« When did Canada's recession end? A WCI poll | Main | Damage report »

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

"Forget currency (mainly, I admit, because I can't really get my head around where it fits in). Assume that all money is bank money. The commercial banks are the central bank's agents for creating money."

And, "Central banks got rid of their fleet of helicopters when currency got replaced by demand deposits as the primary medium of exchange (again, I admit, I'm unclear on where currency fits into my picture)."

This is why we disagree so much. Back to a mortgage. When someone buys a house, does he/she get the mortgage and eventually and usually pay with a check (demand deposit)?

Think about an economy with all currency and NO currency denominated debt. What would it look like?

"Forget currency (mainly, I admit, because I can't really get my head around where it fits in). Assume that all money is bank money. The commercial banks are the central bank's agents for creating money."

Is it more accurate to say "The commercial banks are the central bank's agents for creating currency denominated debt with an interest rate attached and repayment terms attached."?

"Central banks got rid of their fleet of helicopters when currency got replaced by demand deposits as the primary medium of exchange (again, I admit, I'm unclear on where currency fits into my picture)."

Assuming you want to increase the amount of the medium of exchange, should you increase demand deposits (currency denominated debt) or currency?

"If the central bank increases the supply of reserves, and banks respond by increasing the supply of loans, the money supply expands,"

I thought people said supply of loans was determined by capital. Is it the central bank increases the supply of reserves, interest rates come down, and demand for currency denominated debt (a loan) goes up (they are hoping for that)?

"Forget currency (mainly, I admit, because I can't really get my head around where it fits in). Assume that all money is bank money. The commercial banks are the central bank's agents for creating money."

With legal tender laws, is currency used to make the interest payments on currency denominated debt whether gov't (thru taxes) or private?

"But the current economic crisis does not exhibit the symptoms of a problem with the AS curve. We have symptoms of excess supply of goods and labour, and falling prices. That suggests a fall in Aggregate demand, relative to Aggregate Supply. So even if bad banks have caused to LRAS curve to shift left (or move rightwards at a slower pace), that is not the current problem. We have a fall in AD, and current output and employment is demand-constrained, not supply constrained. We have a general glut.

I believe Stiglitz has said something similar to this. Recessions happen with problems on the AS side, while depressions happen with problems on the AD side, almost always currency denominated debt problems.

I'm going to add consumer currency denominated debt.

Why should bad banks cause AD to fall?"

because the people (the lower and middle class experiencing negative real earnings growth) couldn't make the interest payments leading to defaults and should not be borrowing. It was bad bank lending behavior.

It is not the banks that need bailed out. It is the lower and middle class.

Another point. If lower interest rates don't get excess savers to spend, excess debtors to go further into currency denominated debt, and people who were excess savers but aren't anymore because of work hour cuts or wage income cuts to go further into currency denominated debt, what should happen?

Too much: "When someone buys a house, does he/she get the mortgage and eventually and usually pay with a check (demand deposit)?" Yes.

"Think about an economy with all currency and NO currency denominated debt. What would it look like?" Horrible. No pension plans. But OFF-TOPIC!

"Is it more accurate to say "The commercial banks are the central bank's agents for creating currency denominated debt with an interest rate attached and repayment terms attached."?" No.

"Is it the central bank increases the supply of reserves, interest rates come down, and demand for currency denominated debt (a loan) goes up (they are hoping for that)?" *Quantity* demanded of loans goes up. Yes.

"With legal tender laws, is currency used to make the interest payments on currency denominated debt whether gov't (thru taxes) or private?" No. We usually pay by cheque.

Now, back off a little please. Let someone else attack me!

No attack here, generally I thought that was pretty good.

Just one question, which might be a dumb question, but here goes anyway - if someone doesn't want to hold currency what alternative do they have that might keep the money from showing up as a bank deposit. For example, I could buy stock, but then whoever sells me the stock has the money in their bank account.

Seems like the money created by the loans is a hot potato that gets passed around from bank account to bank account, but is always in one bank account or another. Is there somewhere else it could go that isn't occurring to me?

And if not, in what sense do 'ultimate lenders' really choose to lend, lending would equal borrowing as an accounting identity, but it wouldn't really have any explanatory meaning.

Declan: Thanks.

Your "hot potato" metaphor is exactly the same metaphor that "disequilibrium monetarists" use to describe the same process. There is nowhere it can go. Unless someone pays off a bank loan. But then the bank, if nothing else has changed, will just make a loan to someone else. They can't get rid of the hot potato, but in their attempts to get rid of it, by buying stuff or lending it, they cause output and income to rise, and/or prices of goods and assets to rise, until income and prices rise enough that the demand to hold money rises enough, that they decode they want to hold it after all. And that's the new, happier, equilibrium.

You lost me on the last bit, I'm afraid.

Assuming the CB does not intervene, if banks (or most banks) have insufficient capital, do they switch from supplying money to demanding money? The bank itself can satisfy it's demand for more money by selling less money (contracting credit), with the secondary effect that contracting credit can create a recession. Which causes everyone else to demand more money, so they sell less money and keep it in their bank accounts - which helps satisfy the banks demand for more money.


Patrick:

I'm not quite sure what you are saying. I'm going to give two different answers, for two different interpretations:

1. Suppose a bank has insufficient capital, and wants to get more. A banks capital represents its wealth, or net worth. Or shareholders' equity. It's what it owns (assets) minus what it owes (liabilities). If a bank borrows (demands money?), that doesn't increase its capital (unless it borrows by issuing new shares). It just increases its assets and liabilities by an equal amount.

2. If a bank has insufficient capital for its current assets and liabilities, and doesn't or can't get more, it might call in loans, or sell assets. That contracts its balance sheet, and reduces the supply of money at the same time.

I was imagining a bank that had barely sufficient capital to start with. So it won't call in loans, but won't make new ones either. It won't contract, but can't expand.

NR: In the first part of your post, you get everything precisely backwards. I guess that's what happens when you start with "Banks are financial intermediaries".

You and I actually agree when you say "A general glut is caused not by an excess desire to save, but by an excess desire to save in the form of the medium of exchange." I think your definitions are screwy and stick with the common sense definition of saving as "income you do not spend".

You lost me with how banks can make people who don't want to take out more loans take out loans. And the Govt (albeit not the banking sector) still has massive fleets of helicopters they could fire up on Monday.

Also, please note how a falling economy, with collapsing AD, impacts even a well capitalized bank's desire to lend.

Thanks for the response Nick. What I meant at the end of my comment was, if banks create money by lending, and that money has nowhere to go but back to the bank as a deposit, then the view of the bank as intermediary doesn't make much sense to me. The 'ultimate lender' has no agency - they have money so the money sits in a bank. They don't *choose* to lend it, and they won't stop lending it if interest rates on deposits get too low. They can't do anything (in aggregate) *but* lend their money.

The decision makers are the borrower and the banker. If they are both willing, then the money supply expands, if either of them is unwilling or unable, then money supply is static (or decreases as existing loans get repaid).

The person who has a deposit in the bank is just a bystander. The depositor could give their money to a potential borrower directly (in return for a promise of interest payments in the future), but this wouldn't have the same effect as borrowing from a bank, and to the extent that it replaced bank lending, it would have a negative impact of money supply.

Nick: Mises has a great passage on this precise subject.

Winterspeak: I'm glad to see your response.

"In the first part of your post, you get everything precisely backwards. I guess that's what happens when you start with "Banks are financial intermediaries"."

I thought I was being very orthodox there. Nothing especially original. Just textbook money and banking, minus the money! Perhaps it's that last part ("minus the money") you object to. So do I.

"You and I actually agree when you say "A general glut is caused not by an excess desire to save, but by an excess desire to save in the form of the medium of exchange." I think your definitions are screwy and stick with the common sense definition of saving as "income you do not spend"."

Yes! I thought we would agree there. But you and i must be about the only people who do (I exaggerate slightly). "My" definition isn't mine. It's the standard keynesian national income accounting textbook definition of "saving". I agree it's not a useful definition in this context.

"Also, please note how a falling economy, with collapsing AD, impacts even a well capitalized bank's desire to lend."

Yes, because it makes some otherwise good loans go bad. Absolutely. I forgot to include that bit. It's an important deviation-amplifying feedback loop. But it isn't essential to my story, and I need to keep my story as simple as possible.

"And the Govt (albeit not the banking sector) still has massive fleets of helicopters they could fire up on Monday." Filled with currency? Or bank money? I want to leave fiscal policy out of it, because it's hard enough keeping my head straight as it is.

"You lost me with how banks can make people who don't want to take out more loans take out loans."

You misunderstood what I was trying to say. They want to take out the loans, but they don't want to hold the loans as money. They want to spend it. Each individual can spend the money and stop holding it, but in aggregate someone's got to hold it. So in aggregate, they are forced to hold more money than they want to hold.

Declan: OK. I understand you better now. I think you might be right, and making an important point. But need to think about it some more. It's late here.

Hi Nick

Regarding the "hot potato" issue, I'm with you on the point that buying any asset or good/service just "moves the money around." But, are you assuming there are no time deposits, money market accounts, money market funds, and the like, which enable individuals to simply relabel existing deposits as less liquid, interest earning balances while remaining on the same bank's balance sheet? Or, are you assuming those things are "money," too?

Jon: "Mises has a great passage on this precise subject."

Do you remember where? It's so long ago I read Mises on money, I can't remember it all. Wonder if that's one of the places I got these thoughts from? It wouldn't surprise me.

Scott: Hmmm. I don't want to call those things "money", because they aren't media of exchange. Dunno. I have to think about that one.

"It's late here."

Indeed, I hesitated to make my comment since I was worried it might keep you up!

Scott's question is similar to what I had in my mind when I wrote my first comment - does it matter if money goes into, say, a money market fund instead of just a deposit? Does that give people an alternative to bank deposits - or just make the bank more comfortable knowing that its deposits are tied down for a little longer? I can't quite get my head around that.

Is there a difference between price inflating with currency and price inflating with currency denominated debt?

"They want to borrow" or do they? I thought they wanted to save. The bank can lower the price of money to encourage it, but they can't lower it to zero or they won't be able to profit. Capital impairment will increase their desired profit margin. Many would like to borrow if they didn't have to repay, but banks can't lower it below their expected risk and expect to profit. If they can't lower rates enough between the zero bound and their margin to increase lending how can they create money? How do they push on the string?

Combining Scott's and Winterspeak's points, it might be helpful to separate out the act of pricing claims with the act of expanding balance sheets. Any entity that expands its balance sheet is adding to demand.

But in order to expand your balance sheet you need someone else to price your obligation. Here banks play a role in performing idiosyncratic credit analysis as households cannot supply uniform accounting information as in the financial markets.

So it might be more accurate to call banks "appraisers", or "price-setters", because of their ability to price claims on a mark-to-model basis, due to their government backing. That is the only special thing about banks (and playing golf rounds with the Treasury Secretary:P). Once a price is assigned to a claim, your immediate purchasing power is increased by the same amount in exchange for a long term liability. However the claim is priced, the effect on increasing demand is the same.

I think the key issue here is the desire of borrowers to de-leverage, and some evidence of this is that net household borrowing began to rapidly shrink in 2007, prior to Bear or Lehman failing. This means shoving more money to banks is not going to boost aggregate demand. Moreover, the only difference between a "bad bank" and a "good bank" is in the capital structure and underwriting discipline. All that can be fixed by government administratively, and fairly quickly (given the political will) whereas getting household borrowing restarted is harder to do.

"It ain't the loans what matter; it's the money creation that goes along with loan creation what matters. That's why (bad) banks matter. Bad banks won't create money."

So you have an exclusive cause and its effect...and the effect matters but the cause doesn't. Rather meaningless distinction, no?

"A general glut is caused not by an excess desire to save, but by an excess desire to save in the form of the medium of exchange."

No. Savers can just sell their money income for other financial assets. The financial asset sellers aren't net savers -- the buyers are.

Great post.

Why isn't the second view orthodox? Isn't it obvious? Lol.

The connection to currency is redeemability. Of of the zero maturity assets are redeemable in currency.

Connecting up the 3D interest rate structure and monetary disequilibrium isn't very hard. When you hit the zero nominal bound, efforts to reduce nominal interest rates lower creates monetary disequilibrrium.

At a negative nominal interest rate, currency is better than other assets, and an excess demand for currency develops. All the other zero maturity assets are redeemable, and so the excess demand for currency generates an excess demand for them. Their quantities drop. So, excess demand for money in general. Reduced expenditure.

If the nominal interest rates on all money (including currency) can go negative, (passively adjusted with the rest,) then this isn't a problem. Low, perhaps negative nominal interest rates don't cause an excess demand for any sort of money. But at the same time, market interest rates on the short safe stuff can fall, which allows that whole 3D structure to adjust.

Capital constrained banks (ignoring redeemability into zero interest currency) are consistent with monetary equilibrium. They can shift both liaiblity composition to increase in quantity of money. They can shift their asset composition to reduce the capital constraint. (Collect on risky assets and buy safe ones.) And finally, they can lower the interest rates they pay on monetary liabilities. This reduces the demand for money, adds to profitability, and fixed the capital problems.

Having money redeemable in zero interest currency messes this up. The Central Bank has to expand the quantity of zero interest currency to meet any excess demand or else the entire banking system is going to end up with an excess demand for moeny. They can't fix it buy lowering their interest rates paid on money if that makes currency better than deposits. Excess demand for currency.

If the reason for the excess demand for currency is the zero nominal bound, where people are demanding currency because clearing the markets for short and save assets would make them have negative yields (so that currency is better,) then having the central bank expand the quantity of currency by purchasing those very assets in excess demand at the zero bound fixes nothing. More currency, yes, but less of those securities on the market.

The central bank wants to fix the problem, it must buy riskier and longer term assets that are not in excess demand at the zero bound. Notice this changes the 3D shape. The central bank is taking more risk.

Anyway, I think what you call heterodox is the core. But the orthodox stuff is more or less correct too. They must be. It is often two ways of looking at the same thing.

Nick I guess I was describing option 2, so I wasn't playing by the rules.

Nick wrote: "But then each ultimate lender would have to collect information on the creditworthiness or profitability of the ultimate [borrower]."

I find the Treasury Direct market to be a good model. The Treasury loans direct to citizens bypassing the 'intermediaries'. Economists need to get off their tushes and start thinking along the line that many/most loans serve public purpose, are simple enough, and therefore don't need an intermediary.

http://www.treasurydirect.gov/

Nick wrote: "Forget currency (mainly, I admit, because I can't really get my head around where it fits in)."

Most everyone has an 'accounting', including Nick.

Currency is the tangible form of the Fed's 'medium of exchange' and the foundation of their accounting. Take that same currency and put it in a bank and it becomes reserves. Take it out of a bank and it is called cash.

Cash/Reserves are ultimately the only thing to pay taxes with, BofA bank money just won't do, Ford/Walmart/HomeDepot nonbank money won't do either. Neither will Nick's 'medium of exchange'.

"Different people and firms face different costs of funds, or opportunity costs of funds."

Exactly and what monetary policy needs to avoid in order to retain 'independence'. Most 'risk' analysis is used to screw certain classes of citizens.

Why not allow any citizen to offer a Treasury Security to the Fed in exchange for Fed funds at the current rate?

Winslow: "Currency is the tangible form of the Fed's 'medium of exchange' and the foundation of their accounting. Take that same currency and put it in a bank and it becomes reserves. Take it out of a bank and it is called cash."

That's the accounting at the back of my mind. It is exactly as if each bank has a wad of currency sitting in the Bank of Canada, as reserves. But then there's the question of people's choice as to whether to hold money in the form of currency or in the form of demand deposits. That's not the same thing.

Bill: Thanks!

What I call the "heterodox" view was never really "orthodox". As you know, it's a distillation of Yeager, Clower, and "disequilibrium monetarists" like Laidler. Always a minority taste, but it seems to never have been properly understood, because the Lucas revolution swept away the brief flowering of general disequilibrium theory in a monetary exchange economy. So everybody forgot this stuff (if they ever learned it, which few did). I never know how to categorise it. Is Yeager a monetarist? Not really. And Clower is much more Keynesian. Everybody talks about monetary economies, but nobody stops to think about what makes the medium of exchange so special. And those special features only get revealed when markets are in disequilibrium (non-market clearing).

So I called it "heterodox". Partly also, I confess, to gently tease Post Keynesians, who I think don't always realise just how orthodox and mainstream they are, in at least this respect. They are part of the "establishment", in many respects (though not of course in all).

"Anyway, I think what you call heterodox is the core. But the orthodox stuff is more or less correct too. They must be. It is often two ways of looking at the same thing." I can't figure out if there is a fundamental difference, or if they are just two ways of looking at the same thing. Duck/rabbit. What makes me suspect they are different is that I see no way to capture the fact that money is a medium of exchange within the orthodox perspective. It's implicit within the orthodox perspective, of course, because there can't be an excess supply of goods otherwise. But is the orthodox view fully consistent with the role of money as medium of exchange?

Your discussion of currency makes sense within the orthodox view. Just another asset, into which all other assets are redeemable, which happens to have a 0% nominal yield. But it's also a medium of exchange. And it's hard enough getting my head around the weird properties of just one medium of exchange, let alone two. Plus, as far as I can tell, the stock of currency really is (nowadays, in Canada) demand-determined, while the stock of demand deposits is supply determined. Banks helicopter demand deposits, but the Bank of Canada doesn't helicopter currency. Dunno.

Wicksell's nightmare: "So you have an exclusive cause and its effect...and the effect matters but the cause doesn't. Rather meaningless distinction, no?"

'Twas but a rhetorical flourish! But loans don't cause money, because the expansion of loans and the creation of money happen in the same act at the same time.

"No. Savers can just sell their money income for other financial assets. The financial asset sellers aren't net savers -- the buyers are."

Sure. Someone who buys a financial asset with part of his income is saving. But that doesn't disprove what I said, about only the desire to save in the form of money can cause an excess supply of goods.

RSJ: Sorry. You lost me in that comment.

Lord: "The bank can lower the price of money to encourage it,..." You mean lower the *interest rate* on *loans*? (We need to be really careful here to distinguish loans from money, and what we mean by "price".)

If a bank's capital constraint is binding, that means that if it had more capital it *would* find it profitable to expand loans at current rates of interest on loans. If it weren't profitable to expand loans anyway, then the capital constraint cannot be binding, by definition.

Scott and Declan: the simple model at the back of my mind was one where banks' only liability was chequable demand deposits. If banks offer (say) term deposits as well, then an individual can extinguish his demand deposit by switching it to a term deposit. It's just as if he paid off a bank loan. I expect the effect on the supply of money depends on what the bank does next. If it does nothing, the supply of money contracts. If it responds by creating a new loan (or just buying something), then the money supply does not contract.

"most people buy most things not with currency, but with the money created by commercial banks. We pay by cheque, or debit card, or ATM from our chequing account"

And credit cards? A credit card effectively creates a (bank?) loan and a deposit in another bank as it is used, but what about the quantity effect of the undrawn credit line?

Rebel: I think credit cards don't count, because you still have to pay the bill by cheque at the end of the month. It's just a way of postponing payment.

Nick,

Credit cards or lines of credit don't count, but spending with credit cards (or lines of credit) does, as it creates a loan and a deposit for the bank issuing the card, and the deposit is debited and credited to the bank at which the recipient of the spending banks. The reserve account of the credit card issuer is debited and the reserve account of the spending recipient's bank is credited. Paying down your credit card decreases the loan (on the credit card issuer's balance sheet) and debits your own deposits, while your bank's reserve account is debited while the credit card issuer's reserve account is credited.

So, spending with a credit card raises deposits. Paying down the credit card reduces deposits. That's just accounting.

Best,
Scott

Nick,

I was just making the point that as far as demand is concerned, there is no difference between a customer taking out a loan from a bank and a business selling a bond to the credit market. Both boost demand by the same amount (assuming that the proceeds from both are spent to purchase output, and not to purchase other financial claims). So there is nothing special about banks in this regard.

The special thing about banks is that they serve households who would otherwise not get access to the credit markets, and of course banks have certain government backing that allows them to avoid market pricing.

This simplifies the assumption a bit, so you don't need to get into discussions about whether money market funds "count" or whether other financial assets count. You don't need to try tease out whether an increase in some monetary aggregate increases effective demand or not, etc. If you assume that financial claims are liquid (a reasonable assumption), then you can treat all financial claims equally, regardless of maturity. In that case, only new issues (or net borrowing) add to demand, and demand is increased by how much the new issues are priced, independently of the instrument or whether the obligations are priced by a bank or a market maker.

In that case, instead of saying "A general glut is caused not by an excess desire to save, but by an excess desire to save in the form of the medium of exchange.", you would say "A general glut is caused not by an excess desire to save, but by an excess desire to save in the form of financial assets, as opposed to tangible assets". Then you can look at the imbalance between the desire to accumulate financial assets and the willingness to incur financial liabilities generally, with the possible role of government to fill the gap by issuing liabilities.

Well said, RSJ!

Thanks Scott,

I guess the only difference (maybe not a real difference?) I have with MMT is that I don't want the government to always supply all the net financial liabilities needed to fund private sector net financial assets. Even if doing so would not cause inflation and would achieve full output.

In some cases, the private sector can desire to accumulate *too many* financial claims, in which case my preferred resolution would be to try to change the underlying structural issues so that the private sector desires to accumulate the "right amount" of net financial assets -- specifically whatever is needed to fund real productive investment needs modulo a leverage multiple (which should be constant over the business cycle).

Ideally, if productive investment is growing by 3%, then deficit spending should be 3% of GDP. Over time, that number should decrease as a result of increasing productivity. Obviously this will vary over the cycle as that leverage ratio changes, but if the numbers get too far out of whack from this, then you are no longer funding investment, but growing consumption in excess of wages or a current account imbalance.

In this case, I would try to address these structural imbalances via other levers rather than just funding private sector net financial asset desires in unlimited amounts. As an example, if I was emperor of Japan in 1990, and the private sector wanted to accumulate, say 1.5GDPs worth of NFA to pay for all the excess real estate debt, then I would first put the banks into receivership and cause mass debt write-offs, at which point, you wouldn't need to deficit spend so much, but just enough to help recover from the shock of the write-offs. The reduced debt burdened on households would be a more effective demand boost, it would work faster, and it would solve an underlying structural problem more quickly than 2 decades of falling prices and enormous public works. If I was emperor of the U.S. in 2008, I would do the same. I'm not sure what the most popular MMT decision algorithm is in these cases.

"A general glut is caused not by an excess desire to save, but by an excess desire to save in the form of financial assets, as opposed to tangible assets"

- not true in the case of an excess desire to save in the form of existing (as opposed to new) tangible assets

Scott: On credit cards. What you say seems to make sense to me. But at the same time, when you pay by credit card you still have to pay again, when you pay off the credit card. What's different about paying by cheque (or debit card) is that when BMO customer deposit $100 of cheques drawn on TD, and TD customers deposit $101 cheques drawn on BMO, the $100 cancels, and BMO only needs to transfer $1 reserves to TD. I'm not as clear on this as I should be.

But what I really want to respond to is the other argument in the comments above: a 3-way argument between anon, RSJ+Scott, and me(+Winterspeak). I think it's important. And, I think I've got my head clear on it too!

First off, anon's position is exactly the orthodox position. It's what the textbooks say, and what we teach and learn in economics classes. It's what I teach. But I think it's wrong!

Let me restate anon's position: take a closed economy, and one with no government, for simplicity. We divide all goods (real and financial) into two groups: newly-produced goods, and the rest.

Cd + Id =Y means that the demand for newly produced consumption goods plus the demand for newly produced investment goods = the output of newly produced goods. It's a (semi-) equilibrium condition ("semi-" , because it ignores the supply side). And if Cd+Id is less than Y, then firms reduce Y, and you get the general glut of excess supply (assume the output firms and workers *want* to produce and sell, Ys, is constant).

We can define desired saving as Sd=Y-Cd, and re-state the semi-equilibrium condition as Sd=Id.

According to the orthodox view (anon's view), an excess of desire to spend your income on *anything* other than newly-produced goods causes a general glut. It could be money, it could be other financial assets, or it could be real goods (like land and antique furniture) that are not newly-produced.

Here's why I think the orthodox view is wrong. Suppose we start in equilibrium, then suddenly everyone wants to cut consumption of newly-produced goods and buy land. There is an excess demand for land. Even if the price of land is fixed, so you can't eliminate the excess desire to save in the form of land by a rise in land prices, people will want to buy land but won't be able to. Unable to buy land with their income (because nobody is willing to sell), they will change their plans on saving in the form of land. They start looking at second-best alternative plans.

If their second best alternative plan is to go back to wanting to buy newly-produced consumption goods, there's no general glut.

If their second-best alternative plan is to buy more antique furniture, that creates an excess demand to buy antiques, so they can't buy antiques, and I repeat my argument above.

I can run exactly the same argument against RSJ and Scott, if people desire to save in the form of financial assets (except money), instead of land, or antiques. Either the price (interest rate) on financial assets adjusts, until desired savings equals investment again, or it doesn't adjust. If it doesn't adjust, we just get an excess demand for financial assets, so people can't buy financial assets, and have to change their plans, and consider a second-best alternative. Repeat.

It's only if they desire to save in the form of the medium of exchange (money) that we get a general glut. If people can't "buy" more money (sell more other goods) because there's an excess demand for money (excess supply of other goods), they can always "sell" less money (buy less other goods.

It is only an excess desire to save in the form of money that causes a general glut. An excess desire to save in the form of other financial assets (RSJ and Scott), or real goods that are not newly-produced (anon), will only cause a general glut if it leads to an excess desire to save in the form of money.

By the way, anon: since you are commenting here, and clearly know some economics, why not give yourself a real fake name? Every Tom, Dick, and Harry calls himself "anon". ;-)

Nick:

You claim that the quantity of currency is demand determined. I understand, of course. Currency is printed and issued in responses to withdrawals of currency from reserve balances at the central bank. Banks generally withdraw currency from those reserve balances to meet currency withdrawals from depositors.

Why not make interest on reserve balances at the central bank -5%? We can imagine all the nominal interest rates in the economy falling--checkable deposits, short government bonds, long bonds, whatever.

What is the problem? I think you end up with a demand determined quantity of currency equal to total wealth and the nominal interest rate at zero.

If you won't put up with that, then the illusion of passive demand determined currency breaks down.

I favor privatized currency. Without other changes, that makes reserve balances the sole monetary base. But under current conditions, where banks have the right to withdrawal currency from reserve balances, then currency is key.

That it seems demand driven is nothing more than an artifact of interest rate targeting. All money is "demand driven" if we say the central bank changes the quantity to keep an interest rate on target. Or, for that matter, anything else.

The central bank does choose the quantity of currency. And that drives everything else. If they are so lost in their policy tool that they don't see that, then it is just self-delusion. But a clear eyed vision of what is going on doesn't mean that the quantity of currency doesn't have to adjust if they are going to use their instrument.

It is even true if they are using explicit loans, what we used to call discount rate policy. There is a shortage of loanable funds at the rate unless they change the currency they control.

All of you accounting folks--

when people save by purchasing already existing financial or tangible assets, what does the seller do with the money they receive?

You are assuming that their desired money holdings--currency or balances in checkable deposits increase.

Your model has the amount of money people hold passively adjust to everything else.

Monetarism is based on the notion that it doesn't just passively adjust. It might well adjust, but for a reason. And so, the quantity of money and the demand to hold money is the key.

The straw man version of monetarism is that the demand to hold money is constant, so that changes in the quantity of money result in a proportional increase in spending. But that is a straw man. All we ask for is an explanation as to what is happening to either the quantity of money or else the demand to hold money. How is it that the demand to hold money rises to match the increase in the quantity of money?

If the non-govt sector has a desire to accumulate financial assets, the only way the entire sector can do this on balance is via government deficits. That's the only way to for there to be a NET increase in financial assets held by the non-govt sector, BY DEFINITION.

Others are talking about a desire to save by some people that's offset by a desire to spend by those that eventually find themselves with the balances. Right now, though, the private sector as a whole has been attempting to deleverage (and, of course, the desire to save doesn't have to result in a transaction to purchase financial assets, as savers can simply hold deposits or relable them as time deposits, though I doubt that's more than a portion of what's going on in practice). The non-govt sector can't really do that in real time (again, by definition) aside from the effects automatic fiscal stabilizers (such as lowered income leading to lower taxes paid, etc.) that raise the deficit. The other option is to have a change in relative prices or interest such that the desire to deleverage on the whole ceases or is even reversed, which is obviously the preference of neoclassicals. From the MMT perspective, that approach might possibly work temporarily (or it might not), but overall it ignores the lessons of Minsky.

meant to say change in prices, not relative prices.

Nick,

Regarding credit cards, that is an increase in the money supply and a purchase. Like any spending on credit, it may or may not be paid off "at the end of the month." To the degree that it is, yes, you're right in the accounting sense (though the spending happened, for sure).

But, in the aggregate, credit card debt incurred within the month often (usually) is not paid off at the end of the month. Indeed, it's not until you get recessions (generalizing here) that in the aggregate you might see aggregate credit card debt outstanding actually decline.

NR - "Here's why I think the orthodox view is wrong. Suppose we start in equilibrium, then suddenly everyone wants to cut consumption of newly-produced goods and buy land. There is an excess demand for land. Even if the price of land is fixed, so you can't eliminate the excess desire to save in the form of land by a rise in land prices, people will want to buy land but won't be able to. Unable to buy land with their income (because nobody is willing to sell), they will change their plans on saving in the form of land. They start looking at second-best alternative plans."
--why necessary to assume there's not a market clearing price for the money price of land? why can't there be a market clearing price for excess demand for existing assets? agree if you assume not, then the money alternative becomes the source of the glut - but why assume that?

BW - "when people save by purchasing already existing financial or tangible assets, what does the seller do with the money they receive? You are assuming that their desired money holdings--currency or balances in checkable deposits increase. Your model has the amount of money people hold passively adjust to everything else."
--same question; that's what a market clearing price means - satisfaction with the money result

SF - agree as a government response to a glut; separate issue from NR's question about the source or nature of gluts

Nick,

Thanks.

"It ain't the loans what matter; it's the money creation that goes along with loan creation what matters. That's why (bad) banks matter. Bad banks won't create money."

I had been looking without success for a way of putting that thought. You have found the formulation I will henceforth quote.

In return, I will try to suggest how currency fits in. Think ex-ante. The forward monetary variable in the light of which we economic agents make our choices of prices to offer at/expect to accept, plan quantities to offer/buy and amounts to save/dissave is expected MV. Expected MV turns out to be the stock of all assets multiplied by the likelihhood of their being used to settle transactions in the coming period. A perfectly illiquid asset (if one exists) would have a likelihood of zero. A highly illiquid asset class like real property may have a likelihood of (roughly) the order of 0.0001. A time deposit for a year may have a likelihood around 0.1 - it is a thousand times more likely to be used in settlement of a trans action than real property. A high denomination note may have a likelihood of 2.0. A coin may have a likelihood of, say, 100 or 500. The key is that all asset classes may expect to be used in the settlement of transactions to some degree, and to that degree they enter into the expected MV. Currency is just one such asset class.

(This ex-ante view also seems to help when considering how small changes in the expected liklihood of big asset classes being used in settlements of transactions affect monetary conditions. It brings into the picture the unused credit lines (including those of credit cards) and the credit created between non-financial economic agents (my supplier gives me a credit limit at his warehouse, my publisher pays me six weeks in arrears, etc.). But all that is for another occasion.)

Credit card debt isn't an increase in the quantity of money.

Lines of credit, however, impact the demand for money.

The supply of money isn't the same thing as nominal expenditure.

Nearly all spending is funded out of current income. People earn income and spend it.

Credit shifts funds between debtors and creditors. Lenders spend less and borrowers spend more. Repayment means debtors spend less and creditors spend more.


Most credit is not created by banks. Most expenditure is not financed by loans.

Now, checkable deposit issuing banks do make some loans. But treating all spending as if it were due to an excess supply of money is blindness.

If there were no government, imbalances between the supply and demand to hold financial assets can be cleared through changes in prices and yields.

The only problem is with currency. With fixed nominal price, its relative price can only change through changes in the price level. With a zero nominal yield, its yield can only change through expected changes in the price level.

With deposits, they have a fixed nominal yield, and so their nominal quantity needs to adjust unless the prices of all the other goods are to change. Fortunately, the nominal yield on deposits can change.

There is never any need for the government to create more financial assets. Only if the government monopolized the issue of currency, keeps its nominal yield at zero, and makes it the medium of redemption for all the rest of the economy, do problems develop.

We can imagine a situation where the central bank owns all the private financial assets and there remains an excess demand for currency, and no one wants to sell any private securities to the central bank. But the notion that the government must create more financial assets because the private sector wants to hold more is mistaken.

anonTDH: "--why necessary to assume there's not a market clearing price for the money price of land? why can't there be a market clearing price for excess demand for existing assets? agree if you assume not, then the money alternative becomes the source of the glut - but why assume that?"

Agreed. We are on the same page. We can imagine a hypothetical case where the government has price controls on land, so there could be an excess demand for land. But even in that case (I argue), that would not cause an excess supply of newly-produced goods (unless it spilled over into an excess demand for the medium of exchange). (Notice, by the way, I am contradicting Walras' Law when I say that). But if the price of land adjusts to market clearing, we can't get the excess demand for land anyway. And I would make the same argument for financial assets (except money). Whether or not the interest rate adjusts to get rid of the excess demand for financial assets, it can't create an excess supply of newly-produced goods (unless it spills over into an excess demand for the medium of exchange).

On the same point: Scott: "If the non-govt sector has a desire to accumulate financial assets, the only way the entire sector can do this on balance is via government deficits. That's the only way to for there to be a NET increase in financial assets held by the non-govt sector, BY DEFINITION."
Agreed. It's an accounting identity (in a closed economy). But suppose the government does not create a deficit to satisfy the private sector's demand for financial assets. Will that cause an excess supply of newly-produced goods, and a recession? (If those financial assets are not money). I say "no", unless it creates an excess demand for money.

Similarly, if the non-govt sector has a desire to accumulate land, the only way the entire sector can do this on balance is via government sales of land. That's the only way to for there to be a NET increase in land held by the non-govt sector, BY DEFINITION. But suppose the government decides not to sell land. Will that cause an excess supply of newly-produced goods, and a recession?

Bill: I think you were writing too quickly, and meant to say:

"With CHEQUABLE deposits, IF they have a fixed nominal yield, [DELETE and so] THEN their nominal quantity needs to adjust unless the prices of all the other goods are to change. Fortunately, the nominal yield on deposits can change."

Now, the nominal yield on deposits can change, and if it changes by the right amount, and in the right direction, that will eliminate any excess supply or demand for deposits. But will it? My head's not clear on that.

Suppose there's no currency. All money is bank deposits. Start in full equilibrium. Then the central bank increases the supply curve of reserves. Banks expand loans and deposits. There is an excess supply of deposits hot potatoing around the economy. If banks raised their interest rates on deposits, that would eliminate the excess supply of deposits. But would banks do this? As far as I can see, the only reason an individual bank would have to raise the interest rate on deposits is if it wanted to gain reserves (take them away from other banks). But I don't see why individual banks would respond this way if the original shock that displaced us from equilibrium were an increased supply of reserves.

I'm still puzzled.

David: Thanks! (But you had better fix my grammar first!)

But I don't think your way of looking at the spectrum of liquid/illiquid assets works. Apart from money, no other asset is ever used to settle a transaction. If I have land, and want a house instead, first I sell the land for money, then I use the money to buy a house. It's always the medium of exchange, on one side of all transactions.

Even though there's only a difference in degree in the liquidity/transactions cost of all assets, with a spectrum with land and houses at one end, and the medium of exchange at the other, that difference in degree becomes a difference in kind (Clower? Yeager?). The asset that is most liquid (has the lowest transactions costs) behaves qualitatively differently from all the others. It gets used in ALL exchanges.

"Credit card debt isn't an increase in the quantity of money"

Yes it is, unless it is a non-bank like a department store issuing the credit. In that case, you've created a receivable but--true--no deposits. Please show me with balance sheets how money isn't created when someone spends with a bank issued credit card. Good luck.

"Credit shifts funds between debtors and creditors. Lenders spend less and borrowers spend more. Repayment means debtors spend less and creditors spend more."

Wrong. Many (most, even) lenders do not use any funds to create a loan. For example, whether you are a bank, a department store, or a supplier of materials to a business, no "funds" are used. For the bank, the loan creates a deposit . . no prior funds necessary. For the department store, the loan is a receivable . . . no prior funds necessary. For the supplier, the loan also creates a receivable. The lender does not necessarily spend less, and particularly in the case of banks, has no less "funds" available, and no more "funds" available when the loan is repaid (aside from less capital applied).

Bill: thinking about interest rates on deposits some more, let's assume there are no admin costs on demand deposits (or that banks charge separate fees for each cheque, etc., to cover those admin costs). Then each individual bank has two equivalent ways of getting more reserves: borrow them from other banks, including the central bank; attract more demand deposits. With a competitive banking system, with x% desired reserve ratio, the interest rate on demand deposits would always equal (1-x) the interest rate on reserves. With x=0%, the interest rates would be the same.

So if we start in equilibrium, then the central bank increases the supply of reserves, the interest rate on reserves falls, loans and deposits expand, and the interest rate on deposits FALLS TOO. So it will not increase to eliminate the excess supply of deposits.

Nick. . . regarding land and net financial assets, you said what I would have in your response to David.

Bill: I thought your "banks make loans in something other than money" comment was priceless, but I think your "negative nominal interest rates on cash currency" is even better. I'm eager to hear how this will be implemented. I hope it involves flamethrowers.

NICK: Yes, your explanation was the orthodox one, and the orthodox one is wrong. It's wrong because it looks upon banks as financial intermediaries who somehow "loan out deposits". If banks were financial intermediaries who loan out deposits, the other stuff would be right.

Hi Nick,

"I can run exactly the same argument against RSJ and Scott, if people desire to save in the form of financial assets (except money), instead of land, or antiques. Either the price (interest rate) on financial assets adjusts, until desired savings equals investment again, or it doesn't adjust."

I hate to say this, but this dispute may be about stock-flow confusion. Please forgive my wordiness -- I'm too busy to be brief :)

Money as the medium of exchange means that all income flows are in terms of money. But the desire to accumulate financial assets is expressed as a desire to increase your total stock of financial holdings. In other words, I agree that everyone desires to *earn* money, but they do not desire to *hold* money per se. Bonds, stocks, or other financial claims are just as good. So let's separate out two desires:

1. (flows) the desire to increase your holdings of financial assets by earning financial surpluses, defined as earning more money than you consume by purchasing real goods/services whether for consumption or investment
2. (stocks) allocating those holdings among stocks, bonds, etc.

For 2., I claim that portfolio shifts do not affect aggregate demand. Suppose a company borrows money to retire bonds, or sells bonds to retire debt. None of that affects purchases of real goods or services, and none of that will cause a glut, even though it will shift some of the monetary aggregates.

Moreover, if everyone is happy with their current level of profitability, but is unhappy with their portfolio holdings, you may see a stock market crash or yield divergences, but this will not translate into layoffs or changes in output. 1987 is a good example. But that is a rare, as all that people really care about is the total value of their portfolio, and the only way to increase their financial holdings in aggregate is to increase your financial surpluses, or #1. That is what causes gluts.

For #1, the amount of financial surpluses earned, by identity is exactly the amount financial deficits. Everyone who earns money but does not spend it on goods is withdrawing from demand in the goods market -- they are a sink, in terms of cash-flows. A sink can only exist if there is an offsetting source somewhere else in the economy. The source is funded by borrowing or issuing liabilities (here we need to include someone drawing down their own stock of financial assets as borrowing from themselves).

There is no shortage of willing sinks -- everyone wants to be as profitable as possible. The system is constrained by the number of willing sources. Moreover, as Scott pointed out, the issuance of liabilities is not constrained by the desire to accumulate assets, whereas the reverse is true.

This is why if people stop borrowing on real estate in Florida, then the profits of General Mills decrease. Government better step in and deficit spend, or you hope to get people to borrow on real estate in Ohio, etc. When a source disappears, the size of the pie of gross financial surpluses shrinks by an equivalent amount.

When businesses stop being profitable, they liquidate capital and lay off workers. Everyone has a certain cost of capital, below which it is not profitable for them to create output in the first place. That is how output decreases as a result of a decline in the issuance of liabilities. If you are talking about *flows*, you can characterize this as the desire to *earn* money profits that are not re-invested in goods, in excess of willingness to dissave money, as flows occur with money. If you are talking about stocks, you can characterize this as the desire to accumulate financial assets in excess of the desire by others to incur financial liabilities.

"Wrong. Many (most, even) lenders do not use any funds to create a loan"

Good point, Scott.

The same happens when you buy claims in the credit market (even new issues). Imagine a company selling bonds to an investor. The investor liquidates his money market holdings to buy the bond, and the company takes the cash proceeds of the bond sale and puts them into a money market fund. Even though many sales and purchases may have been triggered, effectively all that happened was a swap of a money market asset for a newly issued bond.

Just as with a bank creating a loan, the result of the operation is that the new asset matching the new liability is first assigned to the *borrower*, as the investor's financial net worth is unchanged as a result of buying the bond.

To the degree that the borrower then spends that money by making purchases, they are adding to demand and more importantly, they allow the rest of the sector to earn financial savings. This is the only way that financial savings can occur.

"Only if the government monopolized the issue of currency, keeps its nominal yield at zero, and makes it the medium of redemption for all the rest of the economy, do problems develop...But the notion that the government must create more financial assets because the private sector wants to hold more is mistaken."

It's true that in theory, you might be able to convince people not to accumulate nominal financial assets and to re-invest all their money profits in tangible assets. That would prevent gluts. Trying to eliminate the desire of the private sector to accumulate financial assets all so that the government need not issue liabilities is a hopeless fight against human nature.

RSJ:

"1. (flows) the desire to increase your holdings of financial assets by earning financial surpluses, defined as earning more money than you consume by purchasing real goods/services whether for consumption or investment".

This isn't right. A deficient demand recession is when people are earning more money (from producing new goods/services) than they want to spend by purchasing NEWLY-PRODUCED real goods/services whether for consumption or investment.

And this can mean "the desire to increase your holdings of financial assets", but it can also mean the desire to accumulate real assets like land, or accumulate goods produced in the past like antique furniture.

So if an excess desire to accumulate financial assets can cause a recession, why can't an excess desire to accumulate land, or antique furniture, also cause a recession? In all three cases, some income from the production of new goods is desired to be diverted away from being spent on newly-produced goods.

I have explained why I believe that an excess desire to accumulate land and antiques cannot cause a recession, and I can make the same argument for financial assets (except the medium of exchange) too.

RSJ @12.59. But when the company that holds $100 in the money market fund withdraws $100 to spend it, what happens to the money market fund? It needs to raise $100 from somewhere else.

"why necessary to assume there's not a market clearing price for the money price of land? why can't there be a market clearing price for excess demand for existing assets? agree if you assume not, then the money alternative becomes the source of the glut - but why assume that? ... Agreed. We are on the same page"

conditionally on the same page, yes, but only if you assume no market clearing price - but still why do you assume not? why assume an excess demand for money due to a goods and services glut can't be cleared by pricing for existing assets? this holds the entire market for existing assets inflexible apart from a glut assumption about the market for goods and services - why? if sellers of existing assets like the price they're getting, there's no reason to assume the excess demand for money doesn't get absorbed

RSJ,

When a company borrows $100 and puts in into a money market fund, the fund must then do something with the investment it receives. Normally, it would purchase some security (commercial paper, treasuries, etc.) from another investor. And then that investor must do something with the money. The chain doesn't end for purposes of Aggregate Demand analysis until (1) someone increases their demand for money (2) someone spends it on goods and services.

This is why it is a bit frustrating to see Scott F's recent one-line response to Nick's question, which I thought finally had a chance at getting to the heart of a genuine non-semantic Rowe versus MMT difference after all this time. Normally the MMT camp applies so much operational fervor in showing how step 1 causes step 2. So Nick is asking how step 1 (an increase in demand for financial assets) turns into step 2 (decreased aggregate demand) without step 1a (an excess demand for money, as in the medium of exchange) happens in between.

Your MMF story is not different than where a Corporation borrows money and uses it to invest in a standard mutual fund. The mutual fund then must do something with the money (buy stocks) and so on. Only if the fund (so someone in the chain) decides to increase its actual money holdings does the chain stop and allow for a general glut, because money is special.

Horizontal expansions can go on a long time until debt structure/asset prices become unstable due to low ratio of vertical money (currency/reserves) to horizontal money (what Nick calls bank money or dlr may call money).


dlr wrote: "without step 1a (an excess demand for money, as in the medium of exchange)"

Perhaps the role of government is what you are looking for?


A small withdrawal of vertical money due to government surplus (excess taxation) then leads to deleveraging of the horizontal debt/asset structure as borrows scramble to payoff loans while asset prices collapse.

anonTDH: I am making the standard assumption that the price of goods is sticky in the short run.

Suppose we have an increased demand to save in the form of money (people desire to save part of their income and add a flow of money to accumulate a greater stock of real balances M/P). If the price of goods P falls, that solves the problem, by increasing M/P and satisfying the demand to hold more money. (Unless of course it causes expected deflation which exacerbates the problem by increasing the demand for money).

Is that what you were asking?

Suppose the price of new goods and services is sticky. Prices don't drop, people don't buy, which means they save. But instead of saving in the form of money, they buy existing assets, driving up their price... then there has been a glut of new goods and services, but people haven't saved in the form of money. Sellers of existing assets are happy with the higher price they receive, so you can't really blame the glut in new goods and services on their excess demand for money, can you? Also, those sellers haven't "saved"; they've just exchanged one asset for another ... e.g. a glut of new houses; people bid up the price of existing houses; no resulting saving in the form of excess demand for money.

anonTDH: But has there been a glut in new goods?

Take a simple example. Suppose people currently are holding their target value of financial assets (excluding money). Suppose they decide they are all going to live twice as long in retirement, and want to double their target value of financial assets. So each individual decides to save a little bit more, and slowly increase his financial assets over time. But if the price of financial assets responds instantly to excess demand, the price of financial assets relative to goods doubles instantly, to eliminate the excess demand for financial assets. Having hit their target immediately, they stop saving immediately, and the excess supply of goods disappears immediately. There was only an incipient excess supply of goods.

(All this assumes that the doubling of the prices of financial assets has no effect on the demand for money, of course.)

Nick @6:57

"But if the price of financial assets responds instantly to excess demand, the price of financial assets relative to goods doubles instantly"

No, it doesn't.

The price of financial assets is determined by estimates of future returns over a given time horizon. Prices will not adjust upward unless either prospects for overall returns increase, or unless all other prospects (e.g. investment in tangible capital, rental property, and finally the overall growth rate of the economy) diminish so that you are willing to accept a lower yield as you have no other options. Obviously in the short run, there are frictions and dislocations, cognitive biases, etc, but the price of financial assets is not determined by a desire to have your overall portfolio increase in value -- such a desire is infinite.

In terms of your model, you demonstrated that if everyone wants to hold a good, then transactions will stop. That is true. It holds for bonds, money, land, antiques, and anything else you want to accumulate. No one will succeed if everyone is on the same side of each trade, and everyone has the same list of second and third best preferences. They will all move down this list together at the same time. If they are trying to accumulate land -- everyone will just keep the land they already have and no one will get more. If they try to accumulate money -- everyone will just keep the money that they have and no one will earn any more. You go on down the list and the end result is that no transactions occur. But I don't think this is enlightening, or is an argument that can be used "against" my model.

re: new vs. pre-existing goods

I agree that only purchase of new output counts, but I don't think that purchase of pre-existing output leads to gluts. Whether someone buys a new or used car can lead prices to adjust between new and used cars, or between cars and real estate, but as long as they are spending all of their income on goods or tangible assets, then there will not be a glut.

The person who receives the proceeds will turn around and spend them on something else (again, we are assuming that they are not saving in the form of financial assets). It is only when they do not turn around spend but take that money out of the goods market can there be a glut. Whether they leave the money passively in a bank account, or whether they exchange their bank account for a bond is beside the point. They will allocate their financial assets in such a way as to maximize their returns, but regardless of that allocation, demand for output is decreased.

If this demand to earn more than you spend is not offset by people who spend more than they earn (by incurring liabilities), then the market will not clear and you will have a glut. This does assume that prices do not just adjust downward, but of course network effects and debt service prevent prices from adjusting downward. Contracts and cost of capital effects cause business to liquidate capital rather than instantly re-negotiate all debts/wages/vendor relationships/prices downward simultaneously by some all-knowing auctioneer.

RSJ: "No, it doesn't. The price of financial assets is determined by estimates of future returns over a given time horizon."

In my very simple model, where people wanted to double the value of financial assets they held, regardless of expected rates of return, the prices of financial assets would double. You are moving to a more complex, admittedly realistic model. But it doesn't affect my point. If an excess supply of goods is matched by an excess demand for financial assets, and the price of those assets is perfectly flexible, they adjust instantly to eliminate the excess demand for financial assets and the excess supply of goods (as long as there's no excess demand for money).

"In terms of your model, you demonstrated that if everyone wants to hold a good, then transactions will stop. ...... If they try to accumulate money -- everyone will just keep the money that they have and no one will earn any more. You go on down the list and the end result is that no transactions occur. But I don't think this is enlightening, or is an argument that can be used "against" my model."

Good. You have understood my main argument. I just need to convince you that it doesn't apply to money. Money, as medium of exchange, is weird. It's different.

All the other goods have a market of their own. Money doesn't have a market of it's own. Money (as medium of exchange) appears on one side of all the markets. In a monetary exchange economy, if there are n goods, including money, there are n-1 markets, one for each of the non-money goods. And money is bought and sold in each of those n-1 markets.

And money circulates. Every person has a flow of money coming in, and a flow of money going out, and so there are two ways to increase the stock of money you hold: increase the flow coming in; or reduce the flow going out. If one of those ways is blocked, because there's an excess demand for money, you just switch to the other way. If there's an excess demand for money, you can't sell more goods to increase the flow of money coming in; but you can buy fewer goods to reduce the flow of money going out. And that's what causes the recession.

Hi dlr, Nick @4:58

"When a company borrows $100 and puts in into a money market fund, the fund must then do something with the investment it receives."

In my example, a company sold a bond to an investor. The investor had $100 in a MM fund. The investor sells his MM holdings and buys the bond, and the company sells the bond and buys the MM holdings. So everything is closed. The MM fund assets are unchanged. Obviously many other transactions can happen, but these portfolio shifts are not important.

The company now has a liability (bond), and an asset (MM holdings). The investors in the economy do not have more or less financial assets then they started with. The only role of investors in this process was to price the bond in terms of their MM holdings.

Next, suppose that company sells the MM holdings to buy some capital equipment. You can assume the seller of the capital equipment receives MM holdings as proceeds of the sale. If you draw a line about the initial company and the rest of the private sector, when the company buys the capital equipment with the MM holdings, the net financial savings of the rest of the sector increased by $100. That is because the liability stayed with the company, but the asset jumped ship to the rest of the sector.

And this is the only way that financial savings can increase.

"This is why it is a bit frustrating to see Scott F's recent one-line response to Nick's question, which I thought finally had a chance at getting to the heart of a genuine non-semantic Rowe versus MMT difference after all this time."

OK, here is my shot at it:

My view:

Borrower creates asset and liability and spends the asset into the economy, causing someone else to end up with a financial asset. In terms of rates, a flow of borrowing causes a flow of financial asset accumulation, which is defined as surplus profits, or profits not re-invested in the goods/labor markets. The nominal yield does not equilibrate between the supply and demand for financial assets, but merely reflects risk-adjusted nominal growth expectations (although obviously the amount borrowed depends on changing growth prospects and risk tolerances).

My characterization of the orthodox view:

Businesses/Individuals choose to save in the form of financial assets. Interest rates equilibrate the amount borrowed with the amount saved, so that a businesses decides to save $100 and then keeps lowering the interest rate up until $100 is borrowed and spent, funding the $100 in savings ex-post.

I think, operationally, the orthodox view is not consistent with accounting, or observed economic behavior.

"Normally the MMT camp applies so much operational fervor in showing how step 1 causes step 2. So Nick is asking how step 1 (an increase in demand for financial assets) turns into step 2 (decreased aggregate demand) without step 1a (an excess demand for money, as in the medium of exchange) happens in between."

My point is that no one holds money as a stock, they only hold claims (e.g. deposits, savings accounts, bonds, money market funds, stocks). Money circulates to settle transactions, but there is no special desire to hold cash in your mattress. There *is* a special desire to obtain a *flow* of cash profits that are not re-invested in goods/labor.

But the terminus of this flow is some financial claim, whether a bank account, bond holding, etc, and the market prices of one claim vs. another is such that indifference is achieved between holding one or the other (once time horizons are taken into account).

RSJ @12.18: Kudos to you for a good attempt to explain the MMF transactions. But I'm afraid my brain is not up to following it. Just too many transactions to try to keep my head straight. Sorry.

But I do want to take issue with you on this bit:
"My point is that no one holds money as a stock, ..... Money circulates to settle transactions, but there is no special desire to hold cash in your mattress."

But they do. There is a stock of currency and demand deposits that people hold. I agree that money circulates, and that it's like an inventory. But at any given time there is a stock that can be seen in a snapshot. And people can change the stock that they desire to keep on average. (And if they held each dollar for some fixed period of time, and never changed their minds about how long they wanted to keep each dollar in their pockets, then that says desired velocity of circulation is fixed, and we get into a *very* monetarist MV=PT sort of model.)

" Every person has a flow of money coming in, and a flow of money going out, and so there are two ways to increase the stock of money you hold: increase the flow coming in; or reduce the flow going out. If one of those ways is blocked, because there's an excess demand for money, you just switch to the other way. If there's an excess demand for money, you can't sell more goods to increase the flow of money coming in; but you can buy fewer goods to reduce the flow of money going out. And that's what causes the recession."

I agree! And I would add the following caveat: Any retained earnings immediately become a financial asset. Remember that both currency and demand deposits are financial assets backed by financial liabilities (of the central bank and private banks, respectively).

If desire to hold financial assets is not matched by a growth of financial liabilities, then markets don't clear and you have a glut. I *think* we agree on this point. Moreover, we would both agree that if you have $10,000 in money market funds, and I have $10,000 in bonds, and Joe Potato has $10,000 under his mattresses, then it doesn't matter whether we all switch our holdings, in terms of purchases in the real economy. Moreover, if in aggregate, businesses shift their capital structure to fund themselves more with equity than debt, or more with bank debt than bonds, then none of these shifts in monetary aggregates affect demand or the real economy.

"But they do. There is a stock of currency and demand deposits that people hold."

I agree that both demand deposits and currency have transactional utility that causes people to hold them above and beyond the return provided. But this is just a buffer stock to help settle transactions -- it's a cache.

Who says to themselves "I want to spend $100 less than I make *and* store it as currency."?

My claim is they say "I want to save $100", and then there is a *separate* decision of whether to buy a stock or bond, or hold that money in a deposit account, or withdraw cash and keep it under your mattress.

And I claim that decision #1 affects the real economy whereas decision #2 does not -- at least for an economy with the current banking framework (e.g. the CB would add currency to the economy as people put more under their mattress, if such an event caused overnight rates to spike).

RSJ:

"If desire to hold financial assets is not matched by a growth of financial liabilities, then markets don't clear and you have a glut. I *think* we agree on this point."

No we disagree there (for financial liabilities that are no media of exchange). Either the price (rate of return) on those liabilities adjusts, eliminating the excess demand for both them and for newly-produced goods, or it doesn't, and people can't buy financial liabilities and have to buy something else with their income instead (or add to their stocks of money).

"Moreover, we would both agree that if you have $10,000 in money market funds, and I have $10,000 in bonds, and Joe Potato has $10,000 under his mattresses, then it doesn't matter whether we all switch our holdings, in terms of purchases in the real economy."

Yes, we agree.

"I agree that both demand deposits and currency have transactional utility that causes people to hold them above and beyond the return provided. But this is just a buffer stock to help settle transactions -- it's a cache."

Take out the word "just", and we agree.

"Who says to themselves "I want to spend $100 less than I make *and* store it as currency."? "

I did! As a baby I held no currency, and now I hold $100 on average. So I must have done this.

"My claim is they say "I want to save $100", and then there is a *separate* decision of whether to buy a stock or bond, or hold that money in a deposit account, or withdraw cash and keep it under your mattress."

My claim is that some economists may find it useful to think of there being two separate decisions, but this way of framing the problem is a consequence of the economist's theoretical perspective; it is not a fact about the world. We can take the time derivative of the second budget constraint, and substitute it into the first budget constraint, and get a unified budget constraint. But what both these approaches miss is the structure of markets in a monetary exchange economy. In an economy with n goods, including money, there are n-1 markets, and really n-1 separate decisions, because each of those n-1 decisions is to maximise utility subject to any quantity constraints in the *other* n-2 markets (Clower, Benassy).

You have a "2-decision" way of looking at the world; I have an "n-1 decision" way of looking at the world. And that is radically different. When markets are not clearing, then monetary exchange matters. If the transactions costs of barter were not (usually) prohibitive, then unemployed workers would swap the goods they produce directly with each other, and would circumvent the excess supplies of goods and get to full employment. There could never be deficient demand unemployment if barter were possible at low transactions costs.

"....-- at least for an economy with the current banking framework (e.g. the CB would add currency to the economy as people put more under their mattress, if such an event caused overnight rates to spike)."

Agreed, sort of. If the CB did add currency and/or demand deposits to eliminate any excess demand for money then we wouldn't get a glut of newly-produced goods. But I don't think the spike in the overnight rate would necessarily be a good indicator of whether or not there was an excess demand for money. Remember there are n-1 markets, and so n-1 separate excess demands for money, in principle.

Nick,
Nick,

We have grammar to clarify meaning. Yours does. I won't fix what ain't broke.

Ex-post, you are right about the uniqueness of money. But in a world with sticky pay rates and other prices, it is expectations that do the heavy hauling of adjusting PQ to MV. So ex-ante looks the way to think.

If I think there is a 0.001% chance of your converting your house into money, that belongs in my estimate of the expected MV. If I reckon that chance has increased to 0.002%, that is probably a large change in my estimate of expected MV.

Nick:

I haven't been following these comments.

I don't believe that interest rates on deposits will change to clear up excess supplies or demands for money.

I see no plausible market prices by which an excess supply of money will lead banks to raise deposit rates. Or an excess demand for money will cause them to lower rates.

The Black and Fama parts of BFH and some of Greenfield and Yeager's early arguments made that sort of false claim.

I think deposit interest rates are often sticky in the short run, which allows changes in the prices of nonmonetary assets to generate the liquidity effect. So, you get some kind of short run "equlibrium" with deviations of the market and natural interest rates.

On the other hand, if there aren't sticky, then I think deposit rates will move in proportion to other rates. That might happen in the short run and will happen in the long run. And so, no liquidity effect.

However, when that happens, there is no problem with a zero nominal bound.

That is why having a yield on them is important. It can be negative and so there is never a problem with the zero nominal bound.

The zero nominal bound is solely an artifact of currency. And deposits get tied up in it because of redeemability into currency.

Bill: "I don't believe that interest rates on deposits will change to clear up excess supplies or demands for money."

I'm glad to see your intuition is coming to the same answer as mine.

Nick:

I agree that not all the n-1 goods are perfectly substitutable -- I don't think this is a source of gluts. It's a source of badness which may or may not reach general glut badness levels.

"My claim is that some economists may find it useful to think of there being two separate decisions, but this way of framing the problem is a consequence of the economist's theoretical perspective; it is not a fact about the world"

It is a fact about the world, because yields on financial assets (and hence their price) are not determined by the quantity of borrowing demanded intersecting with the quantity of "lending" supplied, but by the expectation of return. This is a prediction market, and you cannot use the same supply and demand arguments to combine a prediction market with goods markets. There is real world economic behavior driving this distinction.

Prices of securities in liquid markets are not set by the marginal purchaser -- anyone can short a security if they think it is too expensive, and all cash-flows with similar return characteristics are perfectly substitutable: A second or third cash-flow with the same return characteristics is not priced lower than the first. Therefore the "supply" of lending is horizontal with respect to quantity demanded.

A good mental model is to imagine a line of borrowers standing in front of you, the lender. Each borrower asks "how much is my obligation worth?", and you say "X", trading X of your existing financial assets for the borrower's new issue. And the next borrower asks "how much is my obligation worth?" and you say "Y", swapping again. You as the lender can "lend" many multiples of your financial holdings in a single trading session, because each time you swap some of your existing assets for a new issue, you are enabling a borrower's balance sheet to expand, even as your own balance sheet is unchanged.

And you are not going to start increasing the yield if the line is long or short. The rate of turnover does not boost the price, but you will appraise each claim individually, regardless of the quantity borrowed.

This is unlike the Walrassian n-1 goods markets, in which the buyer and seller must come to an agreement on the *quantity* of goods purchased, with any excess demands resulting in purchases in some other market in which people also agree on quantity. The net result is a full overall agreement on quantity within the barter model.

When you introduce money, then there is *no* agreement on quantity in the money markets. The level of financial savings is a one-way decision set by borrowers. You can call this an "excess desire for money savings" if you want, but the key point is that it is excess of whatever the borrowers say it will be, and there is no reason to believe that the increase in liabilities will be sufficient to meet the financial savings needs of the private sector.

Nick, do you agree with what is said here about currency?

http://www.treas.gov/education/faq/currency/legal-tender.shtml

I thought this might interest JKH and others about capital requirements.

http://baselinescenario.com/2009/12/07/importance-of-regulatory-capital-requirements/

Nick's post said: "Too much: "When someone buys a house, does he/she get the mortgage and eventually and usually pay with a check (demand deposit)?" Yes.

"Think about an economy with all currency and NO currency denominated debt. What would it look like?" Horrible. No pension plans. But OFF-TOPIC!

"Is it more accurate to say "The commercial banks are the central bank's agents for creating currency denominated debt with an interest rate attached and repayment terms attached."?" No.

"Is it the central bank increases the supply of reserves, interest rates come down, and demand for currency denominated debt (a loan) goes up (they are hoping for that)?" *Quantity* demanded of loans goes up. Yes.

"With legal tender laws, is currency used to make the interest payments on currency denominated debt whether gov't (thru taxes) or private?" No. We usually pay by cheque.

Now, back off a little please. Let someone else attack me!"

If the builder redeposits the check, I don't see any currency created.

If there is an all currency economy, are there any currency denominated debt defaults? If no, are there any bad banks?

Could you expand on the no pension plans part?

I don't believe commercial banks can create currency, so what do they create?

In a lot of cases, the *Quantity* demanded of loans goes up. I think we need to discuss the demand side like when it might not go up.

Aside from some form of check default, is it really a good idea to make the interest payments on currency denominated debt with currency denominated debt? (I believe we agree that a check is currency denominated debt [see the treas.gov link/post])

@Bill, Nick
"'Bill: "I don't believe that interest rates on deposits will change to clear up excess supplies or demands for money."

I'm glad to see your intuition is coming to the same answer as mine.'"

I agree with this as well, the imbalance will continue as the information moves though the market. Eventually a catastrophic shift will happen to correct the imbalance. I believe this is what Hayek's trade cycle is.
Its a bit heterodox, but I don't think that markets are efficient, but that they are efficiency seeking.

@Too Much:
"Aside from some form of check default, is it really a good idea to make the interest payments on currency denominated debt with currency denominated debt?"

Yes; it allows you to not have to adjust for price changes due to supply side shocks in what ever you are making the interest payments in. For example say you have a silver denominated currency (like US dollars during bimetallism) suddenly new technology allows for much better silver mining and the price of silver plummets (Like it did). Now you have severe inflation.

"It ain't the loans what matter; it's the money creation that goes along with loan creation what matters. That's why (bad) banks matter. Bad banks won't create money."

So the major credit card issuers are bad banks, right? I suspected as much. ;)

I think that what the US government (and other governments that have followed in their footsteps) has been doing to do to attempt to "improve" the economy is very irresponsible and wreckless. It has wasted trillions of dollars bailing out creditors and shareholders of failed institutions with broken business models rather than addressing the structural flaws in the system of too much debt. And this is going to lead to massive problems down the road with regard to our currency and interest rates, in my opinion. And I think that the gold price breaking out to a new high is a strong indication of the reduction in faith and confidence that people have in governments and their fiat currencies. I recently read a good article on this at Gold Price Cheaper Now than at $300 - Hathaway that discusses the Federal Reserve's easy monetary policies in order to try to prevent any sort of deflation from occurring and to try to reflate assets prices. There are also many more articles here that I think are very helpful for any investor to read because they help to explain the investment implications for the dollar, the gold price, and gold mining companies who I believe will continue to benefit from central banks' inflationary programs.

Nick's post said: "Too much: "When someone buys a house, does he/she get the mortgage and eventually and usually pay with a check (demand deposit)?" Yes.

So if I can buy a house with all currency, all "check" (demand deposit and therefore currency denominated debt), or a combination of the two, are you sure that there are not n-2 markets?

Nick said: "We have a fall in AD, and current output and employment is demand-constrained, not supply constrained. We have a general glut."

I agree (believe it or not) about being demand-constrained.

That sounds like an output gap. If so, how should it filled?

RSJ said: "Who says to themselves "I want to spend $100 less than I make *and* store it as currency."?

My claim is they say "I want to save $100", and then there is a *separate* decision of whether to buy a stock or bond, or hold that money in a deposit account, or withdraw cash and keep it under your mattress."

I agree. People are trying to maximize the return on their savings.

"And I claim that decision #1 affects the real economy whereas decision #2 does not -- at least for an economy with the current banking framework (e.g. the CB would add currency to the economy as people put more under their mattress, if such an event caused overnight rates to spike)."

Add currency??? It seems to me the fed is about adding reserves NOT currency. Why is that?

Nick said: "Here's why I think the orthodox view is wrong. Suppose we start in equilibrium, then suddenly everyone wants to cut consumption of newly-produced goods and buy land."

What if the cut in consumption is forced by currency denominated debt defaults, the collateral seized, and the collateral resold as a used good?

Are the reselling of technology goods around 2001 and the reselling of housing good examples?

Doc merlin, I assume that is why there is a legal tender law and a need for an entity to limit currency printing.

Now if there was only a good way to limit currency denominated debt creation!?!

Hi Nick, I know you've moved on from the topic, but could you offer, a summary of what just went on here these past couple of months, maybe reference back to the "what makes a central bank a central bank."

Did anything happen? I've read a ton of words, some used by the same people with different meanings. People getting exercised. And all that sticks out was JKH saying something about viewing ownership as access, to real rents, disagreeing with winterspeak. You trying to convince the MMT's of shifting the curve to the left. And a whole lot of "reserves don't cause lending!" It's just that I'm going through my chemistry exams and realizing they would have been easier with more studying during the term. Economics is like falling through the rabbit hole, but I really appreciated you stating that economists could use whatever variables they like.

This is more a backhanded compliment then anything, thank you and the commenters for the knowledge, there is a lot of dense material.

Hi edeast: yes, it all got pretty dense!

If I were doing it again, I would change the title of my earlier post, because people (understandably) misunderstood the question I was addressing. Instead of "What makes central banks central?" I should have called it "What gives central banks the power to control monetary policy?" Or "Why is the Bank of Canada more powerful than the Bank of Montreal?" My answer would stay the same.

In this post, and the post on "capital, reserves, and loan officers", I would insist that "an increase in the supply of reserves (or loans)" means "a *shift* in the supply *curve* of reserves (or loans)". That is what economists (should) mean by "supply". Otherwise, I would stick to what I said originally. A shift in the supply curve of reserves does cause a shift in the supply of loans and of money. Except perhaps in special circumstances when the supply curve of bank capital is (locally) vertical, so banks are capital-constrained.

An awful lot of confusion is due to people thinking that "an increase in supply" means "an increase in quantity sold", as opposed to "a shift in the supply curve".

ok, thanks.

The comments to this entry are closed.

Search this site

  • Google

    WWW
    worthwhile.typepad.com
Blog powered by Typepad