Two extreme fiscal/monetary worlds

I want to imagine two extreme worlds, at opposite ends of the spectrum of possibilities.

In the first world (the "fiscal" world), all financial liabilities of the government are non-monetary liabilities, "bonds", that cannot be used as media of exchange. People use something else for money. Maybe gold, or maybe money issued by commercial banks. The government is no different from a local government, or a corporation with the power to tax. Government bonds pay interest.

In the second world (the "monetary" world), all financial liabilities of the government are monetary liabilities, "money", that are used as media of exchange. Unlike local governments, or corporations, the government does not issue "bonds". Government money may or may not pay interest.

(The real world lies somewhere between those two extremes, because most governments have both money and bonds as liabilities.)

Let's start with both worlds in equilibrium, and with balanced budgets. Now let's suppose the government wants to increase spending without increasing taxes. What happens to interest rates on government liabilities?

In the fiscal world, the government may be forced to pay a higher interest rate on government bonds to persuade people to buy more government bonds. Nobody would buy the extra government bonds unless they wanted to hold extra, or thought somebody else wanted to hold extra. And if we started in equilibrium, people were already holding as many government bonds as they wanted to hold. Unless it can persuade people to want to hold extra government bonds, the government cannot run a deficit, because it cannot sell the extra bonds.

In the monetary world, the government would never be forced to pay higher interest rate on government money to persuade people to buy more government money. It might choose to pay a higher interest rate on government money to prevent inflation, but it wouldn't be forced to. Even though we started in equilibrium, and people were already holding as much government money as they wanted to hold. The government can always sell the extra money and run a deficit, even if it cannot persuade anyone to want to hold the extra money. Money -- the medium of exchange -- is weird like that. People accept it, not because they want to hold it, but because they know everyone else in turn will accept it from them, even if nobody wants to hold it.

Nobody ever turned down a government contract because they didn't want to hold extra government money and thought nobody else would want to hold extra government money. If they did, it would mean that government money no longer works as money.

The government in the first, fiscal world, might peg the rate of interest it pays on government bonds, and let the quantity of bonds it sells be demand-determined, and so the demand for government bonds in turn determines the deficit.

The government in the second, monetary world, might peg the rate of interest it pays on government  money. This does not mean that the quantity of money it sells is demand-determined, nor that the demand for government money determines the deficit.

The government in the first, fiscal world, has only one degree of freedom. Once it has chosen how many liabilities to sell, it cannot at the same time choose what rate of interest to pay on those liabilities.

The government in the second, monetary world, has two degrees of freedom. It can choose how many liabilities to sell, and it can choose the rate of interest it pays on those liabilities.

The government in the real world is a combination of those two extremes, and it can choose what combination it is. That gives it three degrees of freedom. It can choose how many liabilities to sell. It can choose what combination of liabilities to sell. And it can choose what rate of interest to pay on its monetary liabilities. Because money is different from all other assets, financial or otherwise.

Finally, imagine a "red/green" world that is exactly like the first, fiscal world, except the government issues two types of bonds --  red bonds and green bonds. And suppose that people like holding a mix of red and green bonds, but normally prefer holding green bonds to red bonds, so that green bonds normally yield a lower rate of interest than red bonds. And suppose there is a government-owned and controlled "central bond bank" that sets a rate of interest on green bonds, and lets people swap red bonds for green bonds, and vice versa, as many as they wish, at that rate of interest. So the total stock of bonds is determined by the government's deficit, but the stock of green bonds that people hold (or the mix of green vs red bonds) is demand-determined at that rate of interest chosen by the central bond bank. The government has only two degrees of freedom in the red/green world.

The real world is NOT like the red/green world. Too many economists, of all persuasions, think it is.

One more for the "inarticulate dork" pile. We all keep trying. One day we will all think it and write it clearly.

This was a bit rushed. I gotta go.

Comments

[Peter N]

The behavior of your fiscal world depends on some additional considerations

1) can there can be a Triffin dilemma? Can the demand for safe assets from country A exceed the amount of debt A wants to issue? This amount depends on A's risk adjusted estimate of the servicing costs compared to A's citizens willingness to bear them. Even if rates go negative, A may not wish to issue debt at a nominal profit because the risk adjusted ROI of the servicing costs is too high.

Obviously if A is also an issuer of a reserve currency demand for its debt denominated in this currency (A could obviously also issue debt denominated in a different currency, so the proportion of A's debt in its own currency is A's option) is likely to be high, because of the need for collateral denominated in the currency in which goods are priced. If oil is traded priced in dollars, then hedges in dollars involve no currency risk. The result is what we would expect. If you are the official issuer of a reserve medium of exchange (other issuers are possible through pegs), then the demand both for safe currency and safe debt denominated in that medium will be higher than they would be otherwise. By definition, the official issuer is 0 risk in the sense that it can always issue that currency at par.

So debt issued in a reserve medium of exchange is more likely to be subject to the Triffin dilemma, but the dilemma is possible for any sovereign debt.

2) What is the usefulness of the medium of exchange as a store of value? It could fluctuate in value or quantity or be a wasting asset (as in your sofa example).

3) Does bankruptcy law support repos? This is a bit beyond my pay grade, but it clearly affects the demand for safe collateral.

These issues may create additional degrees of freedom. This is very tricky stuff. You've got all sorts of players looking for ways to game the system or opportunities for arbitrage.

[Ron Ronson]

Suppose that the govt targets NGDP and wants to build a bridge.

In world 1: They borrow the money , build the bridge and adjust the tax rate to keep NGDP on target.
In world 2: They print the money , build the bridge then use interest rates or asset purchase to keep NGDP on target.

If banks suddenly magically acquire huge amount of reserves that they want to lend out

in World 1: As these reserves get lent out the govt can use tax and borrowing to keep NGDP on target
in world 2: The CB can either pay IOR to prevent these funds getting lent out or let the money get lent but sell assets to reduce the money supply to keep NGDP on target.

In all cases the govt can use either monetary or fiscal policy to hit the target - the only variable is who gains and loses as a result of the transfers in income depending upon the path chosen,

[marris]

This is a stupid quesiton, but why are we not in the red-green world? Is it because the Fed is color-blind [they are targeting the full yield curve]?

Should we give the Fed get another degree of freedom for every price it can fix? It probably has infinite degrees of "freedom," even though fixing some prices may cause the economy to blow up.

[Nick Rowe]

Ron: I would change what you wrote to: "In world 2: They print the money , build the bridge then use interest rates or adjust the tax rate to keep NGDP on target."

World 2 has one extra degree of freedom compared to world 1.

[wh10]

I don't like your use of the labels "fiscal" and "monetary" to distinguish these two worlds. Both are fiscal worlds - the government is issuing net financial assets.

If you want to distinguish based on the type of liability, it would be better to say "cash" vs "bond" worlds or something like that.

But I sense you're doing this to bias people towards your preferred macroeconomic policy, which starts with the world "monetary"...

[wh10]

*starts with the word

[wh10]

Although, to be fair, I guess perhaps we need a good definition of "fiscal policy" and "monetary policy" first. Because, when the central bank changes interest rates, they're changing net financial assets as well, through a valuation effect. So there's not such a clear break between fiscal and monetary policy. On the other hand, policies such as QE are asset swaps (putting aside its impact on interest rates), in contrast with fiscal policy.

[Becky Hargrove]

wh10, in the long run all things monetary carefully protect all things fiscal

[JP Koning]

Just playing around with your ideas here...

In the monetary world, if the government issues more money but doesn't raise interest rates, then inflation is the result. I think we agree on that. But say we think of inflation as a fall in the present price of money relative to its expected future value. This adds an expectation of price appreciation to money's already existing fixed interest yield, both of which sum up to the total return on money holdings. So the government can lure people to hold additional money at the same interest rate only by offering an extra bit of capital gains.

While I agree with you that in a monetary world the government isn't forced to pay a higher interest rate, it is forced to pay a higher total return in order to get people to accept money. This is the same as in a financial world, since the government can only get people to accept additional bonds with a face value of $1000 by promising a higher coupon, or a higher total return.

[marris]

@JP Koning

> it is forced to pay a higher total return in order to get people to accept money. This is the same as in a financial world, since the government can only get people to accept additional bonds with a face value of $1000 by promising a higher coupon, or a higher total return.

I think that's wrong (or at least not what Nick is trying to say). I think Nick is distinguishing between (1) getting people to accept money and (2) getting people to hold money. They are not the same. If the government wants to increase spending in the monetary world, they only need to do (1). They get (1) for free because they are paying in the medium of exchange. But they don't get (2) for free. The new money will hot potato through the economy until prices rise to restore the system to equilibrium.

Paying interest on money is a way to slow down the hot potator (or reduce the number of rounds it makes before the system settles into a new equilibrium).

I think.

[Nick Rowe]

wh10: fair criticism. Yep. My choice on names was not obviously the best.

JP: I think that's right. But what marris says is correct too. In the second world, something (P and/or Y) will eventually adjust, as a consequence of the hot potato. Another way of looking at it would be to ask Ron Ronson's question: how many ways does the government have of hitting its NGDP (or whatever) target?

[Andy Harless]

Consider a world where the government can issue green bonds and hold the proceeds in a vault instead of spending it, so there's no longer a constraint that the sum of red bonds plus green bonds equals the real deficit. There's your third degree of freedom back again. Is this world materially different from the real world?

I contend, no, provided we are willing to define the prized characteristic of the green bonds as something called liquidity. The government (including the central bank) can choose how much liquidity to produce, and that is separate from the decisions of how much to borrow for spending purposes and what interest rates to pay on different liabilities.

But does this third degree of freedom have any macroeconomic implications? It certainly has microeconomic implications: liquidity is a product, just like smartphones or nursing services, and we should try to produce the optimal amount. It might have macroeconomic implications in the same sense that, for example, the oil market has macroeconomic implications: liquidity is a critical product, and its availability changes the set of options available to macroeconomic policymakers. (In particular, when your interest rate policy hits the ZLB, you might be able to use liquidity as stimulus in its own right, and you might also use it in an anticipatory way if you are afraid of hitting the ZLB, much as the Fed did in late 2008.)

But I think (and I expect Nick will disagree) that liquidity is, in some sense, just another very important product, much like oil, rather than a fundamental determinant of macroeconomic outcomes. If it weren't for the ZLB (i.e., if the expected inflation rate were always high enough, or if we had a Kimball-style electronic money system), interest rate policy would be fully sufficient to control NGDP, and liquidity policy could be carried out separately, as a microeconomic concern rather than a macroeconomic one.

[Daphne Millar]

Why does government not choose to live in the second, monetary world? It is true that the power that choice gives it could be misused, with inflation as a likely outcome. But it does not need to misuse the power. And the attractions are obvious.

[Nick Rowe]

Andy: "Consider a world where the government can issue green bonds and hold the proceeds in a vault instead of spending it, so there's no longer a constraint that the sum of red bonds plus green bonds equals the real deficit. There's your third degree of freedom back again."

The government in that case sounds very similar to an individual, or a local government, or corporation, that cannot print money, but can borrow money, and which holds a varying stock of money to cover transactions. I expect the difference would be that a national government might (say) buy or sell gold to vary the stock of gold it holds to try to influence economy-wide conditions, like the price level.

Ummmmm. This is weird, but I'm beginning to feel I don't understand my own post. I think that everything I said is right. But when I think about writing down a model, with a demand for bonds that depends on the rate of interest bonds pay, and other rates of interest, and income, and stuff, and writing down a demand for money that, at least qualitatively, looks just the same, I can't see any qualitative difference between my first and second worlds. I think this goes back to the JP vs marris discussion in comments. Comparative statics vs monetary disequilibrium hot potato process. And there's some sort of fallacy of composition issue in the equilibrium modelling of money. Because you can't force any individual to hold the hot potato but you can force all individuals to hold the hot potato. Maybe it's just too early in the morning.

Perhaps the answer to my self-muddle is the old monetarist saying: the government can choose the nominal quantity of money but not the real quantity of money. If the price level were perfectly flexible, the distinction would disappear (and there would be no hot potato)?

[Nick Rowe]

Daphne: most governments do choose to live in a combination of the two worlds, to give them that extra degree of freedom so they can control macroeconomic variables better.

[Nick Rowe]

I'm a bit surprised I haven't got more pushback on this post, from the MMT guys for example. Not that they are the only ones who make what I think is the big mistake of saying that the real world is like the red/green world, in which the stock of government money is demand-determined, because a lot of very orthodox central bankers and New Keynesians say the same thing.

Brad DeLong says he thinks it is worth reading, which reinforces my belief I'm onto something, because he's good on this sort of stuff. But where are the people saying I've got it totally wrong?

[wh10]

Nick, here’s my go, since you asked for it. It’s written as if I was responding point by point, before having read the whole post. Honestly, I don’t think I am disagreeing with you, and I am not sure why you would expect push back from “the usual suspects.” However, I don’t quite understand how the red/green world is working, so I would greatly appreciate some clarification, if you think my response below is misunderstanding your post.

---

I'd like to start with a side note. I think your “fiscal” world is similar to the EU before the crisis. Essentially, by calling the government like a local government, you’re saying there is no central bank to back the bonds of that government (which would issue liabilities as currency). I think *that* is the most important difference between your two worlds, and for that matter, countries like the U.S., UK, Japan, and, I think, Canada. I explain below (if you need explaining).

“In the fiscal world… Nobody would buy the extra government bonds unless they wanted to hold extra, or thought somebody else wanted to hold extra.”

Ah! I think “thought somebody else wanted to hold extra” is a key phrase! This is where the central bank comes in, as per my statement above, because the central bank can act as that “somebody” but with UNLIMITED potential to hold extra. I see the issue as driven by a type of arbitrage condition, in which if you know the central bank will act to keep rates down to x%, you’re going to buy bonds that rise above x% to capture that free profit. If the fiscal world had that capability, it would have the 2 degrees of freedom that the monetary world had (plus the additional degree of freedom to choose the mix of currency and bonds, such that it is really equivalent to countries like the U.S., UK, Japan, and, I think, Canada). Marinner Eccles agrees with me: “So it is an illusion to think that to eliminate or to restrict the direct borrowing privilege reduces the amount of deficit financing. Or that the market controls the interest rate. Neither is true…” (See the full quote at http://monetaryrealism.com/marriner-eccles-explains-it-all/ )

“The government in the real world is a combination of those two extremes, and it can choose what combination it is.”

Let’s clarify – you are talking about government like the U.S., UK, Japan, and, I think, Canada, right? Not all government have the capabilities you are referring to. Also, are you saying these governments can choose what combination of liabilities to sell by having the central bank engage in policies like quantitative easing? If so, I think I am with you.

“Finally, imagine a "red/green" world that is exactly like the first, fiscal world, except the government issues two types of bonds -- red bonds and green bonds. And suppose that people like holding a mix of red and green bonds, but normally prefer holding green bonds to red bonds, so that green bonds normally yield a lower rate of interest than red bonds.”

So we can say green bonds are like those with a maturity of 1 yr in the real-world, and red bonds are like 30 yr bonds in the real-world. People in the real-world are more willing to hold 1 yr bonds than 30 yr bonds.

“And suppose there is a government-owned and controlled "central bond bank" that sets a rate of interest on green bonds, and lets people swap red bonds for green bonds, and vice versa, as many as they wish, at that rate of interest. So the total stock of bonds is determined by the government's deficit, but the stock of green bonds that people hold (or the mix of green vs red bonds) is demand-determined at that rate of interest chosen by the central bond bank. The government has only two degrees of freedom in the red/green world.”

Wait, so how does this work, in more detail? Is this bank buying the green bonds with currency and holding the rate down to x%? And what is this swapping? How does that work?

“The real world is NOT like the red/green world. Too many economists, of all persuasions, think it is. One more for the "inarticulate dork" pile.”

I can’t really speak to this since I don’t quite get what’s going on in the red/green world. Additionally, even if I did, I am not quite sure I would know why you don’t think the real world is not like the red/green world, since you didn’t explain why. Overall, I thought it was pretty articulate, except the red/green part :).

[wh10]

Sorry, would like to clarify a poorly written sentences (although there are plenty more) -

"you’re saying there is no central bank to back the bonds of that government (which would issue liabilities as currency)"

I am saying that the central bank would issue liabilities as currency, not "that government."

[Dan Kervick]

Nick, we're imagining that in the monetary world all government purchases of goods and services are voluntary on the part of the both the buyer and the seller? Or does the government get to dictate the terms of the sale?

[RebelEconomist]

"The government in the real world is a combination of those two extremes"

Really? How is the government in the monetary world (in a normal advanced economy I mean, in which the government is effectively a user of central bank money)?

[JP Koning]

Marris: "I think Nick is distinguishing between (1) getting people to accept money and (2) getting people to hold money. They are not the same. If the government wants to increase spending in the monetary world, they only need to do (1). They get (1) for free because they are paying in the medium of exchange. But they don't get (2) for free. The new money will hot potato through the economy until prices rise to restore the system to equilibrium."

Just playing around with your language, but it seems to me that the government only gets (1) for free if no one is aware that the government is issuing money. If everyone knows whats up, they'll immediately lower prices in anticipation, in which case the government needs to spend more money to get the same amount of services. This raises its financing costs while simultaneously offering money holders a higher return. The higher return comes in the form of expected appreciation of money, or a capital gain, which relates back to my initial comment.

The government can get (1) for free in a 100% fiscal world too. As long as they create new bonds and surreptitiously sell them in the secondary market for bonds, people will accept them at current market value. After the issue, bond price will fall (ie yields will rise), but only slowly through a hot potato effect as people try to offload the excess. On the other hand, if the government were to have pre-announced their intention to issue new bonds, prices would immediately fall in anticipation of the event, forcing the government to pay higher interest rates.

[JP Koning]

"If the price level were perfectly flexible, the distinction would disappear (and there would be no hot potato)?"

Nick, the hot potato effect doesn't actually have to happen to be the hot potato effect, right? Isn't that Chuck Norris? He threatens to set off a chain of hot potato effects, but since people see Chuck coming, the hot potato effect is self-realized without any hot potatoes actually being emitted?

[rsj]

I would propose that having your liabilities accepted is a function of your power/stability, so that if government liabilities are not accepted then no one's liabilities are.

We accept bank liabilities as payment because the government guarantees them. It is illogical to accept bank but not government liabilities, unless you are in some euro-zone arrangement where there is another supra-national entity that guarantees banks other than government. I.e. a hierarchy of governments.

So really, your two models are gold standard on the one hand and a Wicksellian credit economy on the other. If government obligations are not trusted, then no one's obligations are trusted and everyone must transact only in gold. If government obligations are trusted, then a lot of other obligations are dragged in as well (agency paper, money market funds, traveller's checks, trade credit, etc.).

It seems clear to me which is the model I would use when describing a modern economy.

I do not see a smooth continuum from one to the other. I see an isolated point {*} describing the a theoretical pure gold standard economy, which never existed, as transactions have always been done on a primarily credit basis, and a continuum in the second case, in which on one extreme, there are no credit constraints and everyone's liabilities are accepted equally, and on the other extremen, only the government's liabilities are accepted as a means of payment. So let's just drop the first case from consideration entirely as a still-born theoretical model with no historical roots, and examine the second model to determine where on that continuum we lie.

[Nick Rowe]

Dan: "Nick, we're imagining that in the monetary world all government purchases of goods and services are voluntary on the part of the both the buyer and the seller? Or does the government get to dictate the terms of the sale?"

That is the key. Short answer "yes".

Longer answer: "Yes, but there's something kinda sorta like a fallacy of composition thingy here. Each *individual* is willing to *accept* extra money, because he doesn't have to *hold* that extra money, since he can pass it onto another individual. But all individuals *in aggregate* do have to hold that extra money. The government can get any individual to say "yes" to the trade. But if all individuals had a collective voice they might say "no" to the trade, because they don't want to hold extra money.

Still longer answer: this needs a post.

wh10: "Let’s clarify – you are talking about government like the U.S., UK, Japan, and, I think, Canada, right?"

Yes. And not Germany, France, Greece, Spain, or "local" governments.

More later.

[Nick Rowe]

Curses! In my answer to Dan I meant :"Yes, I'm assuming voluntary exchange for the individual buyer and seller."

[Frank Restly]

Nick,

"In the first world (the "fiscal" world), all financial liabilities of the government are non-monetary liabilities, "bonds", that cannot be used as media of exchange. People use something else for money. Maybe gold, or maybe money issued by commercial banks. The government is no different from a local government, or a corporation with the power to tax. Government bonds pay interest."

"Let's start with both worlds in equilibrium, and with balanced budgets. Now let's suppose the government wants to increase spending without increasing taxes. What happens to interest rates on government liabilities?"

"In the fiscal world, the government may be forced to pay a higher interest rate on government bonds to persuade people to buy more government bonds. Nobody would buy the extra government bonds unless they wanted to hold extra, or thought somebody else wanted to hold extra. And if we started in equilibrium, people were already holding as many government bonds as they wanted to hold. Unless it can persuade people to want to hold extra government bonds, the government cannot run a deficit, because it cannot sell the extra bonds."

Interest payments on bonds are a form of government spending. And so I think you need to clarify what you mean by "the government wants to increase spending". Also you need to clarify what you mean by "raising taxes". If a government does not attempt to increase tax revenue and wants to increase government spending, the only method available to it is selling additional bonds. That works until all spending is in the form of interest payments.

[Alex Godosky]

JP:

Just playing around with your language, but it seems to me that the government only gets (1) for free if no one is aware that the government is issuing money. If everyone knows whats up, they'll immediately lower prices in anticipation, in which case the government needs to spend more money to get the same amount of services. This raises its financing costs while simultaneously offering money holders a higher return. The higher return comes in the form of expected appreciation of money, or a capital gain, which relates back to my initial comment.

Prices (except asset prices, generally) are sticky.

Also, I think you mean "raise prices in anticipation", right?

[Peter N]

People and organizations differ in their ability to avoid holding money. For instance if I have a system of net settlement for my customers, then I only have a net exposure to the banking system. Any transactions between customers that net out, I can settle myself, so velocity measured at the customer level can be much higher than what appears at the bank level.

Those who can do this would seem to have an advantage over those who can't. If so, what would the expected log term effect be? Is this mirror image rent?

[Ritwik]

Nick

I'm not quite sure what your reason for the assertion that the real world is not like the green/red world is.

The best argument you can make is that if one rate of interest is necessarily fixed (historically at 0) and the other is allowed to respond to market beliefs, demand for the asset that pays the fixed interest is controllable by the monetary authority and hence, the stock isn't demand-determined in any meaningful sense.

But in world where the monetary authority tried to control both rates of interest, even the power it has over inflation expectations isn't good enough to block the channel of reflux through arbitrage between the two assets.

The relevant degrees of freedom are 4. Total qty of liabilities. Real rate of interest on money. Real rate of interest on bonds. Mix of money and bonds.

The fiscal-financial-monetary complex of the government must choose any three of these 4. Arguably, given that government liabilities may interact in strange ways with private capital formation, the ovreign may only have 2 degrees of freedom.

[Ritwik]

As a corollary, the people that you expect would have criticized your post (MMTers etc.) do believe in 3 degrees of freedom. They just choose a different set of three from you.

[Nick Rowe]

Ritwik: The red/green world has only two degrees of freedom. If you say there are 4 degrees of freedom in the real world, then the real world cannot be like the red/green world.

[Ritwik]

Who sets the rate of interest on red bonds in that world?

[Ritwik]

I'm saying that the total degrees of freedom are 4. Sovereigns have to choose which 3 they want to fiddle with. In your world, or the MMT world. A neoclassical/long-run analysis may even contend that sovereigns have only 2 degrees of freedom, because given an optimal rate of private capital formation, to be able to hit it may require giving up an additional degree.

[Nick Rowe]

Ritwik: Let's go back to the simple case. My first world. How many degrees of freedom does the government have? I say one. It can either choose the deficit or it can choose the rate of interest on bonds. Not both.

Now consider the red/green world. It can choose that red rate and the green rate, and let the market decide the quantities of both and the deficit. Or it can choose the green rate and the deficit, and let the market choose the red rate and the mix. Or any other combination of two.

[wh10]

Nick, if I am understanding you correctly, based upon your interaction with Ritwik, I think I agree with you. We do not live in a red/green world. The reason is that we have a central bank that issues currency, gives people confidence the govt can't default (putting aside political factors), and can peg the rates on red and green bonds to prove it. The CB normally doesn't have to peg the rates, though, because people realize that they always can.

[Ritwik]

Nick

Sure. Now, why do you say the world is not like the red/green world?

Here's how I think about things :

An MMTer believes that the government can (and should) set the red rate(subject to the external constraint of macroeconomic balance), green rate and the total quantity. Thus, the red/green mix is now endogenous.

You believe that the government can (and should) set the green rate (subject to macroeconomic balance), the total quantity and the mix. Further you believe that the quantity of green is in and of itself reflective/causative of macroeconomic balance.

A New Keynesian believes that the government should set the green rate and the total qty. The red/green mix is now endogenous and the red rate cannot be set separately because the external constraint of macroeconomic balance forces the red rate given the total qty of liabilities. Or vice-versa.

The New Keynesian has the same definition of macroeconomic balance (Savings/Investment for flows or Optimal rate of private capital formation for stocks) as the MMTers but affords himself one degree of freedom less because his micro is Neoclassical. Your micro is Neoclassical but you allow yourself one degree of freedom more than the New Keynesian because your definition of macroeconomic balance (right qty of green) is not external to the system, but contained within it.

Is that a fair characterization? Where's the confusion? The New Keynesian appears the most right to me, at least on this count.

[Frank Restly]

Nick,

"Now consider the red/green world. It can choose that red rate and the green rate, and let the market decide the quantities of both and the deficit. Or it can choose the green rate and the deficit, and let the market choose the red rate and the mix. Or any other combination of two."

R = Red Bond Issuance
G = Green Bond Issuance
r% = Red Bond Interest Rate
g% = Green Bond Interest Rate
EX = Non-Interest Expenditures
TX = Tax Revenue

EX + r% * R + g% * G - TX = R + G

Assuming that for each type of bond, the market choses either:
1. Issuance
2. Interest rate

If the market choses issuance for both, the government is left with setting interest rates on both.
If the market choses interest rates for both, the government is left with setting issuance for both.
If the market choses interest rates for red, and issuance for green, the government sets interest rates for green and issuance for red and vice versa.

Of course, this assumes that both red and green bonds are coupon bonds. If they are not then things change. If red bonds are coupon type and green bonds are accrual type, then the government gains a degree of freedom.

EX + r% * R - TX = R + G

Here, the government can set the green issuance and the green interest rate and let the market set the duration. The government can then either set the red issuance or the red interest rate.

[Dan Kervick]

Nick, OK, but that seems to place an artificial limit on fiscal power and bond sales. It's hard to see why a government that has the power to use the law to force people to turn over things of value (i.e. taxes) would decline to use that power to dictate or influence the price at which it buys things of value - or at which it sells bonds.

"Selling" a bond could be seen as similar to taxing somebody, and then giving them a rebate later. Whether the future rebate should be an amount greater or less than the original tax is a public policy choice. The decision to offer positive interest bonds in a voluntary market is is a policy choice to the benefit of people who happen to possess a savings surplus. There is no necessity that such a service be offered.

[Alex Godofsky]

Dan:

It's hard to see why a government that has the power to use the law to force people to turn over things of value (i.e. taxes) would decline to use that power to dictate or influence the price at which it buys things of value - or at which it sells bonds.

Except that we readily observe the government declining both. Even in the places where the government actually tries to impose some degree of price controls (e.g. Medicare) it doesn't actually seem all that willing to turn the screws hard enough to insulate it from changes in the general price level. Yes, the government could conscript doctors, but it doesn't and probably won't.

[JP Koning]

Alex: "Prices (except asset prices, generally) are sticky."

Ok. But then the difference between Nick's two worlds is really just about slow vs fast adjustment. The government can finance itself cheaper via money issuance because the world is slow to anticipate and mark prices up ahead of time. Since bonds react immediately to new issue announcements the government can't run the same game with bonds. Once the market catches on to money issuance, the worlds are alike.

"Also, I think you mean "raise prices in anticipation", right?"

Yes, you're right. Whoops

[Dan Kervick]

Alex, yes that's true. But Nick is giving a very general theoretical argument. In his monetary world, the government is not a market play granted the power to determine what the medium of exchange is, and to issue either interest-bearing or non-interest bearing units of that medium. But in the fiscal world it is suddenly required to behave like any other market participant, and accept market pricing in its procurement of goods and services and its sales of debt. Looks to me like he has artificially baked the differences in degree of freedom into the cake.

[Alex Godofsky]

I don't agree. We actually see the fiscal world all that time - this is how state and municipal governments behave in the US. If California wants to issue $X billion in bonds, it faces very real constraints on what rate of interest it must pay on those bonds. Too low and they won't sell.

California could choose to pay more interest on those bonds than required, but it's not clear that this would serve any public function.

The reason the monetary world government can choose to pay "too little" interest is because 1) its liabilities are used as money and 2) the money prices of goods and services are sticky. This isn't artifice or sleight-of-hand on Nick's part; this is a fundamental part of his model, and one with a lot of empirical justification. At the wrong bond interest rate the fiscal world bond market will fail to clear more or less immediately; at the wrong money interest rate it will take some time for the output market to stop clearing.

In the long run the monetary world government would face constraints on the quantity and/or interest rate, because too much inflation or deflation eventually causes people to abandon your currency. But history shows it actually takes a lot of inflation or deflation to do that.

[Nick Rowe]

JP: "Ok. But then the difference between Nick's two worlds is really just about slow vs fast adjustment."

I'm not sure if that is correct. I don't think it is.

Assume the equilibrium condition for bonds is B=kPY (Cambridge k formulation).

Assume the equilibrium condition for money is MV=PY

We can assume that k and V are functions of r, and that v=1/k

It all looks identical, in equilibrium. Just different ways of saying the same thing. But it isn't.

Suppose I wanted to hold one month's income in bonds.

Suppose I wanted to hold each dollar of money for one month.

Start in equilibrium.

Offer me more bonds, and I don't want them, unless you pay higher interest on the bonds so that my k increases.

Offer me more money, and I say "OK", and hold it for one month then trade it, so my PY automatically expands.

I'm trying to get my head clear enough on this for a second post.

[Nick Rowe]

Think of kids sitting in a circle around a campfire. Assume a hot potato doesn't burn you unless you hold it for more than 10 seconds. Each kid will voluntarily accept a hot potato, as long as he can pass it on to the next kid in less than 10 seconds, who in turn voluntarily accepts it as long as he can pass it on in less than 10 seconds. There is no limit to the number of hot potatoes they will voluntarily accept, as long as they all expect the circle to remain unbroken.

[Steve Roth]

I agree with WH10. What if this post were rewritten without the word "money," instead talking about different financial assets -- currency, reserves, bonds, etc. All of which *embody* "money," a.k.a. credit (in that word's vernacular sense at least) or exchange value.

[Alex Godofsky]

Nick:

Offer me more bonds, and I don't want them, unless you pay higher interest on the bonds so that my k increases.

Offer me more money, and I say "OK", and hold it for one month then trade it, so my PY automatically expands.

I think you are skipping over a point here. If all of the prices of money were as flexible as the money price of bonds is, then I wouldn't say "OK" - the moment Mr. Central Banker walked over to my store with a roll of freshly-printed $100 bills, I would tell him "nope, you're going to have to pay more than the sticker price".

[JP Koning]

I agree with Alex. It just seems to me that if we take the kids-around-fire analogy to its logical limit, why would inflation ever occur? All the kids know in advance that the next kid will take it knee-jerk like without increasing prices ahead of time, so they'll accept a new potato from the potato issuer without increasing their own prices.

On the other hand, if kid A anticipates that kid B might increase his price before accepting the hot potato from A, then A will act ahead of time and increase his price before kid C can pass him the potato. Then C will do the same. We move back down the chain until the potato issuer already faces higher prices the moment the campfire sees the issuer reach into his/her pocket to put another potato into circulation. At this point the potato issuer can purchase less real goods and services with each new potato issued. But that is the same constraint faced by a bond issuer.

[Determinant]

Completely off-topic, but still of interest to Nick (I hope):

I just registered as a delegate to the Federal NDP Convention in Montreal in April to represent my riding association.

You have argued for NGDP Targeting, Nick, so here's the relevant plank from the NDP Policy Guide

1.5(b) (Jobs & Monetary Policy)

New Democrats believe in:

Monetary policy that preserves and creates jobs and which strikes a balance between price stability and full employment.

This may deserve its own thread. Please tell me why a social democrat lefty should embrace NGDP targeting, why it is consistent with the NDP's policy statement and why it would help, not hurt, the lower and middle classes, primarily wage earners and retirees.