Do sofas refute monetarism?

Start with a very monetarist model of the monetary system. Whatever that means to you.

Now let's add sofas to the model.

Suppose that every year1% (say) of the monetary base disappeared down the back of the sofa, never to be seen again. Would that refute monetarism?

I don't think so.

It would affect the supply of base money, because the monetary base would be declining at 1% per year, unless the central bank took offsetting action.

It would affect the demand for base money too, because people wouldn't want to hold as much base money if they knew that 1% of what they had in their pockets would be lost per year.

Sofas could definitely affect the equilibrium price level, and rate of inflation, via their effects on the supply and demand for base money, depending on whether the central bank took offsetting action. But I don't see why sofas would refute monetarism.

Now lets make the sofas magic, and give the central bank a magic wand, so it can control the sofas. By waving its magic wand, the central bank can make 2%, or 3%, or whatever%, disappear every year down the back of the sofa, never to be seen again.

That magic wand is a powerful new instrument for the central bank. Before, it could only affect the supply of the monetary base. Now it can affect the demand too. If it wanted to increase the price level, it could wave its wand and let people know that 2% rather than 1% of the monetary base would be disappearing down the back of the sofa every year until further notice. Even if the central bank kept the supply of base money constant, by printing more to offset what gets lost down the back of the sofa, waving the wand to increase the percentage from 1% to 2% would reduce the demand for base money and cause people to try to spend it before the sofa got it. Waving the wand would make base money a hot potato.

It's still a very monetarist model.

Now let's make the sofas, and the wand that controls them, even more magic. Suppose the wand can make negative percentages of base money disappear down the back of the sofa.

So what? I can't see any qualitative difference between waving the wand to cut the sofas' take from 2% to 1%, from 1% to 0%, from 0% to -1%, from -1% to -2%, etc.

It's still a very monetarist model.

There is no theoretical difference between sofas and paying interest on base money. A sofa that takes 1% of the monetary base every year is like the central bank paying minus 1% per year interest on the monetary base.

(There is no theoretical difference between sofas and expected inflation either.)

Would it make much difference if one part of the monetary base (coins) could get lost down the sofa, but another part (notes) could not? Or if part of the monetary base were electrons, rather than paper or metal? And if all got lost in different percentages down different sorts of sofas? Or if metal, paper, and electrons, could all be traded at fixed exchange rates at the central bank?

For some empirical estimates of how much money disappears down the backs of sofas see JP Koning's post on his work with Mike Sproul.

I wrote this post partly in response to Steve Randy Waldman's and Steve Williamson's posts. Paul Krugman and Scott Sumner have responses too.

(Sorry for not writing a more comprehensive response. But I have already written posts (e.g. like this one) criticising the idea that the stock of money is demand-determined (and my views on that question are too extreme to be orthodox, because I think that M=Ms=/=Md most of the time). And I'm a bit snowed under at the moment. So I'm sticking to this one point.)

Comments

[Sina Motamedi]

Very good.

[Peter N]

This link that I posted in your previous entry is about some of the effects of negative real interest rates.

http://ftalphaville.ft.com/2013/01/15/1334382/commodity-volatility-where-art-thou/

What does a manufacturer do when it's more profitable to store the output and sell it later than it is to sell it now? Are we seeing this now in commodities?

[Nathan Tankus]

do you have any evidence that the amount of "money" in reserve accounts at the fed doesn't change endogenously to the inelastic demand for reserves by banks?

[Frances Woolley]

Nick - nice post - glad to see admin work has not yet entirely destroyed your brain cells, even if you are snowed under.

[Ritwik]

And when we combine nominal interest (on reserves, deposits) with inflation(factor in money demand), we get....real interest rates!

Monetary policy is really really about setting real interest rates. :)

[Nick Rowe]

Ritwik: do sofas set interest rates?

Nathan: "endogenous" means "depends on stuff". The quantity of reserve accounts at the Fed depends on stuff. The quantity of any money depends on stuff. What precisely that stuff is depends on what the central bank is targeting. That's not what the argument is about.

When you borrow $10k from your bank to buy a car, and your bank creates an extra $10k in deposits, does that mean you (or anybody else) *wants to hold* an extra $10k in your chequeing accounts? That's what the (my) argument is about. I say it doesn't. I say that if we started with equilibrium M=Ms=Md, and then you borrowed $10k to buy a car, we are now in disequilibrium and M=Ms > Md. (Which means I am waaay more heterodox than you ;-) )

Frank: storing newly-produced goods is a form of investment. The lower are real interest rates, the greater the demand for investment, including storage, other things equal.

Sina, Frances: thanks!

[Ritwik]

Nick

A sofa that creates or hides base money? Yes.

[marris]

> Before, it could only affect the supply of the monetary base. Now it can affect the demand too. If it wanted to increase the price level, it could wave its wand and let people know that 2% rather than 1% of the monetary base would be disappearing down the back of the sofa every year until further notice.

Can someone explain each sentence here?

Can I think about this in micro terms? For example, the CB raising the disappear rate from 1% to 3% is like me more quickly eating the apples I buy. So I either go to the store more often (increase velocity) or buy bigger batches when I go (increase transaction price).

Or do I need to draw AS / AD curves to see this?

Is the CB's main goal to increase transaction demand for money? Reduce store-of-value demand for money? Or both?

[Nick Rowe]

marris: think of owning money as like owning a consumer durable. If the depreciation rate on the consumer durable increased from 1% to 3% your demand to hold that consumer durable would fall. You would want to own a smaller stock of refrigerators.

The big difference between money and refrigerators is that I would probably want to own a refrigerator even if I were never allowed to sell it again. I wouldn't want to own dollar bills if I could never sell them again. That's why it doesn't make sense to distinguish between the transactions and store of value demand for money.

[Max]

"When you borrow $10k from your bank to buy a car, and your bank creates an extra $10k in deposits, does that mean you (or anybody else) *wants to hold* an extra $10k in your chequeing accounts?"

Would you object to this generalization?

When you borrow $10k from your bank to buy a car, and your bank creates an extra $10k in financial assets, does that mean you (or anybody else) *wants to hold* an extra $10k of financial assets in your portfolio?

I mean, you don't think the disequilibrium depends on how the bank finances the loan (whether via deposit, or bond, or equity, or some least-cost combination), right?

[RebelEconomist]

Totally off-topic, but I sympathise about your admin workload too (as I did your marking). Again, I make the observation that economists (I think) don't follow their own advice, which I expect would be that the most efficient solution to allocating such work is to unbundle it and price it so that those who are most willing to put up with it do it. But I suspect that for "political" reasons, the university authorities would not allow that - ie an astronomical price would reveal the true cost of getting academics to do admin!

[Josh]

Nick,

Is the sofa a black hole or is it a storage technology? I think this example works if it is the former, but not the latter. For example, if the couch is a storage technology, which was where I originally thought you were going with this post, then the demand for money is unchanged.

[errorr]

Change fed to govt and sofa to tax and you can see the MMT point about how the distinction between fiscal and monwtary policy is artificial.

[Frances Woolley]

RebelEconomist "price it so that those who are most willing to put up with it do it."

Two words: Adverse selection.

Josh - or mechanism for transfering household resources from sofa-sitters to sofa-cleaners?

Following on with marris's point about thinking about this in micro terms - wouldn't money eating sofas tend to decrease labour supply? This would be a consequence of expected inflation, too, but the impact is (too me) more obvious with sofas.

[Peter N]

" I say that if we started with equilibrium M=Ms=Md, and then you borrowed $10k to buy a car, we are now in disequilibrium and M=Ms > Md. (Which means I am waaay more heterodox than you ;-) )"

Steve Keen's endogenous money.

Keen has gotten together with someone you may know, M. Grasselli (sharcnet chair in financial mathematics, mathematics and statistics at McMaster)and reconciled his presentation to orthodox terminology. In this lecture he derives his equation from standard principles and shows it's a necessary result of endogenous money. The math is reasonably simple, and if there's a hole in it I (and presumably Grasselli) can't find it.

http://www.youtube.com/watch?v=UzxQcTOs4JA

It's followed by an interesting presentation by Grasselli on austerity. Using the model, he shows the model economy has 2 equilibria with different basins of attraction. He then shows how austerity affects the basins adversely.

http://www.youtube.com/watch?v=_qcXR5P3rck&playnext=1&list=PLqs7-zw9kiAJu9rsKB35f3cZWg6aq24V-&feature=results_video

[JP Koning]

Nick, your sofa reminds me of Mankiw's Gessell-style tax idea from a few years back, in which notes with serial numbers ending in a random number are destroyed. You can flip it around and gift those who hold notes ending in a random serial number with an extra note. Anyways, I agree with your point that there is no difference between the sofa and IOR. In both cases you're just manipulating the expected return on central-bank issued instruments. Decrease the return on any asset and people will want to spend it away hot potato-like until its price falls to the point at which it once again promises a good return.

[Andy Harless]

Sofas can and do refute monetarism -- at least if by sofas you mean magic sofas controlled by the central bank, and by monetarism you mean the proposition that the stock of base money is the primary determinant of aggregate nominal income (or of the price level). A world where the central bank operates by manipulating the demand for money, rather than the supply, is a very different world than the one people like Milton Friedman contemplated. In the end it's all semantics. I don't think Paul Krugman and Steve Randy Waldman have any substantive disagreement (other than perhaps disagreeing with misinterpretations of what each other said). But it is interesting that we may now live in a world where the Fed's decision to undertake an OMO will almost always be taken for microeconomic reasons rather than macroeconomic ones, the only macroeconomic implications being a confirmation of an earlier IOR decision.

[Nathan Tankus]

Nick: when you take any finite period of time there is always a certain amount of people or entities who will be holding inventories or money balances that doesn't conform to their "target" money balance of "target" money balance ratio. this is a definitional part of accounting periods. now back to what we're talking about: reserve accounts held at the fed is a much narrower category then the money supply. reserve balances are used to settle payments between entities that have accounts at the fed or are the fed (foreign governments, the treasury, banks etc) and to settle reserve requirements. since you're canadian (and i'm in canada!) I'll just talk about the simpler canadian case of no reserve requirements. thus the demand for reserves is the amount of reserves required to settle payments. the primary responsibility of central banks is to preserve the integrity of the payments system. since all reserves are needed for is settling payments the demand for reserves is pretty inelastic (it depends on the accounting period. contemporaneous accounting tends to produce very inelastic demand while the more common lagged reserve accounting makes the demand for reserves more elastic until the end of the accounting period). if the banking system as a whole is short the reserves it needs the bank rate will rise until there is a central bank intervention or borrowing from the discount window becomes a more attractive option. the central bank would have a difficult time targeting anything else because if they don't provide banks the ability to acquire reserves to settle payments the banking system would likely collapse. they may choose to have a policy of raising (lowering) their interest rate target if money supply growth is growing at a rate they deem unacceptable (which is what the supposed money supply targets really were), but ultimately they must provide sufficient reserves for settling payments.

one of the major confusions in this post is the conflation of reserves and money. reserves are a form of money, but they are not all money. reserves are relevant to the central bank, the treasury,foreign governments and banks. all banks could have adequate reserves to make payments but the "money supply" (the one relevant to households and firms) could be extremely volatile. some relationship between the money supply and reserves is created by reserve requirements but that is an accounting relationship more then anything else.

ex ante money supply is undefined. as you yourself point out "When you borrow $10k from your bank to buy a car, and your bank creates an extra $10k in deposits". ex post money supply always equals money demanded the same way ex post goods and services supplied equals goods and services demanded.

[marris]

Nick: I think I understand. Me "eating apples faster" example does not capture the "decay" aspect because I still get nourishment from each eaten apple. In your sofa example, my money does not get magically swapped with stuff (or happiness) at 1% every month. It just disappears. If I want to capture that with apples, I should probably think about apples which spoil quickly. Instead of buying 7 apples once a week, I may now buy 3 apples every 3 days. My steady-state fruit basket would have fewer apples.

Frances: Is labor supply micro? :-) I'm not sure what happens to the labor supply. I think there is more than one plausible story.

Story 1: Alice chooses between leisure (direct consumption) and work->income->ipad+iphone+... (indirect consumption). Money decay means less money, which means fewer goods. So her choices are now leisure and work->income->iphone+... Since the work choice no longer offers an ipad, it is less attractive. She chooses more leisure.

Story 2: Bob always keeps $10K in the bank in case of emergency. Every pay period, he allocates his income to buy consumer goods and producer goods. Bob may now decide to save only $5K (less savings will be lost to decay). He increases his spending by $5K (reduces his cash balance). The hot potato probably comes in here. I guess he continues to spend extra each month until prices have risen to the point that a higher balance, maybe $7K, is required for emergencies.

[rsj]

We have a word for when the government takes people's money away -- taxation. That is fiscal policy, not monetary policy. And what you are proposing is a wealth tax, that would discourage hoarding and encourage spending. Not sure whether something like that would actually work, but if it does, it would need to be targeted based on the wealth level, as poor people, at a minimum, would spend less in response to being taxed more.

[Nick Rowe]

Sorry for not responding to comments. But I am reading them, honest! And I thank you for them. My brain is fried, from too many meetings, and dealing with "Alleged instructional offences", and similar stuff.

Josh: I was assuming that when money disappears down the sofa it's gone for good.

Andy: "A world where the central bank operates by manipulating the demand for money, rather than the supply, is a very different world than the one people like Milton Friedman contemplated."

I'm not sure it's theoretically different. Milton Friedman talked about the inflation tax, which theoretically is just like negative interest on money.

rsj, errorr: paying interest on base money has fiscal consequences. So does the inflation tax through printing money. Running a central bank, where there is a spread between the interest rate on the bank's liabilities and assets, is a profitable business. But those profits are small compared to other sources of tax revenue (for low inflation rates). The Bank of Canada earns around $3billion per year, IIRC.

Max: "Would you object to this generalization? When you borrow $10k from your bank to buy a car, and your bank creates an extra $10k in financial assets, does that mean you (or anybody else) *wants to hold* an extra $10k of financial assets in your portfolio?"

You lost me there. If I want to borrow $10k from the bank to buy a car, I don't want to borrow "financial assets". I want to borrow *money*, so I can buy the car with that money.

Nathan: "when you take any finite period of time there is always a certain amount of people or entities who will be holding inventories or money balances that doesn't conform to their "target" money balance of "target" money balance ratio. this is a definitional part of accounting periods."

Let me try this: When we talk about the "supply and demand" for cars that *means* something very different if we are talking about a car dealer, who buys and sells cars and holds an inventory, than if we are talking about regular people who want a car to drive. And when we talk about the "supply and demand" for the medium of exchange, where everyone is a dealer who buys and sells money and holds an inventory, that means something very different from the demand and supply of all other assets. And there is something very wrong with any economic theory which treats the demand and supply of money just like the demand and supply of any other asset. Yet such a wide spectrum of economists, from very "orthodox" to very "heterodox", seem to me to do just that, when they talk about the quantity of money being "demand-determined" when the central bank sets a rate of interest.

"reserve accounts held at the fed is a much narrower category then the money supply."

Yes. I know that. This post is about base money.

JP: agreed. We are on the same page.

Rebel and Frances: I think Frances' "adverse selection" answer might be part of it. But it's not obvious to me that "volunteers" who do it for little or no money might be subject to a similar adverse selection problem: only the power-hungry or crazy would say "yes"! (Umm, present company excepted, of course!) I'm not at all sure our motives for saying "yes" are in line with any vaguely standard view of rationality. University admin puzzles me more than monetary theory does.

[Nathan Tankus]

Nick: this is where the phrase "money" gets everyone mixed up. let's be very specific. the supply of reserves is indefinite ex ante but ex post conforms to the inelastic demand of banks for reserves to meet reserve requirements and settle payments. the supply of bank money is undefined ex ante and is determined ex post by the meeting of the demand for credit and the borrower quality assessments of financial institutions. this will always leave a "fringe of unsatisfied borrowers" (keynes's words).

[Nathan Tankus]

also: supply and demand in bank reserve "market" is different from most other markets: what other market exists where the "product" has a cost of production of zero and an actor is able and willing (indeed must) meet the demand.

[Max]

"You lost me there. If I want to borrow $10k from the bank to buy a car, I don't want to borrow "financial assets". I want to borrow *money*, so I can buy the car with that money."

Yes, sorry. The point I was trying to make was the $10K doesn't have to remain in the banking system as a deposit. Some of it can buy bank bonds and equity, so that everyone remains satisfied with the mix of deposits/bonds/equity. In other words, the creation of $10K of money has nothing to do with how the loan is financed. So the question is, does a money-financed loan create more of a disequilibrium than a bond-financed loan or an equity-financed loan?

[Nick Rowe]

Max and Nathan: maybe my new post says more clearly what I am trying to say.

[Bubbles and Busts: Does the Permanent Floor Affect the Inflationary Effects of the Platinum Coin?]

[...] David Beckworth, Peter Dorman, Nick Rowe and Stephen Williamson (see here, here, and here) also share their thoughts. [...]

[online forex]

Greeting to www.forex.cd- Your own premium information of forex information and reviews.
online forex http://www.forex.cd/