also true of commodities? The MOE is special in that everyone holds inventories of it, as opposed to most commodities, of which only some people hold inventories, but this difference strikes me as non-critical. If there were an excess supply of soybeans, then soybean holders would reduce their demand for everything else in order to replenish their soybean supplies, just as money holders do with money when there is an excess demand for money. Soybean gluts don't cause recessions because prices are not sticky in terms of soybeans. If there is an excess demand for soybeans, the price goes up until inventory holders are willing to sell. If there is an excess demand for money, the price stays the same, and the excess demand doesn't get corrected, so you get a recession. But I submit that any widely inventoried, stickily priced good could cause recessions in the same way. The medium of exchange is special maybe in part because it is so unusually widely inventoried, but mostly because it also happens to be the medium of account and therefore stickily priced. (I think I have argued in the past that, due to indexation, there was, in the 70's, some stickiness to the price of labor in terms of petroleum, which is widely inventoried, and therefore an excess demand for petroleum helped cause recessions even though it is clearly not a medium of exchange.)

[Peter N]

"What I am saying here is that creating MOE is the only underlying reason why banks are special"

Finally. A root of the problem. The sort of banking done in your (and many other peoples') model is only a small part of what a major financial institution like Bank of America or Morgan Stanley does. The top 20 banks of the world have something like $44 trillion in assets. If they find a way to collectively screw up it WILL affect the economy. We've already seen this in events post 2007, so it's not just a theory.

Another root is the assumption that the system will behave according to the preferences of the customers. Remember the old joke - "where are the customers' yachts"? The banks stole at least several hundred billion dollars. We have the court documents to prove it.

Moreover most of this was lost or will be subject to up in litigation for years to come. Morgan Stanley is currently defendant in 44 major lawsuits (for amounts over $100 million, often way over) with more to come. It looks like the banks will have to repay the GSEs so much money, that the government's loans to the GSEs will be profitable - repayment of multiple tens of billions of dollars.

If banks behavior gets bad enough (Countrywide, Libor, Lehman, Bear Stearns, Dexia, Bankia, Spain, Cyprus and Greece, Ireland... and now China?) they will have macroeconomic effects, and it's necessary to look at mechanisms and incentives. After all Moral Hazard has been part of Macro for a while IIR.

[Tom Brown]

Nick, thanks! You wrote:

"cannot create an excess supply of his paintings in public hands" ... "in public hands" ... that's the key there! He certainly can create an excess supply... or "stock" ... he can even mail them for free like AOL DVD's ... such that everybody on Earth is sick to death of them, and trashcans are overflowing... but he can't force them into their hands. Why is "in their hands" so important though?

You write:

"But if everyone else is also trying to sell more other stuff (including selling more IOUs) you won't actually be able to sell more other stuff (including selling more IOUs), because you can't find people who want to buy more other stuff (including buying more IOUs) from you."

So even if interest rates are very low, you can't actually borrow because banks don't want your IOUs, because everybody is trying to sell them IOUs. Is that it?

[Tom Brown]

Nick, so during recessions (times of excess demand for MOE), if in aggregate the public is trying to sell more IOUs than the banks want, and interest rates are very low (historically low), then why is it that long term rates don't go up? Instead it seems like the requirements for "credit worthiness" go up. Or perhaps actual credit worthiness goes down. Or perhaps though low, rates are still high compared to the public's ability to pay them. But then again, if demand for MOE is truly high, then the public SHOULD be willing to pay the rates and in fact force the rates higher, no?

Cullen explains this differently: he says that rather than an excess demand for MOE, instead there's "not enough income relative to desired saving." Also "spending is a function of income relative to desired saving. QE doesn’t increase income or saving so why would spending increase?"

http://pragcap.com/banks-are-special-the-ins-outs-of-money/comment-page-1#comment-152840

http://pragcap.com/when-non-banking-experts-become-overnight-banking-experts/comment-page-1#comment-152629

Commentator Geoff brings up the difference between money and wealth:

http://pragcap.com/when-non-banking-experts-become-overnight-banking-experts/comment-page-1#comment-152918

[Tom Brown]

Nick, I was discussing this further with "Geoff"... he brought up "People don’t want to borrow, even at the low interest rates, because they are trying to save. They don’t actually want more MOE ... they want more ... more income, or savings."

http://pragcap.com/when-non-banking-experts-become-overnight-banking-experts/comment-page-1#comment-152923

My response was the example of underwater homeowners who wanted to take advantage of low rates but couldn't because their homes were underwater and/or they'd lost creditworthiness (regular work). But they didn't really demand more MOE, they demanded a better balance sheet, no? MOE (through borrowing) was the way to get there. Since they'd be paying off an old debt with a new one, MOE levels wouldn't change.

[Frank Restly]

Tom,

"People don’t want to borrow, even at the low interest rates, because they are trying to save. They don’t actually want more MOE ... they want more ... more income, or savings."

Or they want an asset that has a greater potential return than the cost of servicing previously incurred debt.

[Tom Brown]

Nick, it seems I lost one of my comments there somehow. Let me try again. You write:

"The talented artist who rents out paintings cannot create an excess supply of his paintings in public hands."

So the "in public hands" part is crucial there. The artist could certain create WAY more (stock? supply?) than was wanted, and even mail them out to people for free... and trash cans would be overflowing w/ his work, but he couldn't get them into their "hands."

Regarding people selling IOUs to banks for MOE... well I guess I cover that above. I was just rephrasing you (in my lost comment): banks don't want it (IOUs) because too many people are selling. But again, oddly rates are low, so why doesn't excess MOE demand drive long term rates up?

Perhaps it's because creditworthiness is down. Or the definition of creditworthiness is up. Or incomes are down so rates appear high. Some combination of those? I'm going to say mostly that creditworthiness is low (the homeowner's house is underwater, or he lost his job, or both) but a little bit of all that. But again, it seems in the case of the homeowners, it's not debt created MOE that's desired, but a change of debts: an improvement in BSs. A temporary increase in MOE/debt levels is the way to get there. Improving the BS improves the feeling of wealth.

[Tom Brown]

re: Artist: I know we're talking analogy here, but what's the general way to distinguish "excess supply" (e.g. overflowing trashcans with the artists work) from "excess supply in the public hands" (e.g. hanging on walls)? Do these two kinds of "excess supply" go by different names in general? Because certainly the artist can create the former kind but not the latter.

[Philippe]

"if in aggregate the public is trying to sell more IOUs than the banks want, and interest rates are very low (historically low), then why is it that long term rates don't go up?"

Interest rates go down for safe borrowers, and go up for risky borrowers that no-one wants to lend to. So safe banks and governments will have low borrowing costs, and underwater homeowners won't.

[Nick Edmonds]

Nick,

I would tend to agree with you that the ability of banks to issue liabilities that are used as a medium of account gives them a special status and that this status is important macroeconomically.

However this idea risks obscuring the much more important way in which banks matter, which relates to their lending decisions. For many agents, whether they can realise their planned expenditure depends mainly on whether the bank will lend them money. The decisions of banks can therefore be the critical determinant in whether spending takes place or not. The bank is making the crucial decision, not the household or firm. This, more than their role as suppliers of money, is why banks are important. If banks only supplied the medium of exchange by lending to agents who did not have unfulfilled spending plans, they would matter less.

Of course, this control of the purse-strings makes banks very important, but it doesn't make them special - the actions of non-bank lenders can matter in the same way. The difference is that banks account for the majority of lending to borrowers at the margin of credit-worthiness.

Knowing why banks matter is important. It explains why we need to understand what shapes banks' lending behaviour. It's hard to explain the growth and contraction of credit before and after the crisis without understanding how banks operate. It also tells us that we cannot afford to lose sight of what non-bank financials are doing. Although banks are important, they are not special enough to be exclusively so.

[Tom Brown]

Cullen helped bring to light some concepts concerning your artist analogy (same thread I linked to before). Well I can't blame Cullen entirely... these are my thoughts mostly in response to me explaining the artist thing to him.

Artist actually repos his paintings... he agrees to buy them back at a premium. So it's the art patron that makes money here.

Artist burns his painting when he gets it back (loan agreement is burned... and of course in the case of borrowers (artists) and art patrons (banks) the bank burns the MOE too (i.e. debits the account)).

Now after I just "straightened" Cullen out about which was the bank and which the borrower in the artist example, you'll tell me that *I* had it backwards! Right? Or did I get it right? ;)

[Philippe]

you got it backwards. The artist rents his paintings out in Nick's example.

[Tom Brown]

Philippe, yeah I know, but I thought what was important about that example was that the painting (like a loan agreement or most other goods) is unlike the MOE (the "rent"). Isn't that the point he's trying to make there? That that artist can't create an excess supply of rented artwork on the walls of art patrons, but that banks in aggregate can create an excess supply of MOE? Just like borrowers can't in aggregate (in reaction to a recession) force the banks to hold an excess supply of their IOUs (loan agreements)... thus they need to reduce (in aggregate) spending instead (to meet their own excess demand for MOE).

[Philippe]

Nick's point seems very simple in that example. If you are an artist who paints pictures and then lends them to people for a fee so that they can hang them on their wall, you can't supply more paintings than people want overall, because if people don't want the paintings they'll just bring them back to your studio.

[Tom Brown]

Philippe, right. However that's what distinguishes money from all other goods (like paintings). For the right price people will always take money.

So if you're trying to lend out paintings it's a very different business than lending out MOE. So in that sense I see how he's comparing the painter to the bank but nonetheless concluding they are inherently different because of the product they are lending.

Perhaps I was stretching things too far by putting the painter in the borrower's shoes and the patron in the bank's.

[Philippe]

I think Nick's point is that people will always accept money even if they don't actually want money itself, because they know they can use the money to buy other things, because everyone else accepts it. i.e. they can get rid of it.

So the argument is that banks can create an excess supply of money because people will always accept it, even if their only intention is to subsequently get rid of it. If everyone accepts it and everyone also wants to get rid of it, then you can have an excess supply that nonetheless keeps circulating and pushing up prices. It doesn't get 'returned to the studio' as it were. But it seems to me that even if that were true the money could just keep circulating in the financial sector, pushing up asset prices, without leaking out much into the real goods and services economy.

I don't really get the point about the IOUs. In a recession, demand for some IOUs increases whilst demand for others falls. So borrowing costs for risky borrowers can increase whilst borrowing costs for low-risk or risk-free borrowers like governments falls to near zero. Bank deposits are bank IOUs, i.e. bank debts, or loans to banks. So an increased demand for bank deposits (a form of money) is also an increased demand for loans to banks. But only for loans to safe banks!

[Tom Brown]

Philippe, I get what you're saying, however, I'm not sure I follow the "real goods and services economy" point.

I was more interested in relating Nick's definition of a recession (excess demand for MOE) to the balance sheet recession concept that I'm more familiar with. I think it's possible that they might actually be two ways of looking at the same thing.

Cullen, Geoff, Steve Roth, and I have been kicking this around at pragcap. I think this is a nice current summary from Cullen:

http://pragcap.com/when-non-banking-experts-become-overnight-banking-experts/comment-page-1#comment-152980

[Philippe]

"I'm not sure I follow the "real goods and services economy" point"

Monetarists argue that this "excess supply of money" ends up pushing up the prices of goods, causing general price inflation. But it seems to me that any supposed "excess supply of money" could just push up asset prices, without having much effect on general inflation.

"I was more interested in relating Nick's definition of a recession (excess demand for MOE) to the balance sheet recession concept that I'm more familiar with"

The BSL theory claims that recession can be caused by too many people trying to save and pay down debt at the same time, whereas Nick's argument seems to be that recessions can only be caused by too many people wanting to hold money (MOE) at the same time, either in the form of cash or deposits.

But the cost of borrowing for low-risk borrowers falls in a recession whilst the cost of borrowing for risky borrowers rises, which suggests that people don't necessarily want to hold money, they just want safe and liquid assets with some sort of return, which could be government bonds, bank deposits, blue-chip corporate bonds etc.

[Philo]

“I borrow the drawings because I want to hold the drawings; I borrow the money because I want to buy something else.” Well, you might be an art dealer, planning to rent out the drawings yourself to customers. Admittedly, few are art dealers, while everyone deals in the medium of exchange.

The more important point is that it is an oversimplification simply to distinguish between money on the one hand and all other goods on the other hand. Non-moneys differ considerably in *liquidity*—in the ease with which one can get rid of them in exchange for their “true worth” (what they would fetch in a well-functioning market for them—or something like that). The more liquid a good, the more willing people are to accept it in an exchange even if they don’t want to hold it. The more “near moneys” there are, and the nearer they are to pure money, the less the importance of pure money itself—the less appropriate it is to call it *the* medium of exchange. But even in a situation with lots of very-near moneys, the medium of *account* has special importance. So for theoretical purposes—so as to cover near-money scenarios, too--the notion *medium of account* is more basic than *medium of exchange*.

Banks are the more important in our situation because of our paucity of near-moneys.

[Philippe]

I meant BSR (Balance Sheet Recession) not BSL, obviously.

[Nick Rowe]

Andy: assume everyone holds inventories of milk (they're all Brits, and get random urges for a cuppa). Let's see if an excess demand for milk has the same effect as an excess demand for money:

If there's an excess demand for money/milk, everyone either: tries to sell more of all other goods to buy more money/milk; or buys less of all other goods to sell less money/milk.

It just doesn't work for milk, because:

1. dairy farmers sell milk and never buy milk; everyone else buys milk and never sells milk. Everyone both buys and sells money.

2. When we buy or sell milk, we buy or sell it only for money. When we buy or sell money, we buy and sell it for everything else.

[Nick Rowe]

Nick Edmonds: "However this idea risks obscuring the much more important way in which banks matter, which relates to their lending decisions. For many agents, whether they can realise their planned expenditure depends mainly on whether the bank will lend them money."

Imagine a financial intermediary that borrows money and lends money but does not issue its own money. (It's a non-bank FI). Sure, normal financial intermediaries like that matter, by making it easier for savers and spenders to meet each other, but their macro impact on aggregate demand is minimal. If you want to save, and I want to spend, and we meet, you lend me $100, which means you spend $100 less and I spend $100 more, so total spending is unchanged. It only makes a difference if you would have held that $100 in cash if you hadn't met me, and I would have held the same amount of cash either way. Non-bank financial intermediation only affects aggregate demand insofar as it reduces the demand for money for any given supply of money.

[Andy Harless]

Nick,

Dairy farmers drink milk, don't they? If dairy farmers decide they want to drink more milk but there's a price ceiling, they will sell less milk and buy less of everything else. If everyone else's wages are denominated in milk (even if they're paid in money) you will get unemployment, because the people who normally sell stuff to dairy farmers will have to reduce their production.

(I'm assuming that the dairy farmers' increase in milk consumption comes out of inventory -- i.e. they were already producing their maximum output. Now milk isn't really a good example, because it's highly perishable, so you have to turn over your inventories quickly, and therefore you can't consume out of inventories for very long. Petroleum is a better example: most people don't hold inventories of it, but some people hold fairly large inventories, and if they decide they want to consume more of those inventories at the current price, and the price can't rise, and wages are indexed to oil prices, then there will be a recession.)

As far as being a MOE, I don't think that makes money truly "special," but it does mean, as a matter of degree, that money is an extreme, in that it is (1) more liquid than anything else and (2) more broadly and deeply held in inventory than anything else. So if you want to start a recession and you have a choice of goods with which to start it, your best bet is to use the MOE. And banks aren't truly special for producing the MOE, but they are at an extreme in terms of their potential macroeconomic impact relative to other industries (roughly in the same sense that the US and China are at an extreme in terms of their potential economic impact relative to other countries, but more so). Also, in practice, money happens to be the MOA (and there are nominal rigidities), so it actually is special, and perhaps that makes banks special too, but that's another story.

[Philippe]

Tom,

"Fed deposits [are] mostly for facilitating bank MOE (which is more special)"

What does this mean?

[Too Much Fed]

Nick said: "Likewise, an individual commercial bank that accepted a deposit before figuring out if it had potential customers for loans wouldn't be very wise. My bank both advertises for new borrowers and advertises for new depositors."

Let's say you transfer some demand deposits from bank A to bank B into a checking account. You're saying bank B can now lend out these same demand deposits to a borrower. Correct?

[Tom Brown]

Philippe: "more special" ... I don't expect all to agree w/ me on that. That's an assertion on my part: I tend to look at bank deposits as the CORE of our money system: playing the role of MOE in the majority of transactions, especially in the private sector: bank cards, credit cards, check writing, home loans, car loans, automatic withdrawals for mortgages, electric bills, gas, water, power, trash etc. I don't know if that's actually true, but that's how I live my life, so maybe I'm biased there! (I don't like touching or managing dirty, clunky, losable scraps of paper or bits of metal). I see all the rest playing a facilitating role: Fed deposits, cash, etc.

Look at it this way, if we banned cash (my dream) and there was just a single commercial bank, then the private sector would run off of bank deposits entirely. In addition, if there were no reserve requirements then the ONLY role for reserves would be for transactions between the public & banks on one hand, and the gov, Fed, or foreign govs on the other. The domestic private economy would be entirely run off of bank deposits w/ no role for "base money" whatsoever.

Now perhaps in the future... if people just HAVE to have cash, we could invent some way for people to print their own, much like we print out our own coupons now. The banks would LOVE it! They could close down all those ATM machines (they'd litter the landscape like public pay phones do now)... boarded up, graffiti covered. The banks could ditch the stupid vaults, and reduce robberies to 0 (and probably fire lots more people). Plus the gov would no longer need a Mint or the BEP or any of that nonsense.

[Tom Brown]

... in short, there'd be a veritable cornucopia of benefits!!

[Tom Brown]

... wait even more benefits come to mind! No more cash registers (inefficient... those people who make them can get real jobs ;), convenience store robberies (except for stolen merchandise), pan-handlers (they'd only be looking for food)... wow! Now that I think about it, I'm even more in favor of a cashless society!

[Tom Brown]

... it'd be interesting to see what drug dealers do in a cashless world!

[Tom Brown]

... plus think of the surveillance benefits! Keeping us safe from all sorts of stuff!:

http://nsa.gov1.info/utah-data-center/

"If you don't have anything to hide, you have nothing to fear!"

Hahahaha! :D

[Nick Edmonds]

Nick

"Imagine a financial intermediary......"

If I want to save and you want to spend, that's fine if I'm prepared to lend to you. If I'm not prepared to lend to you and you can't spend, there's no way I can save. All that happens when people try to save and other people aren't spending is that nominal income falls. I think both Keynesians and monetarists would agree on this (although the monetarist would focus on saving money as opposed to saving generally).

So I agree that if all a financial intermediary does is lend my money to someone I was going to lend to anyway, then it's not making much difference. And sometimes that's all that's going on. But not always. One of the most important features of the growth of ABS / CDO market prior to the crisis was the channelling of funds to borrowers from sources that would never have made those loans directly. This process had a huge impact on the national structure of assets and liabilities. That's going to have a big macroeconomic effect.

Now you could say "Well, the financial intermediary is not important. All I need to know is that, for some unspecified reason, lenders became prepared to provide funds to a much wider class of borrowers." And you could try and analyse it in those terms. But you'll never be able to explain why it happened without understanding what was going on in shadow banking.

[Philippe]

"I see all the rest playing a facilitating role: Fed deposits, cash, etc."

Ok but what does that actually mean? What does "facilitating role" mean?

If you owe me $100, and you pay me with a $100 note, has that note "facilitated" your $100 debt?

[Philippe]

The problem is that "reserves just play a facilitating role" is a meaningless phrase.

[Philippe]

"if we banned cash"

So are you saying people should be forced to use private banks? They should have no choice in the matter? That doesn't sound like much of a market to me, more like a form of fascism.

[Philippe]

"w/ no role for "base money" whatsoever."

Even in your strange imaginary economy with only one commercial bank (!?!?) where people are forced to use that bank and forbidden from withdrawing their money from that bank (basically a weird monopoly fascist system), there is still a role for base money, because the bank has to make payments to the state, which it has to do with base money. Also if any one else has accounts at the central bank then payments to those accounts would also have to be made with base money.

[Tom Brown]

re: "w/ no role for "base money" whatsoever."

Philippe, you took that out of context... I pointed out that this was with regard to transactions completely within the private economy. I explicitly pointed out, just as you did, that it would STILL be needed for transactions between CB deposit holders + the CB itself, or for transactions involving... well, here, I'll just quote what I said:

"...if there were no reserve requirements then the ONLY role for reserves would be for transactions between the public & banks on one hand, and the gov, Fed, or foreign govs on the other."

That was in the sentence immediately preceding the one you quoted.

BTW, I'm not proposing we establish this weird fascist uni-bank... Hahaha.. I'm simply pointing out that IF we had such a bank (as some MMT folks imply or state we should *** I think! *** not really sure what they think about that), then I was describing how that would affect the role of base money.

But regarding analysis purposes, I think it is OFTEN fine to consider ALL the commercial banks in aggregate AS IF they were one big commercial bank with lots of highly competitive branches... a view which is essentially the same as the uni-bank model. Now of course in some respects this is not a fair way to look at it, ... I'll grant you that.

[Tom Brown]

"facilitating role" ... I warned you that was my opinion and that not all would agree. I explained what I meant: that MOE ... point of sale ... is more and more done explicitly with bank deposits directly .... or on the spot generated credit via a credit card, and cash is getting sidelines somewhat. I think that MOE is the MOST important aspect of money (NOT MOA!) and thus bank created MOE is more and more important... and that's why I see it as the core of our monetary system.

You are free to disagree if you want. I know I'm not going to convince people of my opinion on this.

So viewed in that way, we get the bulk of our money through private loans. Sure ... some comes from bank held Tsy debt (very very little) and some comes from Fed held Tsy debt, but the bulk through private bank loans. Once a loan is made, the borrower can chose to exchange his bank created MOE for outside cash money... or change it back again. But the source was most likely a loan.

The Tsy and the Fed send out very little in the way of cash payments directly to private non-bank entities.

[Tom Brown]

Philippe... hmmm... Wait! Are you the same "phil" that I was just in a thread with on my blog??

[Tom Brown]

... and BTW, although I actually am all about convenience and prefer using a bank or credit cards almost all the time... and having ALL my bill payments fully automated... you apparently missed the subtle ironic undertone in most of that! I realized that it sounded kind of fascistic... and that NSA link about the Utah data center is a joke... it's a parody site (yes, I know the center is real). It's pretty funny actually, you should check it out.

[Philippe]

"I think it is OFTEN fine to consider ALL the commercial banks in aggregate AS IF they were one big commercial bank"

Do you think it is fine to consider all shops as if they were one big shop? All businesses as if they were one big business?

It's not much of a market if you can consider all the businesses as if they were the same single business.

The reality is you can't consider all the commercial banks as if they were one big commercial bank because they're not one big commercial bank. That's the basic problem. And an imaginary world in which everyone is forced to use one bank and forbidden from withdrawing their money from that bank doesn't have anything to do with the real world. So its not actually a useful example for describing the real world.

"bank created MOE is more and more important"

But what do you mean by "important"? For example in a gold standard system there are far more banknotes and bank deposits in existence than there is gold bullion or coin. Does that mean that banknotes and deposits are more "important" than the gold?

"we get the bulk of our money through private loans"

You're making the mistake of assuming that because most money is things like M2 this means we get the bulk of our money through private loans. You're getting things mixed up.

[Tom Brown]

Philippe,

Yes, base money is important. Yes it serves an important role, even in a one commercial bank world. I'm not saying it's not important, but the fact is with inflation targeting or CB funds rate targeting or a combo of both (what we've typically had: CB funds rate short term, e.g. every six weeks, and inflation rate long term), then the quantity of money is endogenous in both the short run and the long run. It's the nominal rate of interest that is exogenous in the very short run (6 weeks or less, for the Bank of Canada anyway), but endogenous in the long run. Summary: it's all endogenous long run.

[Tom Brown]

... and BTW, cashless isn't really so far fetched:

http://www.cbsnews.com/8301-202_162-57399610/sweden-moving-towards-cashless-economy/

http://www.wired.com/business/2012/08/canada-will-beat-us-to-cashless-economy/

and in terms of percentage of value of all transactions in the US, it's only 0.2% (from the Cleveland Fed):

http://www.clevelandfed.org/Forefront/2012/winter/images/13_02_cash.jpg

[Tom Brown]

"Do you think it is fine to consider all shops as if they were one big shop? All businesses as if they were one big business?"

I could be wrong, but I'm pretty sure that's done all the time in macro economics. Do you see Nick using the word "aggregate" here? Aggregation, from what I can tell, is a big part of macro.

[Philippe]

There's a difference between macroeconomic forms of aggregation and treating a sector (banking for example) as if it were a simple monopoly, which is what you want to do. It's not a monopoly, so it doesn't make sense to describe it as one.

Your example with one monopoly bank which people are *forced* to used and *forbidden* from withdrawing their money from, has absolutely no similarities to the system as it exists in reality.

[Philippe]

the central bank supplies reserves at a given (target) interest rate (subject to other requirements such as collateral and adequate capital) so in that sense the quantity is endogenously determined - the CB sets the price and lets quantity float. I don't think that means what you seem to think it means. The nominal funds rate is not endogenous, the central bank sets the funds rate.

[Philippe]

"in terms of percentage of value of all transactions in the US, cash is only 0.2%"

Yes the total value of cash transaction is relatively small, because people tend to use cash for relatively small transactions. The largest value of transactions goes through wire transfers, and this category includes central bank wire transfers via settlement systems like Fedwire. Fedwire is the system through which payment in reserves, i.e. reserve settlement, occurs in the US.

"According to the Fed’s data, average daily dollar volume of payments settled via Fedwire – the Fed’s real-time gross settlement system – was 17 percent of GDP in 2011 (Federal Reserve 2012). This is a fairly typical percentage that one sees across countries, as Fullwiler (2008) shows. In other words, within 5 to 10 business days, the value of payments settled using central bank balances eclipses that country’s annual gross domestic product (GDP)."

(Scott Fullwiler: An endogenous money perspective on the post-crisis monetary policy debate, page 8)

http://www.rokeonline.com/roke/post%20crisis%20monetary%20policy%20debate.pdf

[Philippe]

"the nominal rate of interest that is exogenous in the very short run (6 weeks or less, for the Bank of Canada anyway), but endogenous in the long run."

It's not endogenous in the 'long run'. The central bank always *sets* the base rate (Fed Funds rate etc). The CB might decide to raise or lower the interest rate to try and control inflation or change expectations, but this is a decision that the central bank makes. It decides to change the interest rate, the interest rate doesn't change 'endogenously'.

[Tom Brown]

Philippe, if they are targeting inflation, it is. The rate changes as a function of what's required to keep inflation on target.