What's wrong with New Keynesian macroeconomics -- an MM perspective

There are many things right with New Keynesian macroeconomics. It is a very good synthesis of many strands in Monetarist, Keynesian, and New Classical macroeconomics. But this post is not about what's right with New Keynesian macroeconomics.

There are also many things wrong with current New Keynesian macroeconomics. New Keynesian macroeconomists are aware of many of those things. But this post is not about all the things that are wrong. It's about some of the things that are wrong. I'm coming at this from a "Market Monetarist" (aka Quasi Monetarist) perspective.

"New Keynesian macroeconomics" can mean a lot of things. What I have in mind here might better be called "Neo-Wicksellian" macroeconomics, because monetary policy is seen as interest rate policy.

1. New Keynesian macroeconomics makes no sense whatsoever in a barter economy. And yet nowhere is the assumption of monetary exchange made explicit. You can't see money in the model, but it must be there somewhere, or the model just wouldn't make any sense.

1a. The producers in New Keynesian models are imperfectly competitive. They choose to set price above marginal cost. If two producers have price above marginal cost, both could gain by a barter deal in which they both produce additional output and exchange it at the ratio given by their posted prices.

1b. Even if we replace imperfect with perfect competition, and assume prices are fixed, it would be impossible to have a recession with excess supply of all goods in a barter economy. Two producers who wanted to sell more but couldn't find customers would simply produce additional output and do a barter deal.

There is something fundamentally wrong with a model that implicitly assumes monetary exchange but does not make this assumption explicit, and does not even have money in the model.

2. New Keynesian models explicitly assume that money is the medium of account. But again, there is no money in the model. Prices are set in terms of a good that does not exist.

3. New Keynesian models explicitly assume that the nominal rate of interest is set by a central bank. Central banks have special power over monetary policy only because they issue irredeemable money. Yet that irredeemable money does not appear in the model.

4. New Keynesian models have only one transmission mechanism for monetary policy -- the effect of current and expected future interest rates on desired consumption and investment.

4a. We know that a permanent change in the stock of central bank money will change the long run equilibrium values of all nominal variables but will leave interest rates unchanged.

4b. We know that monetary policy will still affect nominal variables in the long run, and real variables in the short run (assuming sticky prices), even in a world with no interest rates, no borrowing or lending, or where interest rates are set by law. So the interest rate channel cannot be the only possible transmission mechanism for monetary policy.

4c. We can imagine a world in which central banks implement monetary policy in many different ways than by setting a rate of interest. For example, they can set the price of a commodity like gold, and vary that price. Or they can give away money for free, and vary the quantity they give away.

4d. Central banks that issue irredeemable money aren't even banks. A bank both borrows and lends. It has assets and liabilities. An irredeemable liability is not a liability, so central banks don't have liabilities. They don't even need assets, unless they choose to buy back some of their money in exchange for those assets. A central bank, since it is not a bank, does not have to borrow or lend, and need not have anything to do with interest rates.

5. New Keynesian macroeconomics asserts that central banks must lower current or expected future real interest rates in order to get an economy out of a recession. This assertion is, I believe, often false. The IS curve will probably slope upwards in a recession. Recovery from a recession is compatible with an increase in real interest rates.

6. This to me is the killer. New Keynesian models lead good economists, who correctly diagnose the monetary nature of the recession, at the same time to believe that monetary policy is powerless at the zero lower bound. And recommend fiscal policy instead. This is like correctly diagnosing magneto trouble, then recommending we all get out and push the car, rather than fixing the magneto. I just refuse to accept that that's the best we can do. We need to understand that monetary magneto better, and learn how to fix it. And it is my frustration with this lack of correspondence between diagnosis and policy prescription that has lead me on my three year search for something better.

I have been eclectic in my search, taking ideas that seem useful regardless of their source. Some are Monetarist. Some are Old Keynesian. And Clower is as much a Keynesian as a Monetarist. I side with Silvio Gessell against Keynes on the role of money in general gluts.

Like other "Market Monetarists" I think that monetary policy can cure what is at root a monetary problem. But we don't all agree on everything. Lars Christensen provides a good survey (pdf). This post is in part a response to Arash Molavi Vassei. Scott Sumner responds here. Josh Hendrickson here.

All equilibrium theories have a disequilibrium story on the side. If the demand curve for apples shifts right, that creates excess demand for apples at the old equilibrium price, so individual sellers can raise their prices above other sellers' prices and still sell their apples, and this process is what gets the price to the new equilibrium. In monetary economics we call this disequilibrium story the monetary policy transmission mechanism. The interest rate transmission mechanism is the New Keynesian disequilibrium story. It's not the only possible story. It's not even a very good story, as I argue above.

My MX6 developed "magneto" trouble last Summer, 100kms away from home. Replacing the alternator is a 2 hour job, and I didn't want to do it at the side of the road. So I bought a new battery at Canadian Tire, replaced the old battery when it finally died, and that got me home. I replaced the alternator the next morning. There are circumstances when a bodge job is the best you can do. But it's not really satisfactory.

Comments

[Max]

I prefer the term Monetary Policy Zealot (MPZ). :-)

A general remark about policy ideas. MPZs seem to be driven by the desire to find some intricate, indirect, bank shot solutions. But the more remote the connection between policy lever and what you are trying to control, the less likely you are to be successful, and the harder it is to argue that if the policy doesn't seem to be working, we just need to do more of it (alternative hypothesis: the policy is ineffective).

[Min]

Let me raise what is mainly a question of rhetoric. Why is at least some fiscal policy not monetary policy?

Nice Rowe: "New Keynesian models lead good economists, who correctly diagnose the monetary nature of the recession, at the same time to believe that monetary policy is powerless at the zero lower bound. And recommend fiscal policy instead. This is like correctly diagnosing magneto trouble, then recommending we all get out and push the car, rather than fixing the magneto. I just refuse to accept that that's the best we can do. We need to understand that monetary magneto better, and learn how to fix it."

I am not addressing the question of whether monetary policy is powerless at the zero lower bound.

Suppose that hoarding is the problem, or the manifestation of the problem. That is, there is not enough money **in circulation**, even though there may be enough money in the economy. Too much money is being used as a store of value, rather than as a medium of exchange. In that case, is not a solution, if perhaps a partial or imperfect solution, to put money into circulation? And would such a solution not be a monetary solution? Even if it is called a fiscal solution?

As an illustration, not a recommendation, what if newly printed money were distributed to the poor, who would spend it? Or to local gov'ts, which would also spend it? That is called fiscal policy, but is it not also monetary? (I have other illustrations in mind, but they would get more deeply into economics.)

As for searching for the best solution, I applaud that as an intellectual exercise, as well as one that may have practical payoffs. However, under our current circumstances, I hear the sailor's cry, "Any port in a storm!" :)

[J.V. Dubois]

Min: I think it may not be the same. Look at this older post from Nick Rowe: http://worthwhile.typepad.com/worthwhile_canadian_initi/2011/09/a-response-to-paul-krugman.html What I get from this article is, that there is one line of difference between good and bad response to a recession, no matter if you call it fiscal or monetary. And that is its reversibility.

If you spend newly printed money buying assets that cannot be sold back to private sector, such as giving it to the poor, or building a spaceship for Mars mission (given that such projects would not be started in ordinary times), than this has real impact on the long-term budget constrain http://worthwhile.typepad.com/worthwhile_canadian_initi/2011/04/functional-finance-vs-the-long-run-government-budget-constraint.html either in form of higher public debt and/or inflation.

On the other hand if you use newly printed money assets that can be easily sold back to people (such as buyng stocks, corporate bonds, land or even foreign currency) , then impact on public position is very small. I think this has something to do with how you view the position of the government in an economy. Now there is a free lunch around and government may decide what to do with it. It can either eat it or it can give it to private sector. So if you think that government is more efficient, then by all means lets build that Mars spaceship. But if you think that markets are doing better jobs at allocating resources, then it makes more sense to heed Nick's advice for non-standard monetary policy.

So the point is that monetary polices enables us to smooth the business cycle in a way which does not distort broader political debate - such as the role of the government in economy. Let's politicians do this things based on the people's will, we as macroeconomists will just deliver as stable environment for this debate as possible.

[anon]

Min, I disagree with J.V. Dubois. I think that fiscal policy can be seen as "monetary" in that it's an attempt to decrease demand for money balances. Informally, that can be described as "increasing money in circulation".

Incidentally, this means that the most effective kind of fiscal policy is subsidizing expenditure (investment plus perhaps consumption of final goods), not increasing the government deficit.

[Bill Woolsey]

Dubois:

You very much express my views.

The monetary authority's job is to maintain monetary equilibrium. In my view, this is best understood as slow, steady growth in nominal expenditure on output.

Whether the government should provide more public goods or redistribute income funded by present or future taxes, or fund the current level of expenditure by present or future taxes should be determined on other criterion than maintaining monetary equilibrium.

[K]

"Yet that irredeemable money does not appear in the model."

Well that's OK, since it isn't there. The Bank of Canada keeps about $25M of reserves in the system just to smooth the settlement process. The power to control rates comes from the threat of leaving excess reserves in the system or leaving negative reserves in the system overnight. But except for rounding errors, temporary settlement system bugs and trader mistakes, it doesn't happen. Where is all this irredeemable money of which you speak?

Go to the Bank of Canada site and look for yourself. There is no conceivable relationship between the quantity of reserves and the quantity of the medium of exchange for goods and services.

[Bill Woolsey]

K:

I visited Canada last summer. They had hand-to-hand currency there. Or was I wrong?

[Min]

J. V. DuBois: "But if you think that markets are doing better jobs at allocating resources"

I do not want to get too far afield, but I happen to think that there are suboptimal economic equilibria (or quasi-equilibria in a chaotic system). If they exist, that would mean that you can have long periods of time during which markets do not do such a good job of allocating resources, no?

[K]

Nick: "We know that a permanent change in the stock of central bank money will change the long run equilibrium values of all nominal variables but will leave interest rates unchanged."

You *can't* change the stock of reserves. Add excess reserves: rates go to zero. Leave negative excess reserves: rates go to infinity. Immediately. I've said this before, but you haven't replied. I think you don't believe me.

"New Keynesian models lead good economists, who correctly diagnose the monetary nature of the recession, at the same time to believe that monetary policy is powerless at the zero lower bound."

*Monetary* policy, I.e. swapping reserves for T-Bills *is* powerless because banks are *totally* indifferent to holding reserves or holding T-Bills. Swapping T-Bills for risk assets, on the other hand, may do stuff. But it ain't monetary policy as there is no money money involved. And if it's not monetary policy, then I guess it's fiscal/industrial policy for the financial sector.

Bill: We do! But those deposits are fully redeemable at the option of the bearer. So they can't be used for monetary policy.

[Nick Rowe]

Min: "But the more remote the connection between policy lever and what you are trying to control, the less likely you are to be successful,..."

But if the underlying problem is an excess demand for the medium of exchange, monetary policy does seem to be closer to the problem.

"...and the harder it is to argue that if the policy doesn't seem to be working, we just need to do more of it (alternative hypothesis: the policy is ineffective)."

Fair point, but one that could also be made against fiscal policy.

Min: "That is called fiscal policy, but is it not also monetary?"

Yes, it's both. As JV says, I see one problem as the reversibility. It's a lot easier to do a helicopter operation than a vacuum cleaner operation.

anon: yes, fiscal policy can be seen as "monetary policy by other means" -- raising velocity (decreasing the demand for money. But, as Bill says, fiscal policy is also supposed to be evaluated by other, microeconomic, criteria.

Bill. Yes. We need to remember though, if real interest rates on government bonds are low, that does increase the NPV of government investments, under standard cost-benefit analysis, so more government debt-financed projects ought to be undertaken on purely micro public finance grounds. Keynesians like Brad DeLong who argue for deficit spending on those grounds are surely right. (I made the same argument myself a couple of years back, in an old post.)

http://worthwhile.typepad.com/worthwhile_canadian_initi/2009/06/monetary-and-fiscal-policy-ought-normally-move-together.html

K: there is a lot more to monetary policy than swapping reserves and Tbills today. In fact, that is a very minor part of monetary policy. The most important part of monetary policy is strategy, not tactics. What do we communicate we are going to be targeting over the next decades?

"Bill: We do! But those deposits are fully redeemable at the option of the bearer. So they can't be used for monetary policy."

Bill's talking about currency. It's irredeemable. (Except via indirect convertibility, because the Bank of Canada *chooses* to target CPI inflation at 2%.)

[K]

Nick: "Bill's talking about currency. It's irredeemable"

No, it isn't. I was using the word "deposit" to make the point that paper money is a demand deposit of the CB, which it is. Lets say that Canadians no longer want to hold the $50Bn, or whatever, of paper currency that they currently hold, because they decide that some other means of payment (credit card, debit card, whatever) is better. They take that money and deposit it at the commercial banks, who then take it and exchange it at the CB for reserves, which they then *promptly* dump back to the CB through the OMO's that the CB will *have to conduct* unless they want the policy rate to drop to zero. The supply of paper money is *not* at the discretion of the CB (though they can affect the *demand* for paper money by changing interest rates).

"The most important part of monetary policy is strategy, not tactics. What do we communicate we are going to be targeting over the next decades?"

I think you are skating. We are talking about the failings of the NK perspective on the economy. You'll certainly get no disagreement from NK economists that the target could matter and there are lots of potentially interesting ways to frame it. Woodford, for one, seems very open to NGDP level targeting. If this is all we are talking about then I think we can stop and go have a beer right now. This is simply not an "MM vs NK" issue.

So lets not mix up what we are targeting with *what we are going to do if the economy doesn't oblige our wishes*. The CB needs real tools to effect it's target. Why is anybody going to listen to them pontificating about how things are going to be if they are utterly powerless to do anything about it if things don't turn out as they say. If they have no real tools in the present they can't create expectations of what they might do in the future either. They are just another think tank (though an exceptionally pompous one - The Holy Grail's Black Knight comes to mind).

The policy interest rate *is* such a tool. And so is the exchange of riskless securities, for risky ones (though not a *monetary* tool). But the supply of reserves is *not*. And the supply of paper money isn't either. Where's the disagreement?

[Nick Rowe]

K: I can "redeem" a $20 BoC note for two $10 BoC notes. But that doesn't count. Some people (chartered banks) can "redeem" a $20 BoC note for $20 BoC reserves. I would say that doesn't count either.

The only thing a central bank can ever really do is change its balance sheet, and make commitments to change its balance sheet in future. (That statement is commonplace among BoC people I have heard speak.) And the only part of its balance sheet that is special, that differentiates it from other banks, or from you and me, is the irredeemable "liabilities". Any talk about interest rates, even if true, is true only as a consequence of those two underlying hard facts. Those two facts are the ultimate tool. And that has always been true.

[wh10]

Nick,

I'd like to see you address -

"Nick: "We know that a permanent change in the stock of central bank money will change the long run equilibrium values of all nominal variables but will leave interest rates unchanged."

You *can't* change the stock of reserves. Add excess reserves: rates go to zero. Leave negative excess reserves: rates go to infinity. Immediately. I've said this before, but you haven't replied. I think you don't believe me."

Thanks!

[Nick Rowe]

wh10: this is absolutely bog-standard monetary theory. Start in one equilibrium, with reserves of %R and currency of $C, and prices at $P (P is a vector). There exists a second equilibrium with reserves $2R, currency $2C, and prices $2P. (All nominal variables (measured in $ units) are doubled, and all real variables (not measured in $ units) unchanged. The units of the rate of interest are 1/time. This is called the Neutrality of Money/Quantity Theory of Money (not to be confused with MV=PT, which is sometimes also called the "Quantity Theory". New Keynesian macroeconomists agree with me on this point.

[rsj]

OK, for the audience, where in the above does Nick pull in a fast one, trying to nullify the dynamic argument K made with an argument only applicable to constant economies, via a faith-based appeal to *long run* hand-waving?

[Nick Rowe]

rsj: less snark please. If you disagree, then say so.

[rsj]

Or to heap even more ridicule, as it is well known that there is an equilibrium relationship between the height of the population and the height of their dwellings, given by D = 4/3H, and as the latter is a homogenous equation, it must be the case that we can cause the population to grow by building taller buildings.

After all, the effect of the modeler's choice of units when describing a static equilibrium must be the same as the effect of an economic agent's adjustment of quantities within the dynamic equilibrium.

[Nick Rowe]

rsj: I will let that comment stand, only because you may not have read my previous comment before posting it.

Now, read what I said in the post, about the monetary policy transmission mechanism. "All equilibrium theories have a disequilibrium story on the side......"

[Bill Woolsey]

I think open market operations are monetary policy regardless of the risk chacteristics of the assets purchased. The notion that it is only "monetary policy" if the riskless assets seems confused.

I will grant that having the Treasury shift the composition of its debt might impact nominal expenditure in the economy "ceteris paribus." Further, I would grant that having the Treasury sell short term debt and use the proceeds to make risky loans or just purchase existing risky debt might impact nominal expendenditure in the economy "ceteris paribus."

Also, it is obviously true that if the central bank changes base money however much is necessary to keep interest rates at some level, then base money cannot simultaenously at some different level.

No one is disputing that.

For example, if people deposit currency into the banking system, and the bank of canada must remove those reserves if it doesn't want the policy rate to zero.

Well, I don't know about zero, but yes, the interest rate would fall if the bank of canada didn't contract the quantity of reserves to keep it from falling. The interest rate was targeting won't be targeted any longer.

[Scott Sumner]

Nick, Great post. I have a couple responses to K's comment:

1. It's wrong to assume that monetary stimulus lowers rates. A $50 billion drop in the demand for money is expansionary, if the central bank doesn't react by reducing the monetary base. But it's wrong to assume that because it's expansionary it will reduce rates. Throughout history, monetary stimulus has usually raised rates.

2. So what if the central bank had to let rates fall to zero, Nick wasn't assuming they were targeting rates. If they don't target rates, central banks can make their money irredeemable. They may choose to make money redeemable for bonds, but don't have to. Most importantly, their ability to respond either way (redeemable or not) gives them control of nominal aggregates. That's why other institutions don't have this control over nominal aggregates--they aren't able to make their liabilities irredeemable. In practice central banks target inflation, and make just enough base money irredeemable to hit their inflation target.

3. The supply of reserves (actually base money) matters because they aren't perfect substitutes to other assets, even other risk free assets. If they were, there would be no interest bearing public debt, the government would simply borrow by printing money.

[Nick Rowe]

Bill: agreed. One picky disagreement with this:

"Also, it is obviously true that if the central bank changes base money however much is necessary to keep interest rates at some level, then base money cannot simultaenously at some different level."

Unless perhaps, at the same time, the central bank makes a commitment to change (or leads people to expect a change) in *future* levels of the monetary base.

Just one example of that is paying interest on reserves. The CB makes a commitment: "If you have $100 in reserves this year, ceteris paribus, you will have $105 in reserves next year."

[rsj]

Nick,

No, I didn't read your comment. Sorry.

But I don't think you read K's.

A dynamic equilibrium is not a disequilibrium. It is just an equilibrium evolving in time in response to the different behavior of actors, which is also occurring in time.

In this case, the central bank is one of the actors, that is doing things in time and within the model.

If, for whatever reason, there is a given quantity of reserves, an interest rate, and a price level, then the effect of the CB increasing reserves only slightly will cause the interest rate to drop to zero, as K pointed out. Decreasing them only slightly will cause the OIR to climb and keep climbing. That's why its impossible for CBs to target reserves -- demand for reserves is almost perfectly price-inelastic. CBs can only set rates, and must always provide exactly as many reserves as the banking system needs, otherwise rates will be zero (if the CB provides more reserves than are needed) or banks will go bankrupt (if the CB supplies less).

The above story is completely compatible with being able to relabel your units. Apples and oranges. And it has nothing to do with whether we are in equilibrium or disequilibrium or whether money is neutral or not in the long run.

It's just a plain-Jane constraint on central bank behavior. Central banks lose control of the quantity of reserves the moment they set themselves up as lenders of last resort, or commit themselves to maintaining positive rates.

You have to decide how you are going to define your institution. Does it pick a rate and then promise to lend to all comers at the given rate, or does it not have this capability? If your CB is a lender of last resort, then your choice variable is not going to be the quantity of reserves.

One can imagine central banks that auction off a fixed quantity of reserves to the highest bidder. But they wouldn't be anything like a lender of last resort, and we have decided not to adopt this institutional structure.

[Nick Rowe]

Scott: thanks!

"Throughout history, monetary stimulus has usually raised rates."

OK, you are saying that, historically, the IS curve slopes upwards (well, if that's real rates).

I ought to have a closer look at the 82 recession. I *think* what happened there is that real rates first went up, then down when the economy was in recession, then rose again as the economy recovered. One day I need to get my head around the exact timing of the correlation between r and Y we would expect to see, and see if it roughly matches up with the data.

Your 2. Yep. You explained that better than me. CBs can *choose* what to redeem their money for (if anything) and how much of that thing to redeem it for, or how much to redeem. That choice, and commitments they make about their future redemptions, *is* their monetary policy.

[rsj]

Bill,

"I think open market operations are monetary policy regardless of the risk chacteristics of the assets purchased. The notion that it is only "monetary policy" if the riskless assets seems confused."

Yes, but the actual argument is that it is *only* monetary policy if government obligations are purchased. Purchase of risky debt is monetary and potentially also fiscal policy if the borrower defaults. The legislatures are greedy about keeping control of fiscal policy to themselves, and so constrain the CB to buy only riskless debt, in order to minimize fiscal spillovers.

[rsj]

And, for the same reason, central banks are not allowed to sell debt, write "NGDP" contracts, engage in bets or trades with the private sector, etc.

Of course, one can imagine any institutional powers in their models. A model can have central banks buying baskets of consumption, oil fields, or lottery tickets.

But in our reality, governments must make up for any defaults on assets that central banks purchase -- it comes out of their own bottom line -- and therefore they naturally insist that central banks purchase only their own obligations. This obviously minimizes the fiscal risks to the government, and there is typically more than enough government debt to supply whatever level of reserves and currency the private sector needs.

[Nick Rowe]

rsj: "A dynamic equilibrium is not a disequilibrium. It is just an equilibrium evolving in time in response to the different behavior of actors, which is also occurring in time."

Agreed. Fully agreed. It's just when we tell a disequilibrium story, it's much easier to tell it in the context of a stationary equilibrium. Otherwise, whenever we say "X increases", we have to say instead "X increases more than it would otherwise have done." (This gets especially problematic when you are talking about, say, loosening monetary policy to prevent deflation. Does it cause inflation? Well, yes and no.) But if you understand all statements as referring to changes *relative to an alternative time-path, which may or may not be a horizontal time-path*, it should all be the same.

"Central banks lose control of the quantity of reserves the moment they set themselves up as lenders of last resort,...."

That's an important point, but I would go both further, and less far. In a sense, the central bank loses control of monetary policy altogether during those times when the need to act as LoLR is a binding constraint. Because, it is in effect (I think, is this right?) committing to redeem BMO liabilities for BoC liabilities.

[Min]

Nick Rowe wrote:

"Min: "But the more remote the connection between policy lever and what you are trying to control, the less likely you are to be successful,..."

That was somebody else. :)

But thanks for your reply to my question, as well. :)

[Nick Rowe]

rsj: "Yes, but the actual argument is that it is *only* monetary policy if government obligations are purchased. Purchase of risky debt is monetary and potentially also fiscal policy if the borrower defaults."

That is one standard distinction. But I have always found it problematic. Suppose for example that a central bank buys and sells gold, and does nothing else. Perhaps at a constant price, perhaps at a variable price. Is that a *pure* monetary policy? You could argue that the real value of gold may fluctuate, so it's a risky asset.

[Nick Rowe]

Min: Sorry. I got Min and Max muddled!

[K]

Nick: "The only thing a central bank can ever really do is change its balance sheet, and make commitments to change its balance sheet in future. (That statement is commonplace among BoC people I have heard speak.) And the only part of its balance sheet that is special, that differentiates it from other banks, or from you and me, is the irredeemable "liabilities". Any talk about interest rates, even if true, is true only as a consequence of those two underlying hard facts. Those two facts are the ultimate tool. And that has always been true."

I think we are in absolute, full agreement here. So lets imagine that the BoC decides to revoke paper money. I think we must be in agreement that paper money is not important in determining monetary policy (If not, I'm happy to debate it). The BoC continues to conduct policy by *threatening* to change reserve balances in in the interbank payment system (but continues to target *zero* reserve balance). Now the BoC has *no* liabilities and no plans ever to have any liabilities. But they still have the power to control the interest rate for collateralized interbank balances by threatening to add (or remove) a little bit of net reserves in the system thereby causing someone to have to leave their money at whatever the CB decides to pay on reserves (or have to borrow at the discount window - not literally infinity, but really not something anybody wants to do). So the CB can continue to run monetary policy without ever having any actual outstanding liabilities. Just threats of temporary overnight balances (that never happen). 

Bill: "I think open market operations are monetary policy regardless of the risk chacteristics of the assets purchased. The notion that it is only "monetary policy" if the riskless assets seems confused."

I didn't say that. I said it's only monetary policy if *the money supply* is involved. A QE operation can be broken down into two parts: 1) exchange of reserves for T-Bills and 2) exchange of said T-Bills for risk assets. There is no reason to conflate the two. The first does nothing. To the extent that the second operation transfers risk from the financial sector to the government it can do stuff. But it doesn't change the money supply so it could just as well be done by the Treasury. If government wants to borrow short term in order to buy bonds or private risk assets I think that's better described as fiscal/industrial policy.

Nick: 1) Treasury borrows via t-bills and buys gold (fiscal). 2) CB buys t-bills (monetary).

[rsj]

Nick,

"But I have always found it problematic. Suppose for example that a central bank buys and sells gold, and does nothing else. Perhaps at a constant price, perhaps at a variable price. Is that a *pure* monetary policy? You could argue that the real value of gold may fluctuate, so it's a risky asset."

I did, too. Until I realized that

1) CBs only purchase finite maturity assets
2) CBs use cash-flow based accounting

So according 1) + 2), as long as no one ever defaults, the CB will not lose money and Treasury will not be on the hook to cover the loss. At least in theory.

[Min]

Nick Rowe: "Sorry. I got Min and Max muddled!"

Maybe I should spell my name with an "h". ;)

[Jacques René Giguère]

J.V. Dubois:
"than this has real impact on the long-term budget constrain http://worthwhile.typepad.com/worthwhile_canadian_initi/2011/04/functional-finance-vs-the-long-run-government-budget-constraint.html either in form of higher public debt and/or inflation."
a) no inflation as long as you are below some reasonnable distance from full emplyment.
b)The functionnal finance argument is a red herring. It was invented as a sop to those who opposed deficit financing by telling them we would run "surplus in good times" even if we know that it would contractionary ( I am not talking here about restrictive fiscal policy to counteract an export boom or similar.)

[Nick Rowe]

K: "I think we must be in agreement that paper money is not important in determining monetary policy (If not, I'm happy to debate it)."

We *might* have to debate that, I'm not sure. I *think* it may be important that the general public have direct access to the Bank of Canada's money. At the moment, paper currency is the only way the general public has that access. It doesn't matter that it's paper. But I think it does matter. Otherwise, we would be talking about something similar to a gold exchange standard where the public was not allowed to hold gold. We could even imagine the commercial banks setting up their own clearing house, where they did not settle on the books of the Bank of Canada, and so, in effect, just made their liabilities redeemable (at par) in each others' liabilities. (This is my late colleague TK Rymes thought-experiment). At this point, the Bank of Canada would have lost control of monetary policy completely. This, I think, is related to what happens when interbank lending markets seize up.

Leave that aside.

"So the CB can continue to run monetary policy without ever having any actual outstanding liabilities. Just threats of temporary overnight balances (that never happen)."

OK. I would add the words (that never happen *in equilibrium, because if the threat is credible it need never be carried out*)

But why does the CB have to use this power (or the threat to use this power) to target interest rates? Why couldn't it use this power to target something else? We know that, in the long run, the Bank of Canada doesn't use this power to target interest rates. It uses this power to target the CPI. And the only reason it can't target the CPI directly, in the short run, is because: StatsCan only publishes monthly data on CPI, with a lag; too much of the CPI is sticky. I can imagine a world where the CB doesn't have anything to do with banks at all. (OK, the Lender of Last resort function would be tricky). It buys and sells the TSX300. Or bricks. Or NGDP futures. Does central banking really have anything to do with banks? Yes I know it does now, today. But the ability of ivory tower academics to ignore current realities is our strength, as well as our weakness.

"Nick: 1) Treasury borrows via t-bills and buys gold (fiscal). 2) CB buys t-bills (monetary)."

OK. CB buys gold. Monetary or fiscal?

[Nick Rowe]

rsj:

"So according 1) + 2), as long as no one ever defaults, the CB will not lose money and Treasury will not be on the hook to cover the loss. At least in theory."

Suppose the CB buys only 30 year government bonds. Then nominal interest rates rise, and the prices of those bonds fall. If held to maturity the CB takes no loss (according to some accounting measures). But if the CB also needed to do a large open market sale of bonds, to withdraw money from circulation (for whatever reason) the market value of its bonds might not be enough to enable it to do this.

The ultimate asset of the CB is not on its books at all. It's the potential PV of future seigniorage. Its ultimate liability is the government's insistence that it hand some of that seigniorage over to the government every year.

[K]

Nick: "OK. CB buys gold. Monetary or fiscal?"

It's both. That was the point I was making by splitting it into the part that can be done by the fiscal authority and the part that can only be done by the CB.  

"I can imagine a world where the CB doesn't have anything to do with banks at all."

Me too! Now we're having fun!

They could, for example, provide money by temporarily swapping it with anybody against extremely liquid, transparently priced collateral (repo). They could also *buy* a diversified pool of liquid assets (instead of lending against them). This would be like gold-standard money except we could use a broad pool of capital assets. If they want to *increase* the amount of money in circulation they could *pay* interest on their money. (Yes it's weird - interest *paid* on money causes inflation). This would decrease the exchange rate of money vs the CB's assets but it wouldn't be a tax on the holders of money since they would get proportionately more money. And banks could fund themselves by selling bonds and stocks in the free market. Just like everyone else!

And there's no limit to how fast the CB could create more money (or take it away). And if there's only one kind of money with no inflation/seignorage tax, the link between the money supply and the rate of inflation would probably be a lot firmer. Just like Milton Friedman would have wanted. Oh and wait... no one would have an incentive to try to earn higher private seignorage profits by inflating debt and the money supply. And we wouldn't have to guarantee the liabilities of banks since they would no longer have importance for systemic stability. Which means the moral hazard would be gone, and they'd probably act a lot more responsibly. But, alas, we dream!

[vjk]

rsj:

"I did, too. Until I realized that

1) CBs only purchase finite maturity assets
2) CBs use cash-flow based accounting
"

Not sure how (1) or (2) is relevant.

I am with Nick on the idea that it is hard to split CB activity into purely monetary and fiscal.

When the feds engaged in QEi's, they clearly exposed themselves to the interest rate risk. When they bought agencies paper, they added the default risk exposure. When they bough gold, they exposed themselves to the usual commodity price volatility risk. Don't know if they hedged :)

Only repo activity with the loan maturity of day and two can be considered "pure" monetary policy during normal times due to negligible interest rate exposure. The rest appears to be a mongrel of sorts.

[vjk]

K:

"Now the BoC has *no* liabilities and no plans ever to have any liabilities.
"

If the BoC has no liabilities, that means there will be no interbank market which means that the whole system will come to a standstill and the overnight rates will go through the roof. There has to be some amount of grease in the system to make those cogs move. Granted, the BoC has very little of it in the system, about $150M daily last time I looked, but not zero.

Besides, in order to threaten credibly, the CB has to deliver on those threats from time to time through OMOs. Otherwise, everyone will stop believing the threats.

[rsj]

"Suppose the CB buys only 30 year government bonds. "

LOL, OK, then they would lose money. You don't need to be so exotic: a CB would lose money if it lent to a bank and the bank went bust, and also the collateral went bust. So you will get some fiscal side-effects.

The point being that Congress forced the CB to act in a very prudent way to limit the fiscal side effects as much possible.

[Nick Rowe]

K: "But, alas, we dream!"

It's our job to dream. It's a dirty job, but someone's got to do it. To see the current "reality" through a different lens, and so suggest a better one. The hardest part is escaping the current way of viewing the world, which is so wrapped up with the current policies. (Hmmm. Am I starting to sound like Keynes, or Marx?)

[Bill Woolsey]

Suppose there is no government debt.

The government paid it all off and keeps it budget balanced.

The central bank only purchases private securities, and they have have some credit risk.

Are these open market operations monetary policy? Or are they all fiscal policy?

Suppose the issue of hand to hand currency is fully privatized. The central bank issues no currency at all and all of its liabilities take the form of interest bearing reserve balances. Any profit or loss is distributed among the banks. In other words, there is no seignorage collected by the government.

The central bank purchases securities. Is that monetary policy?

Does it make a difference if these are government bonds that it buys? If there are no government bonds, or if there are such bonds but the central bank can only buy private bonds, is it still monetary policy?

Does the policy become more or less monetary depending on how much risk is involved in these bonds?

From my perspective, all of this talk about it is really fiscal policy and not monetary policy because of risk is based on institutional assumptions that are not at all essential. The notion that the government just has to make profits out of the system and how the government profits are impacted is what makes it monetary policy or fiscal policy seems... wrongheaded... confused...

Think about private central banks of the past. Did they only buy government bonds? Did they pay out all of their profits to the government? Did they do monetary policy?

[JKH]

It's CB monetary policy so as long as its an asset swap.

CB asset swaps (which amounts to CB asset liability management) result in CB profit or loss on the income statement.

The profit or loss is the fiscal contribution or by product of monetary policy. Some contribution, positive or negative, is unavoidable.

Doesn't matter if the asset swap is CB liabilities for government bonds or private sector assets; it's monetary policy.

As I understand it, one technical definition of a "helicopter drop" is a distribution of currency "ex nihilo" from the central bank balance sheet - which means a debit to central bank capital and earnings. That is not an asset swap. It is fiscal policy - a prospective negative contribution from the CB income statement. Also, its fiscal because it creates income for the recipient.

Fiscal contributions from monetary policy can be negative for various reasons - interest rate risk, credit risk, etc.

[JKH]

Buying gold is monetary policy.

Paying Carney is fiscal policy.

[rsj]

Bill, it's not either/or. Think of each operation as being decomposed into a fiscal and monetary component.

But in practice, when central banks are independent, governments constrain their ability to engage in fiscal transfers by limiting what the central banks are allowed to do. Central banks can't just buy and sell anything they want.

The fiscal component is the result of the change in the government's balance sheet as a result of the CB operation, with the understanding that the government "owns" the central bank and has rights to all seignorage income, or net interest income.

In terms of "confused" and whatnot -- well, you can focus on what you think is important. I don't think the quantity of outside money is important at all. To me, focusing on that is confused, when interest rates and income are what should be focused on. But the reason for monitoring changes to the government's balance sheet is that the reverse changes are occurring in the private sector's balance sheet. It's important to know whether the private sector is gaining income or merely swapping assets. If the BoE is defending the pound, making George Soros a billionaire -- and forcing the taxpayers to provide those billions, then that has a different impact on the private sector than if the bank of england is merely swapping bills for sterling.

[Lars Christensen]

Nick, thank you for a great post. I really need to thank Scott and Josh as well. You guys just gave me a couple of more pages to put into my update version of my paper. I we should not forget Arash - I think he did the Market Monetarists a great favour in challenging your (our) views on (dis)equilibrium. With this kind of discussion the views of MM become even more clear and well-defined. I think that is very good.

[Lars Christensen]

Btw Nick...I would love to see a post on Clower's contribution to money and markets...I guess that is the name of the collection of Clower papers. Is Clower, Yeager, Laidler and Friedman the foundation for MM?

[Nick Rowe]

Thanks Lars. Yes, Arash's post, like your paper, gave us the stimulus to try to clarify some of our thoughts. I think the only real foundation of MM that all of us have in common is dissatisfaction with the idea that monetary policy has "shot its wad" when it hits the ZLB. A post on Clower's contribution would be interesting. But I would do a bad job of writing any sort of post like that.

[Lars Christensen]

Nick, personally I am far too ignorant of Clower, but I feel that his contribution to monetary theory needs a lot more of renewed attention. He has obviously been important for your thinking so I would love to hear how you see his views compared to what you think today and for example to Scott or even Yeager. Or maybe you better start planing for the first MM seminar!

[Nick Rowe]

Lars: "The fox knows many things, but the hedgehog knows one big thing." Clower was a "hedgehog". This is it:

Patinkin had a Walrasian model of monetary economics. Clower said: "Hang on. In a Walrasian economy there is one big market where all n goods are exchanged; in a barter economy there are (n-1)n/2 markets where 2 goods are exchanged; and in a monetary economy there are (n-1) markets where 2 goods are exchanged, one of which is money. And, if the price in one of those markets is wrong, either buyers or sellers will be quantity constrained, and that will affect their demands and supplies in all the other markets."

[Scott Sumner]

Isn't K's proposal the same as Woodford's "moneyless" economy? Or more precisely an economy where the net amount of money is zero, and monetary policy is done by adjusting the demand for money rather than the supply of money? It could well happen that way someday, but I'd hope we would have stopped targeting interest rates by then.

[Jim Rootham]

Is there any real point to all this?

In particular, what is the different policy proposal for an economy at the zero lower bound, and has any economy recovered by using that policy?

[Patrick]

THIS IS SIMPLY NOT DONE.

Sorry. I had to vent somewhere.

[K]

vjk: "Granted, the BoC has very little of it in the system, about $150M daily last time I looked, but not zero."

Most of this year, net reserves have been very near $25M. In the last few weeks it's gone up to about $400M, which might mean that *somebody* is having trouble in the interbank market and is using the SLF (the discount window). No matter what, it's pennies.

"If the BoC has no liabilities, that means there will be no interbank market which means that the whole system will come to a standstill and the overnight rates will go through the roof. "

Of course, yes, not literally zero. As close to zero as the system can handle without small errors causing big problems.

"Besides, in order to threaten credibly, the CB has to deliver on those threats from time to time through OMOs."

They are delivering *all the time*. They might put extra cash in the system if they see that rates are trading high, and then, in the afternoon as everyone scrambles to unload the hot potato and rates are dropping, they'll come back in and remove the excess *before it ever actually occurs*. The threat of an overnight balance can be very real if the BoC has already been in the market. So usually they end up doing trades during the day that, by the time of settlement (18:30 or whatever) add up to net zero.

Nick: yup, it's *worthwhile* taking the time to think about how things ought to be in a better world! 

[K]

Bill: "The central bank only purchases private securities, and they have have some credit risk."

In my thought experiment I was imagining a "mutual bank," ie. only one type of liability: money. I was also imagining that anyone could redeem money in return their share of the banks portfolio of assets. Monetary policy, then, consists exclusively of "stock splits". The bank anticipates that prices of consumer goods will fall, eg., by half, so it splits each share in two; a *purely nominal* change, with *zero* redistributive effect. You can achieve this in any 100% reserve system if the commercial banks were required to pass on the policy rate into deposit accounts.  *That* is the purest form of monetary operation. To the extent that anything else can effectively accomplish *that*, it is monetary. Otherwise we can call it fiscal.

"The notion that the government just has to make profits out of the system and how the government profits are impacted is what makes it monetary policy or fiscal policy seems... wrongheaded... confused..."

I don't want governments to make profit from the system. I want money to be a neutral veil. The point is to avoid social engineering through money.

[K]

Scott: "Or more precisely an economy where the net amount of money is zero, and monetary policy is done by adjusting the demand for money rather than the supply of money?"

I *think* it's quite different. Money wouldn't be created by risk averse investors lending to risk seeking ones. It would be created by monetization of capital assets. Investors exchange units of the market portfolio for units of the CB. And my proposal is to meet whatever monetary target by adjusting the quantity of the unit of account by splitting it (paying interest). But that wouldn't change the market value of the total medium of exchange which is given by the value of the CBs assets. And it doesn't redistribute wealth since everyone maintains a claim on the same fraction of the central banks assets.

I see much less potential for runaway money creation or destruction in such an economy so I think the need for stabilization would be smaller. That said, there'd still be changes in velocity so an NGDP level or similar target strikes me as eminently practical.

[no exit]

you can't have a barter society that includes cash. accumulated cash overwhelms accumulated barter... unless you're bartering raw supplies like oil, gold and diamonds... but then you're bartering it for cash. cash is the killer app that killed barter... which is now so old skool. i hear it's two chickens for an angiogram at the bunny ranch...

[Sandwichman]

Mind/body dualism, Nick. Failed metaphysics. An economy is not a car.

[JW Mason]

It's very hard to argue with monetarist position on formal grounds. (Believe me, I've tried.) Telling a coherent story about a shortfall of demand for currently-produced output that doesn't involve excess demand for money may be possible, but I have not figured out how to do it.

But arguing with the policy conclusions that monetarists like Nick want to draw, may be easier.

I see at least two problems with the leap from the diagnosis of excess money demand to the prescription of a monetary solution.

First, what is "money"? Here, it seems to be base money, high-powered money -- reserves and currency. But clearly, the stuff that actual economic units -- firms and households -- have excess demand for, is not just reserves (which they can't hold anyway) and currency. It's some broader class of assets that are money-like in the sense that (1) they are safe and convenient stores of value and (2) they can be used for settling contractual obligations. (People talk more about 1 but 2 is probably more important.) Of course money also includes anything that can quickly and reliably be turned into money.

So to get from excess "there's demand for money" to "the solution is monetary policy" you have to jump over the process that connects the "money" that central banks control to the "money" that is actually in excess demand. There's a lot of complex institutional detail in that process, but Nick just jumps over it. But without understanding that, you really have no idea if a particular form of monetary policy is actually increasing the stock of the money that's in excess demand. This may be hard to do, especially since the moneyness of different assets are often going be endogenous to policy, and to whatever problem created the need for policy. It's a lot easier to find the illiquid end of the spectrum -- it's hard to know when reserves are going to be more money than T-bills and when they are not, but they are both surely a lot more money than an hour of labor. So policy has a more secure purchase when it tried to increase demand for the illiquid stuff, rather than the supply of the liquid stuff.

Second, even though a pure coordination problem model probably can't describe what we see in real recessions, there still can be an important coordination problem aspect to them. Specifically, the monetarist solution (fix the problem from the money-supply side) assumes that additions to the total stock of money will find their way to the particular units with excess money demand more or less automatically. It seems to me that it should be clear from the past few years (if not long before) that this simply isn't how the financial system works. There are huge transaction costs in many credit transactions that make it very possible for some units to be experiencing excess demand for money while other have excess supply. Again, the effectiveness of monetary policy as it's normally conceived is probably much more dependent on the specific institutional structure of the financial system than someone like Nick wants to acknowledge.

It's true that a Walrasian barter economy without money cannot experience recessions. It does not follow that the best framework for thinking about recessions, is Walrasian barter with money added.

[Phil Koop]

Excellent comment by JW Mason @ 01:54am.

[Nick Rowe]

What Phil said. I'm going to have to come back to this later.

[vjk]

K:
"which might mean that *somebody* is having trouble in the interbank market and is using the SLF
"

Not necessarily. Hard to say from the weekly stat table as the overdraft loan terms are between 1 and 4 days so their impact on the CAP total balance may be negligible. Appears to be due to bond acquisition by BoC (monetary policy ) and notes in circulation fluctuations.

[vjk]

CAP -> CPA

[Kevin Donoghue]

JW Mason: So to get from excess "there's demand for money" to "the solution is monetary policy" you have to jump over the process that connects the "money" that central banks control to the "money" that is actually in excess demand. There's a lot of complex institutional detail in that process, but Nick just jumps over it. But without understanding that, you really have no idea if a particular form of monetary policy is actually increasing the stock of the money that's in excess demand.

To my mind the more fundamental problem is that there is only a notional excess demand for money. At depressed levels of income households show no inclination to add to their holdings of money. It’s reasonable to believe that at full employment they would hold more, but that doesn’t, of itself, mean that the central bank can get us from here to there.

At times I get the sense that what Nick and other MMs want is a model which elevates radical monetary policies above fiscal policy. But if they actually have such a model I haven’t seen it. Of course there’s nothing wrong with searching for a model which will justify what your intuition tells you is the right policy. Keynes was advocating public works long before he wrote the GT. But right now the MMs aren’t so much a school of thought as a school of instinct.

[Sandwichman]

Yes, Josh, it's transaction costs all the way down! Fixing the economy's "magneto" may take a lot longer than two hours at the side of the road. A lot longer.

[Sandwichman]

Inestimably longer.

[Too Much Fed]

K: "But, alas, we dream!"

Bill Woolsey said: "Suppose there is no government debt."

K, let's take that a step further. Can you imagine an economy with no currency denominated gov't debt and no currency denominated private debt?

[Simon van Norden]

"This is like correctly diagnosing magneto trouble, then recommending we all get out and push the car, rather than fixing the magneto. I just refuse to accept that that's the best we can do. We need to understand that monetary magneto better, and learn how to fix it."

I'm not sure why you'd want to stop at understanding the monetary magneto; why not keeping pushing until you can create a complete set of Arrow-Debreu contingent markets? I think policy is supposed to fall by the wayside at that point.

And I'm not being snarky....I'm trying to point out that there are many pretty-valid approaches to policy formulation and that differences in what they take as "exogenous" can be key. Is it fair to put Nick's approach on a continuum somewhere between those who just worry about "what can pass the Republicans in the House?" and those who think only in terms of perfect, frictionless and complete markets? Can all three approaches be defended?

[Greg Ransom]

This is an important point. But it also cuts against the Keynesian / fiscal "solution", which suffers from exactly the same sort of problem, often in spades.

And who will arrive at these solutions more competently: the U.S. Congress? the Federal Reserve? or distributed decision makers & learners "crowd sourced" to the markets?

And what if we replaced the Central Plan makers at the Fed with a "crowd sourced" distributed system of independent learners & deciders to "solve" the problem, i.e. via private banking and the denationalization of money? I.e. the solution of Bagehot & Hayek.

Mason writes,

"the monetarist solution (fix the problem from the money-supply side) assumes that additions to the total stock of money will find their way to the particular units with excess money demand more or less automatically. It seems to me that it should be clear from the past few years (if not long before) that this simply isn't how the financial system works'

[Greg Ransom]

Foster & Catchings & Sen. Wagner & many other Americans were advocating public works to eliminate deficiencies of demand before Keynes, in the 1920s ... Herbert Hoover was such a fan of Foster & Catchings that he wrote a complementary introduction to one of their books on the topic.

The economics profession proved collectively incapable of refuting the arguments of Foster & Catchings, in a world-famous prize competition to refute their work.

In many ways, Keynes in 1936 didn't do much more than put a fancy "modern" and "scientific" imprimatur rooted in Marshallian economics on the prior ideas of Foster & Catchings, by tacking on an absurd monetary gloss effectively assumed away both economic scarcity and any genuine guidance role for prices across time, especially in the domain of production goods and inputs to production.

In other words, just as Foster & Catchings instructed the "instincts" of Sen. Wagner & Herbert Hoover in the 1920s, so to it seems did they also set the scientific background atmosphere for in which Keynes'"instincts" eventually grew and advanced.

Kevin wrote,

"Keynes was advocating public works long before he wrote the GT."

[Determinant]

There is a bit of unreality floating around this thread that the Central Bank does not have liabilities. It is a simplistic but true observation that the CB's balance sheet must balance. Assets = Liabilities + Owner's Equity. There has to be "Something" in all those categories. Determining what that something is or should be the point of monetary policy, but you can't go neglecting one side of the balance sheet and focus exclusively on the other because the math will no longer work.

Greg: And Hayek didn't fully realize that a monetary economy can be strangled by its money market. Not enough money = failure to clear = recession. Money is a noose around the economy's neck. Tighten it and you strangle it.

Adjusting prices downward or injecting more money comes to the same thing: paying for defaults. Downward price adjustments means borrowers pay, increasing the money supply means lenders pay. We have been going around in circles for seventy years over who should pay. But ultimately somebody will pay.

[Jacques René Giguère]

Determinant:
For an accountant, there must be something in these categories ,anything, just so there is some fictional balance.
The CB balance its sheet with TB. These TB are backed by the Full Faith and Credit of the Gov't. Ultimately the faith that the economy will run and the Canadian Revenue Agency will collect taxes. The same as with a personnal margin of credit.The bank hope we will have an income and pay back. Nothing of value to sell in a functioning market.
I won't go in whether the CB has "liabilities". But it has no assets in the sense a business has assets. There is nothing but hope. And the fact that Oz functions and we don't want the push the curtain too much.

[rsj]

Jacques, that is true for any bank. It has loans on the asset side and deposit/bond obligations on the liability side. Just because the assets are not tangible does not mean that they are pure hope. They are, in fact contracts, not wishes. The real side of the economy orients itself in such a way as to try to fulfill those contracts, even to the point of doing things that might seem silly if one ignores the financial contracts and looks only at the underlying tangible assets.

[Greg Ransom]

The collapse of "shadow money" and liquidity generally is part of Hayek's explanatory machinery.

So Hayek did realize that the "money" market was part of the "strangulation" process. He did realize that "not enough money" was part of the failure to clear at the turn of the boom / decent into bust -- he writes about that, including discussions of the disequilibrium caused by the "secondary deflation / depression". And it does this very early and very late and consistently throughout his career.

In other words, all of the work of your remark here is done with the weasel word "fully". Eliminate "fully" and your remark is false.

The ambiguity and cognitive near meaninglessness of your remark is indicated by the fact that there is no economists you could not say the same thing about .. if you leave in the word "fully".

D said,

"And Hayek didn't fully realize that a monetary economy can be strangled by its money market. Not enough money = failure to clear = recession.'

[Jacques René Giguère]

rsj: as long as goods and services are not delivered , contracts are wishes. Even a court order and a baillif can't change that.

[rsj]

But Jacque, that's a simplistic view. First, debt is always rolled over, and expectations of return are just as, if not more important. Government debt, as its backed by taxes, is pretty damn strong. Even if the government has trouble collecting taxes now, it only needs to send a signal that it will be able to collect taxes at some future point, and the debt is treated as risk-free today.

Similarly, take the point of view of a firm. It is having trouble selling goods today and earning a high margin today. So it lays off a lot of employees, closes some divisions and takes efforts to regain its higher margin. If the investors believe the efforts are credible, then the stock price of the firm goes up, even though it has less tangible capital and less revenue as a result of the liquidation.

So the relationship between tangible stuff and the value of promises can often go in the opposite direction of what you would expect. And typically the tangible stuff follows, rather than leads, the promises.

[A]

New Keynesian macroeconomics makes no sense whatsoever in a barter economy

If economics is supposed to be a science, then one should base one's macroeconomic theory on actual evidence. There is no evidence that a barter economy has ever existed:

http://www.nakedcapitalism.com/2011/09/david-graeber-on-the-invention-of-money-%E2%80%93-notes-on-sex-adventure-monomaniacal-sociopathy-and-the-true-function-of-economics.html

As such, attacking the New Keynesians for (apparently) not being able to make sense of a barter economy is no better than attacking chemists for not being able to explain phlogiston, or attacking physicists for not being able to explain the workings of a time-travelling DeLorean. It's fantasy, not science.

[A]

J. V. DuBois:

But if you think that markets are doing better jobs at allocating resources

Firstly, no-one reasonably believes this always. We don't trust the market to allocate resources for many things (police, justice, education, etc). So simply asserting that the market is better is not an argument for the Central Bank handing money over and leaving it to the market - it ignores "market failures" in other words.

Secondly, it's not simply the "market" we're talking about, is it. If the Central Bank does QE, then it benefits the already wealthy:

http://www.centreforum.org/assets/pubs/credit-where-its-due.pdf

(report from UK perspective)

who aren't exactly in need of desperate help. QE as it stands is not neutral. The Central Banks have already picked winners. The question should be about making sure those winners are the right people.

What's wrong with Min's suggestion of doing QE and then mailing out cheques to those who need the money? And after all, they're the ones most likely to spend it, rather than save it, thus increasing monetary velocity. And for those that save, it would also help them deleverage quicker.

Finally, the focus on monetary policy here misses two points:

1. We're not really talking about monetary policy as it used to be thought. Nor is what Nick is suggesting not fiscal policy. At best, we've established that there is no neat dividing line between fiscal and monetary policy. You can call these proposals "monetary if you want, but I see no reason not to call them "fiscal" either.

2. It misses the point of what problem we're trying to fix. The problem is not simply "lack of growth". The main problem is huge amounts of unemployment (and too low wages), causing immense human misery. It is this misery that we need to end first and foremost (similarly, lets end world hunger before we start talking about productivity rises in Ethiopia). And as one man once send, monetary policy has long and variable lags. This is not going to help the average working class individual fast enough. Government can directly provide jobs. Let it do so while we wait for other interventions (monetary, quasi-monetary, fiscal, call them what you want) to have their effect.

[A]

Greg:

And who will arrive at these solutions more competently: the U.S. Congress? the Federal Reserve? or distributed decision makers & learners "crowd sourced" to the markets?

I'm not interested in a Hayek v Keynes showdown debate, so I'll just mention one weakness with your argument:

Your argument is that "crowd-sourcing" as you term it, is the more trustworthy solution. That is, it isn't an argument that government could never solve the problem, only that it's much less likely than the market to do so. The problem with this argument is that it is vulnerable to a reductio ad absurdum. After all, if we can't trust the government to choose the correct solution to the crisis, then why should we trust the government to read Hayek or his heirs, and implement Hayek's ideas correctly?

That is, if the government is so bad at decision making, then why would it opt for Hayek? My argument has this form:

P1: Governments can't be trusted to do central planning correctly in the slightest, not even monetary policy.
P2: The best form of monetary policy is to not do it, from Hayek.
C: From P1, government will fail to not do monetary policy. It will instead do monetary policy, which is bad according to P2.

So if the government won't listen to Hayek (who in your opinion was right), then who will it listen to instead? By continuing to advocate Hayek rather than some more moderate position, you (by P1) fail to get yourself heard by the government, thus defeating the point of your advocacy of Hayek.

In other words, if Hayek was right, then there's no way to get government to listen to what you want it to do (or top doing). Advocating Hayek is self-defeating.

[Also, on your bit about the market being better because its "distributed decision makers & learners", this is totally flawed. Not everyone has equal standing in the market - some have more money/purchasing power than others, and so their actions in the market carry more weight. A better form of crowd-sourcing and "distributed decision makers & learners" would be either direct democracy (i.e. everyone's voice counts equally, no arbitrary-ness in who gets how much say), or deliberative democracy (that is we decide what to allocate resources on based on discussion, with everyone getting the same free speech rights, but we reach a consensus rather than a simple vote). Even these have weaknesses - compare the American populations general support of creationism or intelligent design with scientists support of evolution for instance. Very often the wisdom of crowds is replaced by the madness of crowds. In short, what system is best for finding out truth? Every system is flawed. Let's not get ideological about this. And "what is the best way to allocate resources?" is not a question of the same type as "did mankind evolve?". The latter is about truth, the former about morality. There is no right answer. So the solution must be a political one. Choosing the market above all else is inherently a political choice. You've decided on the means before we've even decided on the ends we're aiming for. There are other consequentialist arguments against government, but this Hayekian one is not a good one IMO.]

[Matias Vernengo]

You seem to think that the natural rate hypothesis, that is accepted in New Keynesian models, is fine. No problem with the fact that there seems little evidence for it?

[K]

Too Much Fed: "K, let's take that a step further. Can you imagine an economy with no currency denominated gov't debt and no currency denominated private debt?"

I think so. As Nick once pointed out, nominal debt is a lot like glass (and equity like rubber? I forget). It breaks, and causes huge dead weight losses when it does so (the lawyers take half). The principal benefit of debt is tax treatment. So companies and individuals lever up until the marginal increased expected cost of default approaches the marginal tax benefit. And the financial sector encourages favourable tax treatment of debt (like the US mortgage deduction) because it increases borrowing and the money supply. But without those tax reasons (there are other minor effects), Modigliani-Miller says there is no benefit to corporate leverage, since investors can choose their *own* level of leverage. I think that's true. And margin borrowing against liquid stocks should be more efficient than borrowing against illiquid corporate assets. So change the tax code and we would have way less debt and far less dead weight losses from bankruptcy. I don't see why we couldn't have just equity.

Could governments equity finance? Maybe they could issue "stock" that entitles the owner to a proportion of future revenues. Could work, and seems more transparent, i.e.: "This year, in order to finance current services and tax cuts, we sold to private interests 1% of what could have been future government services. Once we've sold 100% of future revenues, you'll be all out of services. Forever."

The MMT crowd would have the government not finance *at all*. I.e. all their debt would be monetized. I think this is a bad idea because the market mechanism for government debt pricing is really important. It is our consensus estimate of the ability of the government to shift current expenditures onto future tax payers. Higher yielding bonds prevent governments from taking on liabilities that the market estimates that future tax payers are going to want to pay for. The owners of the government debt can then monetize it at the CB at its fair market price if *they* have a desire for increased liquidity. But allowing any issuer (including government) to circumvent market demand and directly monetize their debt results in a moral hazard that is likely to be hyper-inflationary. That's why we need independent central banks.

Your question is important because, as Nick discussed in a post a month or two ago, If you back the CB's balance sheet with nominal liabilities and the CB has no equity, then the value of the liabilities (money) of the CB can be undefined if they are not equal in quantity to the amount of nominal assets. So CB's need to issue equity, *or* they need to own a substantial amount of non-nominal assets (equity). But I don't think governments need to "equity" finance for my central bank proposal to work. The CB just needs to hold a lot of private sector equity securities relative to the amount of nominal assets.

[Determinant]

Greg:

Here is how Hayek got in wrong on money and recessions: http://www.skidelskyr.com/site/article/interpreting-the-great-depression-hayek-versus-keynes/

"These Hayekian and Keynesian forecasts turned into explanations of the slump once it had happened. In a lecture in Cambridge in 1931 Hayek, newly brought to the LSE by Lionel Robbins, expounded the theory of his book, Prices and Production. The slump was due to a crisis of over-investment –overinvestment in relation to the amount of consumption people wanted to forego –financed by credit-creation by the banking system. The slump was merely the process of eliminating the unsustainable investments, those not financed by genuine savings. Government pumping more money into the economy would merely prolong the agony. The quickest cure would be for people to save more, to bring about a recovery in private investment. Richard Kahn records that this lecture was received in total silence by Hayek’s Cambridge audience. To break the ice, Kahn asked: ‘Is it your view that if I went out tomorrow and bought a new overcoat, that would increase unemployment?’. ‘Yes’, Hayek replied, turning to a blackboard full of triangles, ‘but it would take a very long mathematical argument to explain why.'"

Wrong, Mr. Hayek. A Depression is a state-space transition from a supply-constrained economy to a money-constrained economy. Once that transition is made, the economy will permanently settle into a lower state of consumption and investment. You can try to collapse the economy's capacity, contracting its Production Possibilities Frontier, but that both eliminates capacity, creates involuntary unemployment and extracts a terrible human cost.

Further in a money-constrained economy the clearing rate of interest cannot be attained by the market (it lies beyond the area of attainable rates for a given money supply). In this case investment will not work to grow the economy because economic growth to make savings turn into investment requires a clearing rate of interest. This isn't possible in a Depression state-space.

No, don't try to invest your way out of it first. First inject more money, get rid of involuntary unemployment, transition to a full capacity economy and then, only then, invest your way to growth. But the first step is to create more money. Inflationistas be d**ned.

[K]

I meant "Higher yielding bonds prevent governments from taking on liabilities that the market estimates that future tax payers *aren't* going to want to pay for"

[Nick Rowe]

Scott: I don't think Woodford is at all clear on whether his "cashless" economy is also a "moneyless" economy.

Jim: well, economies that went off gold (raised the price of gold) in the 1930's seemed to recover, even though interest rates were near zero. We need the modern equivalent.

Patrick. Agreed. But I'm not American, so try mostly to stay out of US politics. But I thought Scott Sumner's response was right. You could argue that in this case, if US monetary policy is screwed up, it affects a lot more than just the US.

K @11.23. Neat thought experiment. The Quantity Theory thought-experiment is very much like a stock split. Some of the "backing" theorists have used that analogy before. But you are turning it into a (possible) policy recommendation. Filed at the back of my mind for possible future use.

no exit: ? I know cash kills barter (mostly).

Sandwichman: metaphors are like motorbikes; they don't run on 4 wheels.

JW: "But arguing with the policy conclusions that monetarists like Nick want to draw, may be easier."

Agreed. That monetary policy can always cure a monetary problem does not follow as a matter of pure logic. (But, dammit, it *ought* to!)

What is "money"? I see the medium of exchange function as essential; the medium of account function as secondary, but still important. Yes, money is also some sort of store of wealth, but so are hundreds of other goods too.

"Of course money also includes anything that can quickly and reliably be turned into money."

I don't agree with that. Near monies may be near monies at the individual level, but not at the aggregate level. Plus, a difference in degree can become a difference in kind. There is a spectrum of goods with different degrees of liquidity, but the good at the most liquid end of that spectrum becomes used in a totally different way. It becomes the medium of exchange.

"There are huge transaction costs in many credit transactions that make it very possible for some units to be experiencing excess demand for money while other have excess supply."

Agreed. A representative agent model is a bit of a failure here. For example, it is (especially) the unemployed who have an excess demand for money. So to suggest that there can't be an excess demand for money, because anyone with an excess demand for money could just sell bonds, is missing the point.

"It's true that a Walrasian barter economy without money cannot experience recessions. It does not follow that the best framework for thinking about recessions, is Walrasian barter with money added."

OK, but if you really do "add money" (in some essential way) to a Walrasian model, it is such a different model that it isn't really Walrasian any more. This was the critique of Patinkin, that he added money to a Walrasian model without adding it in any essential way (people didn't need to use money to buy and sell goods). Patinkin's model is formally identical to my "bling" model. Take a barter economy, then add M/P to the utility function. Let M be the stock of bling. An excess demand for bling cannot cause a recession.

http://worthwhile.typepad.com/worthwhile_canadian_initi/2009/05/imagine-theres-no-money.html

Kevin: "To my mind the more fundamental problem is that there is only a notional excess demand for money. At depressed levels of income households show no inclination to add to their holdings of money. It’s reasonable to believe that at full employment they would hold more, but that doesn’t, of itself, mean that the central bank can get us from here to there."

That's what happens in a standard ISLM model, with a downward-sloping IS and Md/P=L(Y,i). But, with uncertainty removed, other assets would become safer and more profitable, so the stock demand for money might fall. Also, remember that the unemployed have an excess flow demand for money in the labour market, and presumably an excess stock demand too. If an unempolyed individual succeeded in selling his labour, he would spend most of his extra flow of money income.

[Nick Rowe]

Sandwichman: Asset prices would respond to a change in monetary policy in much less than 2 hours. But yes, it is better to build a magneto that doesn't break than to keep fixing it when it does break. Then we don't have to waste 2 hours at the side of the road, or keep buying new batteries.

Simon: "I'm not sure why you'd want to stop at understanding the monetary magneto; why not keeping pushing until you can create a complete set of Arrow-Debreu contingent markets? I think policy is supposed to fall by the wayside at that point."

Money (even bad money) is a lot cheaper and more effective than a complete set of A-D markets. That's why we use it. Even Zimbabwean money survived, and beat barter, until the very end.

Greg: I'm not against private monies. But, as far as I can see, truly independent (i.e. not redeemable at par in government money) seem to fail the test of the market. With a few exceptions, people just don't use LETS and similar monies.

Determinant: "There is a bit of unreality floating around this thread that the Central Bank does not have liabilities."

I see it the other way around. You can always force a balance sheet perspective onto a central bank if you push hard enough. Just call an irredeemable currency a "liability", even when it isn't. It's pushing it a bit when we make a regular firm's balance sheet balance by saying that equity is a liability, but at least shares are a promise to pay a share of the profits. BoC currency isn't a promise to pay anything.

Jacques: yes. The stuff on the "asset" side of the CB's balance sheet is even weirder. Most of it (bonds) are promises to pay those "liabilities" that appear on the other side of the balance sheet.

Makes me think that Scott Sumner is really right on this. A central bank is not a bank in any way. The whole balance sheet is some sort of self-referential joke, just to keep the accountants happy.

A: you really missed my point on that bit. Since we live in a monetary economy, it's perfectly OK if Keynesian models only work in a monetary economy. But it's not OK if they don't explain *why* they only work in a monetary economy. It's not OK if the model cannot explain why it won't apply equally well in a barter economy. It's not OK if a model that only works in a monetary economy doesn't say anything about money.

It's like having a physics model which ignores friction and yet only works if there is friction. Something is wrong.

Matias: I'm generally more or less OK with the natural rate hypothesis. I see that as a separate question, not related to the problems with the New Keynesian model from a monetarist perspective.

[J.V. Dubois]

JW Mason: "it's hard to know when reserves are going to be more money than T-bills and when they are not, but they are both surely a lot more money than an hour of labor"

This reminds me of the old Keynesian critique of Quantity Theory that velocity actually fluctuates. The easiest response is - you are no better, multiplier is not much more stable. Do you really understand the intricacies of the government projects and how effectively fiscal stimulus flows to units with excess demand for money? If government would choose a fiscal stimulus in form of "iPad for everybody" the result could be huge profits for Apple, maybe a few more people hired in retail - and of course a lot more cash reserves on Apple bank accounts.

So we are back in the realm of reversibility. If you try fiscal stimulus and fail, you may end up with a lot of wasted production. There is no way back, resources were already spent and wasted. You failed to eat that free lunch. On the other hand if you conduct monetary or "fiscal/monetary" policy of buying private assets such as corporate bonds, or stocks - you may always sell them back and try to pour money somewhere else where it has better chance to reach those units craving for money.

Actually I may go further. As Nick mentions in a lot of posts, monetary policy is mostly about expectations and credibility. If government says that they plan to do 10 000 unspecified projects as part of a fiscal stimulus, no one can imagine what effect this will have. Investors may just wait to see what people that recieve that government money want to buy. On the other hand, if central bank says that the policy changed to NGDP level targeting and that for starters they will aim stocks and corporate bonds of most labor intensive industries until it reach certain value, it is quite likely that private investors will pour their money there without need for CB to spend a cent on those stocks/bonds. Remember recent example of Schweizerichse Nationalsbank and you get the idea of what power words of credible institution have. If, as I think we all agree, this recession is one huge coordination failure, and if words, not spending, can solve it - hy not give it a try?

[K]

Nick: "So to suggest that there can't be an excess demand for money, because anyone with an excess demand for money could just sell bonds, is missing the point."

Right. But which point is it missing? I think it's missing the point that if borrowing costs are not the same for everyone, there is no unique, *independent of wealth distribution*, economic equilibrium. This has been known since (moneyless!) CAPM, and probably before. Krugman and Eggertson merely repeated that point in a modern (moneyless!) NK framework.

[Kevin Donoghue]

Nick, thanks for the reply but you lost me. With Y below potential and L(Y,i)=M/P, we’ve got notional (but not effective) excess demand for money. But I don’t see what you mean by “uncertainty removed” unless it’s just whatever uncertainty caused money demand to jump in the first place. That’s not reassuring since it just means the economy might pull itself out of the slump. Similarly it doesn’t help to know that the unemployed would spend income if they had it, when in fact firms don’t see any profit in employing them. But of course you know that, so you must be saying something more which I’m missing.

...if you really do "add money" (in some essential way) to a Walrasian model, it is such a different model that it isn't really Walrasian any more.

Likewise if you introduce rationing of any sort into a Walrasian model then it isn’t Walrasian any more. So what’s the case for saying that money, rather than quantity constraints, is essential? Now, if by “adding money” you simply mean ruling out barter I’ve no problem with that. But then labels like Market (or Neo or Quasi) Monetarism are misleading. What you’re really promulgating is No-Barter Macro. I’d be happy to sail under that flag. (Though it sounds a bit like No-Bullshit Marxism, a movement of which I know nowt, in that it defines itself by what it seeks to remove from the doctrine to be reformed, rather than by what it seeks to add.)

[Min]

Kevin Donoghue: "To my mind the more fundamental problem is that there is only a notional excess demand for money. At depressed levels of income households show no inclination to add to their holdings of money"

No comprende, sennor. Don't households with depressed levels of income have a demand for money so that they can spend it? I may have a demand for ice cream, but that does not necessarily mean that it goes right into my freezer. ;)

[Nick Rowe]

Kevin: my "uncertainty removed" was just (poor) shorthand for "if AD recovers, risk of default falls, and fear of future unemployment falls, and risk of real assets falling in price falls, etc.". In other words, we can imagine a lot of reasons why the demand for money might be higher in a recession, even though Y is smaller than Ys.

"Monetarism" (with any qualifiers) is not a very precise label, agreed. But No-barter Macro, with sticky prices and quantity constraints, with a quantity of the medium of exchange actually appearing in the model, and people using it and holding it for some explicit reason, would be a good start.

[Nick Rowe]

Kevin, Min: money is weird. It's not like other assets. Suppliers of fridges can only increase the stock of fridges in public hands if the quantity people want to hold exceeds the current stock. Suppliers of money can increase the stock in public hands even if the current stock of money equals the desired stock. Just buy something. To be willing to accept money does not imply a willingness to hold more money. The stock of money is supply-determined in a way the stock of fridges isn't.

[Min]

Kevin Donoghue: "Similarly it doesn’t help to know that the unemployed would spend income if they had it, when in fact firms don’t see any profit in employing them."

Gee, it seems to me that the fact that the unemployed would spend money if they got their hands on it is widely ignored by policy makers today, and that is one reason for our continued slump. If firms do not hire them, then there are other ways to get money into their hands. And since they would spend it, to get money into circulation. And they would be better customers for employers. Remember, it is customers and potential customers that create jobs.

[Too Much Fed]

K's post said: "Too Much Fed: "K, let's take that a step further. Can you imagine an economy with no currency denominated gov't debt and no currency denominated private debt?"

I think so. As Nick once pointed out, nominal debt is a lot like glass (and equity like rubber? I forget). It breaks, and causes huge dead weight losses when it does so (the lawyers take half)."

By "break", do you mean it can be defaulted on or paid off?

And, "Could governments equity finance?"

"Your question is important because, as Nick discussed in a post a month or two ago, If you back the CB's balance sheet with nominal liabilities and the CB has no equity, then the value of the liabilities (money) of the CB can be undefined if they are not equal in quantity to the amount of nominal assets. So CB's need to issue equity, *or* they need to own a substantial amount of non-nominal assets (equity). But I don't think governments need to "equity" finance for my central bank proposal to work. The CB just needs to hold a lot of private sector equity securities relative to the amount of nominal assets."

I like to look at it these ways:

savings of the rich plus savings of the lower and middle class = balanced budget(s) of the various levels of gov't(s) plus the dissaving of the currency printing entity (CPE) with "currency" and no bond/loan attached

The CPE has a price inflation target. The CPE balance sheet looks like this:

assets = the amount of goods/services in the economy

liabilities = the medium of exchange supply (all "currency" and no gov't debt and no private debt in this case)

owner's equity = zero

Would an all "currency" economy be considered an equity financed one?

[Too Much Fed]

Min said: "If firms do not hire them, then there are other ways to get money into their hands."

It is called retirement and funding it correctly.

[Too Much Fed]

What is "money"?

Currency plus demand deposits because that is what circulates in the real economy. It does not include central bank reserves because they are a type of "medium of exchange" that only circulates in the banking system covered by the fed.

[Min]

Min said: "If firms do not hire [the unemployed], then there are other ways to get money into their hands."

Too Much Fed: "It is called retirement and funding it correctly."

20 is a little young to retire. ;)

[Too Much Fed]

Min, that is true, but there should be some people in their 50's or 60's who have worked long enough (30 years or so) they should be able to retire. If those people retire, then either someone moves up a job (promotion) and someone at the bottom gets hired, or someone gets hired.

[Min]

Too Much Fed: "Min, that is true, but there should be some people in their 50's or 60's who have worked long enough (30 years or so) they should be able to retire"

Ready or not, many of them are being forced to retire. And instead of talking about reducing the age for getting Social Security, politicians are talking about raising it!

[Greg Ransom]

Your argument is contradicted by facts.

Note well. Lawrence White and others have testified in the past few week before Congress on the topic of free banking.

"By continuing to advocate Hayek rather than some more moderate position, you (by P1) fail to get yourself heard by the government."

[Greg Ransom]

I'm not a scholar of the history of private banking, but my impression is that this is more misleading than factual:

Nick writes,

"Greg: I'm not against private monies. But, as far as I can see, truly independent (i.e. not redeemable at par in government money) seem to fail the test of the market. With a few exceptions, people just don't use LETS and similar monies."

Where it has been allowed -- and has not been crippled by pathological regulations -- private banking & private money did very well.

[Greg Ransom]

Determinant, the whole issue is made a cartoon by your discussion, and that of Skidelsky, who is notorious for not knowing his Hayek.

There is a lot to this -- and much of the heart of Hayek and the complex relation of his explanatory mechanism to the great complexities of history are completely missing from you simplistic account.

Your sense of certainty on the matter is completely unearned.