« How to videos by MumblingProfessor | Main | Dealing with the Baby Boom: You Only Die Once »

Comments

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

But it *really* is not a put. Like I said in your previous post, it would be a put *if* the bearer had the right to sell it back to the CB at the target CPI. But they don't. They have the right to buy the consumption basket at the realized CPI, whatever that happens to be. You explicitly allowed for variance of the CPI process. Where is this mechanism by which *if* the CPI differs from the level target, the holder gets to buy the consumption basket at the target level? It doesn't exist. So there is no put.

Also if there were a put, why wouldn't the bearer also be short call? After all, the CB makes a commitment to no *less* than 2% inflation as well as no more. If the two-way commitment were really contractual, bearers would be long put and short call. I.e. they would be short a money forward contract (or equivalently, long a consumption basket forward). This is not actually the case though, since the government does not have the power to guarantee consumption (only capital assets can transport consumption) which is why inflation commitments often fail, especially when made in huge quantities. I have said before that inflation targeted money is a doomed and dangerous consumption forward contract and that it's the institutionalization of the consumption guarantee that is at the root of macro instability.

K: Suppose a company promised to buy back its shares in the open market if the price ever fell below the strike price. It would be as if I had a put on those shares, even if I don't sell them myself.

"Also if there were a put, why wouldn't the bearer also be short call?"

If I understand the terminology correctly, I think you are right. It's symmetric, under a symmetric inflation target. But since I'm looking only at the Bank of Canada's liability, and since it can print new notes for (almost) free, I don't need to look at the call option. I think.

Very brilliant. But I'd like to ask you, Nick, what's is the main consequence for MP

Nick: "It would be as if I had a put on those shares, even if I don't sell them myself."

Not really. The company can, if they want, target its share price (so long as it's a low price) on a given future date by committing to paying the necessary dividends. By doing that they might depress the uncertainty of the stock price on that date. If they reduce the uncertainty to zero, then options expiring on that date will be worthless (and debating whether you hold a put is pretty pointless).

But lets assume that they aren't perfectly successful at targeting the share price. In that case you will be left selling the shares at whatever the market price turns out to be. If, on the other hand, you actually had a put, you'd be able to sell your shares at the target price regardless. That's totally different.

The real consumption forward is the real return bond. You are trying to argue that money (or a regular bond) is a consumption forward (or put) under inflation targeting. All you are really saying is that under successful inflation targeting there is no difference between a regular and a real return bond, which is true. But you can't claim that in a world where inflation targeting can fail, that the two are the same.

I think K is in the right of it, Nick. I mean, you can only talk about putting currency to the CB at strike under the assumption that the CB intervenes as soon as the CPI deviates infinitesimally from target. But that is equivalent to the assumption that inflation targeting is continuously successful, in which case the options have no volatility and therefore no value. (It strikes me that, financially, the CB has a perverse incentive to miss it's target, since doing so allows it to buy money cheaply and sell it dearly.)

Anyway, if you are looking for a way to value a CB relaxing the assumption of perfect inflation targeting, you might do something with a first-passage structural approach. The value would be the time integral of seignorage earned under the probability-weighted integral of stopping times (the CB stops when it runs out of assets to sell for money, or, less plausibly, when the world runs out of assets it can buy.)

Nick, Disregard my comment at the previous post, or perhaps combine it with what you have here.

You are right that there is no liability associated with the actual $20 bill. And you are right that there is a liability associated with the possible need to contract the monetary base to keep inflation at 2%. But there's also an asset associated with the potential need to expand the base to keep inflation at 2%. Indeed if the base trends upward over time (and it does) those assets quite likely exceed those liabilities. Hence when I speculated in the previous comment section that the sale price of a central bank in an auction would be roughly equal to the monetary base, I understated things in one respect; the sale price might be even higher, in anticipation of future seignorage. Working in the other direction is the fact that investors might demand a higher rate of return that what's earned on government bonds, due to the risk.

I think you are doing a great job of gradually zeroing in on the answer. I don't know if puts are the right term, but either way these derivatives involve both assets and liabilities, with the net effect ambiguous. But the cash itself is not a liability. Or if it is, then there is some other asset that should be on central bank balance sheets, which is roughly equivalent to the base.

Luis: thanks! I confess I wasn't thinking of useful policy implications. Just trying to get my head around something that had always puzzled me. But if central bank solvency and independence is a concern, the value of its liabilities matters.

Phil and K: OK, since the BoC doesn't target inflation exactly, my put option metaphor won't work exactly either. But at this stage, if I can improve on "liability=money base" then I'm way ahead of the game. And yes, it's only "as if" there were an American put option, even if inflation were always exactly on target.

If the BoC wanted to maximise profits, it would choose a much higher inflation rate, as well as bring in 100% reserve requirements and stop paying interest on reserves.

I don't agree with the put option analogy.

The central bank is obligated to make a paymetn to money holders if they take a loss on the money?

No, the central bank is obligated to maintain the value of the money by reducing the quantity to meet the demand necessary.

All debtors have to reduce their debts when creditors no longer wish to lend to them.

Now, with a fixed quantity of base money, if people don't want to lend to the central bank (or government) by holding this base money, then it will depreciate in value. If the central banks (and government) are obligated to prevent this, then they reduce the debts.

It is possible, of course, to change the form of debt. For example, if will lend to the government with one year bills and not 3o year bonds, the government can issue 1 year bills to pay off the 30 year bonds when they come due.

The same thing with currency. If people don't want to lend to the government as much by holding currency, then the government can pay that off by issuing 1 year bills or 30 year bonds or whatever.

When interest is paid on most base money, if this causes some kind of puzzle--it isn't really money--then the problem is with the framing of money.

And, of course, treating base money as a liabilility makes it similar to all the other money--the checkable deposits.

Further, suppose we have a gold standard then switch to a price level target. Everything is the same, but rather than open market operations aimed at keeping the price of gold on target, it is a broader price index. But now, the base money is no longer a liability and instead there is a put option on the CPI?

Of course, if we instead must have the government print money as needed with no worries about its purchasing power, or else, but base money on a constant growth path, so that if the real demand for it changes, then its purchasing power will change, then maybe it is no longer a liability.

Notice that the two alternatives are the ones that interested old monetarists. Out of control spend money like they want or quantity rule.

Scott: "But there's also an asset associated with the potential need to expand the base to keep inflation at 2%. Indeed if the base trends upward over time (and it does) those assets quite likely exceed those liabilities."

Yep. Agreed.

Bill: "The central bank is obligated to make a paymetn to money holders if they take a loss on the money?

No, the central bank is obligated to maintain the value of the money by reducing the quantity to meet the demand necessary."

I'm saying that those two things are effectively the same (unless you can vary the interest paid on money). The payment the central bank makes to money holders is buying the money back from them, which also reduces the quantity of money to meet the demand at the existing price.

Nick:

If you want to introduce the option vocabulary into the discourse, you may want to take a look at "the real options" (http://en.wikipedia.org/wiki/Real_options_valuation).

I personally find the attempts to apply the financial options methods and experience elsewhere misguided and futile, but who knows ? I may be wrong !

Nick:
Your basic insight seems to me to be correct, but where does it take us?
Hugh

vjk: thanks. I have found that thinking in terms of options, just occasionally, helps me clarify my thinking.

Hugh: Good question. I don't know. This post was a bit of an accident, when the idea just came to me. I haven't thought that far ahead yet.

You say (rightly I think) that the monetary base is not a liability of a central bank. You then seem to say that the OBLIGATION on a central bank to churn out more monetary base in the event of inflation falling below 2% IS A LIABILITY.

I don’t see that the potential obligation to turn on the printing press is much of a liability. And the potential obligation on the government / central bank machine to do the opposite (i.e. confiscate dollars from citizens via extra taxes) is not a liability either.

I’d be happy to take over the above two “liabilities” from the Canadian central bank / government anytime. Please send suitable plates for turning out $100 bills. Plus a pallet load of income tax forms which enable me to demand money with menaces from Canadian citizens would be appreciated. I look forward to doing business with Canada.

Ralph: "I don’t see that the potential obligation to turn on the printing press is much of a liability."

It isn't. That's an asset.

"And the potential obligation on the government / central bank machine to do the opposite (i.e. confiscate dollars from citizens via extra taxes) is not a liability either."

Central banks buy back notes. The obligation to do so is a liability. Governments take back notes. The ability to do so is an asset.

The car analogy is flawed in that the use value of the car doesn't depend on the existence of the put option. The use value of the car is entirely independent of your future actions after you sell it, whereas the use value of money is entirely predicated on the anticipated future actions (or inactions) of the central bank. This post ignores the fact that money provides useful services, and thus it leaves no way to account for the value of those services.

So I'm going to stick with my position in the comments to the earlier post: money is a liability because it represents an obligation to provide services, and the profit earned on those services shouldn't be recognized until the services are delivered. (Again, it's an unsecured non-recourse liability, so the central bank can also profit by defaulting and failing to provide the promised services, but that's a one-shot profit which destroys the central bank's future profit potential. Generally limited liability implies that the owners of a corporation can profit by its insolvency, but we still recognize the full liabilities while it operates.)

Now I do think Scott is right that "the sale price of a central bank in an auction would be roughly equal to the monetary base" and "might be even higher," but I don't think this means that money isn't really a liability. A central bank has a franchise that is worth something. Since its liabilities are unsecured and non-recourse, the franchise must be worth at least as much as its assets, or else it the buyers would choose to default and liquidate.

But I don't think the franchise value comes from the liabilities' being fake. Rather, there is a bunch of equity that isn't recognized on the balance sheet until the bank is sold. If a sale actually happened, the proper accounting procedure would be to recognize that equity as "goodwill." It's the same as with any other corporation that has developed intangible assets. The only difference here is that the unsecured non-recourse nature of its debts allows us to put a lower bound on the value of its intangible assets (assuming the purchasers won't choose to liquidate, and I think we can agree that they won't).

I agree with Andy's characterization of money as an unsecured liability. Not sure about non-recourse. I'd add senior and perpetual to the mix, as in senior unsecured perpetual liabilities. And I like the idea of adding a convertibility feature to it, as Nick described, since central bank deposits can be converted into bonds if certain things happen. It's a floating convertibility feature though. And central bank liabilities can even pay interest too in some situations as in the Fed's IOR. So a central bank issues senior unsecured perpetual liabilities, potentially interest paying, with floating convertibility into bonds.

Nick, your last 2 posts on money-as-a-liability seem to me to be sub-chapters within your post on store of value. Having read and re-read David Laidler's excellent Taking Money Seriously I respect where these posts are coming from. My reading of that book was that (among other things) David was reacting against the literature that treated money as little more than a store of value to be held in a portfolio with other stores of value, essentially abstracting from its means of exchange role. He was also justifiably reacting against Walrasian models that could only fit money into the model as a store of value, since in such a model money has nothing to do unless it was treated as little more than an asset. Anyways, you know the details better than me.

My sense is that in order to get their point across, the push back Laidler and others mounted against money as a store of value packed more of a punch by denying the store of value nature of money altogether. Attacking the idea of money as a liability is a sub-category of this, since if central bank issues are liabilities, then they are also stores-of-value. But this seems like throwing out the baby with the bath water. It seems to me that the way to take money seriously is to figure out how to configure both its medium-of-exchange and store-of-value aspects together harmoniously.

So with regards to the latter, as an investor looking for good stores-of-value, were BoC liabilities to fall by 4% in 2011, I'd want to own them as stores-of-value due to their floating convertibility feature. If the convertibility feature were for some reason unceremoniously yanked, I'd be an investor in BoC liabilities as stores-of-value at some much lower level as they remain senior liabilities of the BoC. Wearing the hat of a consumer, not an investor, I'd be a holder of BoC money (here I would never think of them as liabilities) as a convenient medium of exchange no matter what their value, hat tip to David Laidler.

Nick, Thanks for clarifying your various points. But I’m not sure that central banks’ occasional need to “buy back notes” is a liability in the normal sense of the word liability.

To “buy back notes”, i.e. rein in monetary base, a central bank just sells bonds. No big liability there. As to where a CB gets the money to pay interest on those bonds, it can just print it. No big liability there either.

And if the required deflationary effect is not big enough, the CB governor just needs to ring up the finance minister and tell him to raise taxes and extinguish the money, or hand the money over to the CB who will extinguish it. No big liability there either.

This is all a world away from me undertaking to buy something, perhaps at a loss. That undertaking might bankrupt me, or might lead to a reduction in my net assets. That’s what I call a real liability.

Andy: OK, the car provides driving services to the new owner(s); the $20 provides monetary services to the new owner(s). What does the central bank need to do to ensure it continues to provide those monetary services? Replace old worn notes with new ones, so OK add in replacement printing costs, which I ignored. Make sure the money doesn't depreciate too quickly, by for example, targeting 2% inflation. And I *did* include that obligation. What else am I missing, which is costly for the central bank to provide? Stamp out counterfeiting? OK. We could include that cost too, as an obligation.

Ralph: "To “buy back notes”, i.e. rein in monetary base, a central bank just sells bonds. No big liability there. As to where a CB gets the money to pay interest on those bonds, it can just print it. No big liability there either."

The possible need to sell assets is a liability, that offsets those assets. The CB doesn't pay interest on the bonds it sells, the government does. But if the CB sells the bonds it owns, it no longer gets the interest.

Scott:

You say that base money trends up.

Well, if we have a quantity rule that says base money trends up no matter that, that is true.

If instead, nominal GDP is targeted, then base money will trend up only if the demand to hold base money trends up. That is not necessarily true.

What will happen in 100 years?

What happens if people start using Euro currency? Or Chinese currency? Or they use electronic payments for everything?

Well, we can outlaw that. Or impose reserve requirements. Heck, must mandate that everyone accumulate currency and put it in a shoe box.

The old monetarists treated all money like it was zero-interest hand-to-hand currency issued by government spending and subject to a quantity rule. The assumption was that there is always a budget deficit representing the issue of money. Other taxes, plus money issued according to the rule equals goverment spending. And there could be a debt financed deficit if you want.

From my perspective, the deficit funded by money issue plus the deficit funded by other debt is the total budget deficit. The national debt is the issue of government currency plus all the other government debt.

But, I grant that if you assume the quantity of currency is on a fixed by law trajectory, then the old moentarist approach makes sense.

But I think it is the wrong way to look at real money. Quantity rules are a bad idea. Money has never been issued in that form.

Think of paper money on a gold standard. If you leave gold and just say, "print money and spend it," then sure, it isn't a liability. But if you say, gold is a bad medium of account, stabilize something else, the paper money is still a liability.

If the central bank is independent, then it must keep assets to match the money. Like liabilities. And it is obligated to maintain the value of the money. And that is true of liabilities. A dollar debt when it is due must have a market value of a dollar or there is bankruptcy. A dollar debt due continually must be worth a dollar continually.

A regular bank must match liabilities with assets so that when the demand to hold its liabilities falls, it can reduce the quantity to match the demand and keep their value at par.

This is the same with the nonmonetary liabilities as well. The bank issues nonmonetary liabilities and funds assets. It has to have assets to match the liabilities so that _if_ people don't want to lend to the bank any more, it can pay them off.

Now, if the monetary authority is just a branch of the treasury, then it is funding a deficit by two types of debt. There is zero interest debt, the currency and the interest bearing debt. The national debt is the outstanding currency and the outstanding interest bearing debt. The amount of currency it issues is limited by the demand to hold it. If peopel want to lend less to governmetn in the form of currency and are still willing to lend by T-bills, then the government must shift its funding. This is exactly what it must do if no one will buy 30 year bonds but will buy T-bills.

Of course, the governmetn can reduce currency by running a surplus, but this is what the government does when it pays off debts--pays down the national debt.

Let's see...

base money is not like debt because:

base money has zero interest -- wrong.

base money should always increase -- wrong.

base money is at a historic high, and the need to greatly reduce its pressing. But it really isn't a liability? There is no need to worry about the assets?

Obviously I am biased by three things:

Indirect covertibility as a hypthetical possibility.

Index futures converibitiliy as a contractual obligation for central banks. And I don't see it as an put option. Of course, it is a futures contract and not an option.

Privatization of hand to hand currency.

Interest on reserves.

But those reforms together, and all money looks like liabilitities.

Now, if you insist on reserve requirements (rasing them?)

If you insist on keeping private banknotes illegal..

If you insist on zero interest on reserve balances..

Then the paper gold framing might work. I think all of those things are bad.

I think there should be no reserve requirements. Banks should be free to issue private
redeemable currency if they want, and the central bank should pay interest on reserves, but
a fixed amount below other short term market rates.

And the amount of base money should always adjust to the demand to hold it.

It is best understood as a liability.

Of course, I advocate index futures _convertibility_.

I also advocate privatization of currency and interest on base money deposits. And so, I favor getting rid of this stuff you want to treat like paper gold.

Clear you mind of old monetarism. It is not a useful framing.

The demand for base money doesn't necessarily grow. Or rather, you can make it grow if you create enough regulations. For example, make everyone accumulate a certain amount each year.

Nick:

Suppose GM stock sells for $60, and every GM share includes a put option on GM stock that allows the bearer to sell the stock back to GM at an exercise price that declines at 2% per year. The GM shares are still GM's liability, and so are the puts. It wouldn't make sense to say that the put is GM's liability while the share is not, and it doesn't make sense to say that a put on a paper dollar is the central bank's liability but the paper dollar itself is not.

I'm willing to acknowledge, for reasons Andy specifies, that base money is, technically, a liability of its central bank. That said, Nick originally framed the issue in terms of what to teach his students. The question they were asked was whether base money is a liability of its central bank, and they all answered yes. They're all right, technically, but the answer still seems unsatisfactory. It would be like me asking 'do you have a dog?', and you answering 'yes' because you have a chihuahua. You'd be right, of course: a chihuahua is a kind of dog. However, a chihuahua is not a typical dog, and merely answering 'yes' is likely to be misleading in a way that it wouldn't if instead you had, say, a border collie.

I don't think it would be wrong to say that base money is less of a liability to its central bank than a mortgage is to an average homeowner. That seems to be part of what troubled Nick; without qualifying exactly how base money differs from other liabilities he felt as though he was misleading his students.

Nick,

The question of whether base money is a liability of the CB is a major discussion point in an essay I just posted yesterday and today at New Economic Perspectives.

http://neweconomicperspectives.org/2012/03/the-public-money-monopoly-pt-i.html

See the section entitles "Money, Credit and Central Bank Liabilities." My answer: no. But I would be interested in getting your reaction.

Andy is right. Bill is right. Base money is a liability of a central bank.

Nick has got some valid points, but he is violating the principles of accounting. Nick should be measuring the value of central bank's equity, not the value of the central bank's liabilities.

Accounting at market values is the most relevant accounting. 20$ note is a liability whose market value is $20.
Suppose the demand for base money increases, and another $20 is printed. If market values are used for everything, including goodwill, central bank's liabilities have increased by $20, central bank's assets have increased by $20 too (maybe the central bank has bought some financial asset with that $20 bill). Central bank's equity and goodwill have increased too, say by 13$. In order to arrive at the exact calculation of $13, you need to calculate the value of those american put options.

Increased demand for base money means that the market value of central bank's equity has grown, and that the market value of central bank's goodwill has grown. Unless the central bank targets deflation which means that the expected real return on cash is way above the market rate. In this case the market value of equity declines. So there is a natural limit to deflation under a fiat money standard.

Suppose the inflation target is changed from 2% to 3%. The effect on the market value of central bank's equity is ambiguous. The effect on the size of the monetary base is ambiguous.
On one hand, the inflation-adjusted value of monetary base is now smaller, the value of central bank's equity is higher, so the larger monetary base is warranted. On the other hand, consumers might switch to the alternative suppliers of liquidity services, especially if interest on reserves is not paid. This effect reduces the market value of central bank's equity, and the smaller monetary base is warranted if the new inflation target is to be hit.

123 (TMDB), I'm more with Nick here. I see the prevailing accounting norms regarding central banks as useful bookkeeping conventions, but not true measures of economic reality.

Personally, I think that there is something odd about whole idea that the central bank has financial assets, liabilities or equity. We might as well view the CB as a kind of black hole, or black box or black vault. Money disappears from private sector bank accounts at its command and is destroyed; money appears on bank accounts at its command and is created. But whether, or how much, money exists inside of the black CB box is of absolutely no economic significance. We could adopt an alternative set of accounting conventions according to which the CB never possesses any money, and everything would work just as it does now. Or we could adopt a convention according to which the CB always possesses an infinite amount of money, so that receipts and payments never affect its balance, and things would still work out as they do now.

As the monopoly issuer of monetary units that it creates virtually cost free, the application of the same system of accounting standards to the CB that are used for external users of the currency is an inherent distortion of reality.

@Dan Kervick

Accounting for cash at face value is a very useful accounting convention. Other accounting convention are not so useful, as we really need to measure the central bank equity and goodwill at market values so the economic reality is reflected.

The issuance of cash is usually very profitable. But this does not change the market value of cash liability. The market value of central bank equity and goodwill grows instead.

Some companies were able to fund themselves very cheaply and thus profitably in the commercial paper market. But the resulting gain is reflected by increasing the value of equity, not by reducing the value of commercial paper liabilities. Another example is Warren Buffet's insurance float liability that has zero cost and is virtually permanent.

Dan: I have read your post. I disagree with some of the things you say in the first half of it, but I am in basic agreement with what you say in the second half, from your section "Money, Credit and Central Bank Liabilities" on down.

Basically, you can read me as saying "Yep, base money is not a liability of the central bank, except maybe...." and this post concentrates on that "except maybe" part.

Funny thing is, when it comes to the question of whether base money is or is not a liability of the central bank, there is no major difference between: MMT; what I was taught by my Chicago monetarist professors back in the 1970's; what Keynesians like Willem Buiter have been saying for ages.

In the olden days, we used to ask the same question in a slightly different way: "Is money net wealth?" And we would argue that (outside, central bank) money is net wealth, because it is an asset to the public, with no offsetting liability to the central bank. Pesek and Saving argued that money issued by a monopolist central bank was net wealth, while money issued by a competitive commercial bank was not net wealth. Which I still think is basically the right answer. (P&S argued that it was the monopoly vs competitive distinction that mattered, as opposed to Gurley and Shaw who argued that it was the outside vs inside distinction that mattered.)

Bill: suppose the government were very soon about to lose its de facto monopoly profits from being the issuer of base money. Then Pesek and Saving would say that outside money would no longer be net wealth, because the PV of monopoly profits would go to zero. And my analysis would say that the value of the put options would rise to equal the value of the outstanding currency, because all those put options would be exercised very soon (unless the government paid competitive interest on currency).

Mike: suppose people liked wearing GM stock as jewelry to flaunt their wealth, and that only GM stock was fashionable to wear as jewelry (so that GM had a de facto monopoly on the "stock as jewelry" market), and would be willing to buy and hold GM stock even if it paid no dividends, was non-voting stock, gave no right to any share of GM's profits, and was expected to decline in price at 2% per year.

123: suppose the (real) demand for currency increases by $20. Suppose the CB satisfies that demand by selling a $20 note for a $20 bond to keep inflation on target.

If we know for sure that this increase is permanent (and if the variance is unchanged), then the aggregate value of the put options is unchanged. The CB has an additional $20 bond in assets, and no increase in liabilities. Its equity has increased by $20.

If we expect this increase might be temporary, and the CB might have to buy back that $20 note in future, the value of the put options will also increase, but by less than $20. Its equity has increased by less than $20.

"Suppose the inflation target is changed from 2% to 3%."

Some of the existing put options are exercised immediately, as the demand for money falls. So its assets shrink. But at the same time, the value of the remaining put options shrink, since nominal demand will now be growing at 3% instead of 2% for any given time-path of real demand from now on. Plus, the value to the central bank of people exercising the call option *may* also increase (depending on the elasticity of demand for money, and where it is on the Laffer curve).

Dan: Here is a (very oversimplified) restatement of your post:

Part 1. Government bonds are not a liability of the government.

Part 2. Government currency is not a liability of the government.

Here is a (very oversimplified) statement of my response:

Anyone who has a monopoly on doing something has a potential for monopoly profits from doing that thing and has net equity from the present value of those monopoly profits, where assets exceed liabilities.

You are right on part 2, because the government has a monopoly on issuing currency.

You are wrong on part 1, because the government does not have a monopoly on issuing bonds.

Again, that's an oversimplification, of both your and my positions. But it's a good place to start thinking about these questions. (This is just Pesek and Saving's argument generalised to bonds.)

Nick: "And we would argue that (outside, central bank) money is net wealth, because it is an asset to the public, with no offsetting liability to the central bank."

Depending what you mean by that, it doesn't strike me as right. Imagine an entirely equity financed railway monopoly. Imagine that there is nothing clever, or efficient about the operations of the company. They are just a tax collector. The equity value is nothing but the NPV of monopoly rents. Is the value of the stock "real wealth?" You bet. Those rents are valuable. Does the existence of the railway as a monopoly contribute *net* wealth to the economy? Absolutely not. The rent extraction is distorting of transportation services and therefore welfare reducing. Same with outside money.

But maybe I misunderstood you. My bottom line is that inside money can be efficient. Outside money is not.

Nick, just one thing: Although I argued in my post that central bank money is not a liability of the central bank, I don't think that qualifies as the "MMT position." My understanding is that some of the MMT economists agree with that statement and some don't. I was trying to articulate a position that I thought was a best theoretical fit with the MMT view of government as the monopoly issuer of the currency, and the agent responsible for any net increase in financial assets in the non-governmental sector. But I think some of the other MMTers make sense of those views within a slightly different framework.

A lot of MMTers are big fans of Innes. I am too as far as his historical account of money and credit goes, and think his credit theory of money is correct in its general thrust. But I demur at the claim that all money is a credit interest. I jump off at the last stop and say that all money except base money is a credit instrument. The latter I see as the legally established final means of payment for all financial debts and obligations, but as not itself a further IOU.

But I demur at the claim that all money is a credit interest.

should read

But I demur at the claim that all money is a credit instrument.

Nick, I agree with the economic substance of your comment. However, it is important to use the single correct accounting language. Otherwise the differences between various schools of thought would be exaggerated.

"If we know for sure that this increase is permanent (and if the variance is unchanged), then the aggregate value of the put options is unchanged. The CB has an additional $20 bond in assets, and no increase in liabilities. Its equity has increased by $20."

Here the substance is right, the language is not. The correct description is that the CB has an additional $20 bond in assets, and 20$ increase in liabilities. Its equity has increased by $20. Its intangible asset (goodwill) has increased by $20.

"Part 1. Government bonds are not a liability of the government.
Part 2. Government currency is not a liability of the government."

Both bonds and currency are liabilities. The debate should be about the value of intangible assets on CB and treasury balance sheets and the nature thereof.

The central bank has intangible assets related to the future profits from issuance of cheap liabilities, and it also has intangible assets (or liabilities) related to future profits (or losses) from its investing activities.

Government treasuries have got various intangible assets too, but I think their nature is mostly non-monetary. The intangible assets of Greek treasury have lost their market value during the last five years, and the net worth of Greek treasury is now negative.

Even though German debt to GDP ratio is higher than the Spanish one, it is Germany that has lower financial costs. The value of German treasury intangibles has grown, partly due to better economic prospects, partly due to higher credibility, and partly due to the fact that German bonds are very liquid.

"Pesek and Saving argued that money issued by a monopolist central bank was net wealth, while money issued by a competitive commercial bank was not net wealth. "

This is wrong. Before the crisis, the market value of bank equity was way above the book value, this is the very clear evidence of intangibles. In Pesek's language, M2 was partly net wealth before the crisis. After the crisis the value of commercial bank intangibles has shrunk (but not disappeared). The value of central bank intangibles has grown.


Nick, suppose x% M2 growth rule is implemented. In this case, M2 is not a liability of the banking system, even though it is a liability of each individual bank. So it is better to use language that is additive and to discuss intangible assets instead.

"The value of that put option, and the liability of the Bank of Canada, is always less than the value of the monetary base."
One of the parameters driving the value of this option is the riskiness of the CB asset portfolio. If the asset portfolio is riskier, put option volatility is higher, and the option is more valuable. On the other hand, this effect could be counteracted by the increased expectations from the economic profits from the asset side activities, so the net effect on CB equity is not clear.

""Pesek and Saving argued that money issued by a monopolist central bank was net wealth, while money issued by a competitive commercial bank was not net wealth."

Central bank distribute the income from net wealth to treasury, and competitive central banks distribute the income from net wealth to the users of M2.

It is like the difference between the patented and generic drugs.

I agree with Bill and others that currency is a liability of CB. Otherwise even government bonds, if they are issued when interest rate is lower then long-term growth, are not liability.

I also contest your statement in response to Dan, "You are wrong on part 1, because the government does not have a monopoly on issuing bonds." They have monopoly on issuing bonds safe bonds with nominal returns, don't they?

Also your statement that "the government has a monopoly on issuing currency." can be contested. It is true that technically the government has monopoly on issuing currency, they may not really force people to use it. There is still an option that people will start using other currency if government does something nasty the very same way as people have option not to buy government bonds if they do not trust them. Also your definition is quite weird, since government bonds are "liability" that is paid by government currency. CB issues the government currency and since CB is part of the government, then government bonds cannot be a liability for government, right? Now I see why this topic evokes MMT for you, because it is exactly what MMTers are saying. And the answer is exactly what is the answer for MMTers. There really is something we may title as a "Long Term Currency Constraint". It does not matter that CB is monopoly issuer of government currency the same way as it does not matter that government is the monopoly issuer of government bonds.

I think it all comes back to your original article where you discussed if debt is the burden on next generations. The same is valid for currency. Issuing $20 bill now and not 50 years later is in this manner a liability. It has carries inherent opportunity cost since it could have been done later in the future. $20 Bill as medium of exchange may be needed for transactions, but it can simultaneously be used also as a store of value and is the function that makes it a liability.

Nick; Don't have time to read all the comments on this thread, but didn't you slip a derivative somewhere? You're comparing an inflation target to put on the CPI. Aren't you really thinking of a price-level-target policy?

And there is a limited spread of maybe 2 % pts either way, that the bank is expected to reasonably keep inflation targeting a priority, else a recession or hyper-inflation probably excuse the bank from needing to worry about inflation targetting.
I can't read ads like Maclean's at present...would there be any cost savings in building more new F-18s? Certainly save money on pilots. Many parts might be not made anymore and engineers have moved on. They are basically like F-35s without the partial-stealth. I'd like to see slow UAVs used to deploy overpressure tents for pandemic civil defense. But we'd need to build over-pressure tents (made of plastic thus oil), and invent logistics UAVs. A problem of low corporate tax rates is it retards new industries. Corporations under-invest in new industries out of some game theoric market share and profits scenario (traditional product maxes short term profits I think). Forced R+D in concert with J.Flaherty tax rates would help. Europe has also lowered corporate tax rates, their R+D level is a little higher; not sure how much R+D undoes the market forces product line false maxima.

Simon: you spotted it! Yep, I slipped a derivative. I'm really talking about price level path targeting here. It was too hard to describe it accurately for inflation targeting, so I fudged. I don't think it's a big deal.

JV: "I agree with Bill and others that currency is a liability of CB. Otherwise even government bonds, if they are issued when interest rate is lower then long-term growth, are not liability."

If the interest rate on government bonds is permanently below the growth rate of NGDP, because the economy, so the government can run a sustainable Ponzi scheme (and has some sort of monopoly position so that private agents can't run competing sustainable Ponzis, and push the interest rate up) then I would say that government bonds too are not a liability. The government never needs any future tax revenue to pay interest and principal on those bonds, it just rolls them over forever. And, in that case, debt would not be a burden on future generations.

Yep, the question of the extent to which the government has monopoly power and can earn monopoly profits on money and bonds is arguable. It depends on the demand curve facing the government in each case, and whether it slopes down or is horizontal. It's clearly true on currency, since some currency is willingly held even in hyperinflation, when it's paying very negative real interest rates. On government bonds it's less clear.

123: I'm not convinced that talking about "intangible" or "goodwill" assets of the CB is any clearer.

"Central bank distribute the income from net wealth to treasury, and competitive central banks distribute the income from net wealth to the users of M2.
It is like the difference between the patented and generic drugs."

Yep.


Nick: I think it is still a liability even the Ponzi Scheme could be run permanently. Imagine that we would find a huge reservoir of some rare nature resource that is being continuously filled by a small stream of newly created resource. We could observe that for last 200 years this income stream was growing steadily by some trend percentage.

Now there are several ways how to go about this. It could well be so that the first generation to discover this reservoir will consume the whole legacy in a spending stream and leaving only the income from the stream to their children. They could as well consume only as much of the resource from the reservoir as came into into came from the stream. They can even completely abstain from using the resource and leave it for future generations enlarged by the stream that came into the reservoir during their lifetime, just in case that their children would need it more then themselves. Or they could do something in between. The point is that whatever they do has some impact on future generation. Imagine that it would be even possible to pump out more resource now that would decrease the resource yield during next period.

I say that there is no qualitative difference between consuming just a smallest fraction or consuming everything. Every single gram of resource consumed lowers the endowment of future generations, the observed prudence and or opulence is just matter of scale, not qualitative change of behavior. I think the same applies for government debt or any similar promises of future payment (stores of value) that cash can serve as well.

Nick:

About GM stock as jewelry:

Don't forget the puts. As long as GM agrees to buy back each share for $60, or 1 oz of silver, or whatever, then shares would trade for $60. I can't tell if you think the usefulness as jewelry would make shares sell at a premium above $60. I can imaging that shares could still sell at $60 (-2%/year) in spite of the jewelry demand. If that's the case, then GM's assets would have to be sufficient to buy back every share at $60, or else there would be a run on GM as people rush to redeem.

Nick:"I'm not convinced that talking about "intangible" or "goodwill" assets of the CB is any clearer."

Your story explains a lot and put option view is a new and good analysis technique.
While there are lot of technical reasons why intangible asset explanation is better (patents, monopolies and networks are usually accounted for on the asset side; if base money is expected to grow, expected profits from the non-yet-existing base may be significant; assets are additive etc.), the main reason why after telling put option story it makes sense to switch to intangible assets is the communication reason. I see that you and Andy have the same underlying CB economic model, but you had a heated debate about the terminology. Scott accepts both approaches "But the cash itself is not a liability. Or if it is, then there is some other asset that should be on central bank balance sheets, which is roughly equivalent to the base." It is not clear if there is any model-based difference between your and Bill's opinion. And it is easier to pinpoint the source of disagreement with MMT if an empty debate "what is a liability" is avoided, and discussion is turned towards intangible assets and future profits (I remenber Krugman said to MMTers that when the annual seignorage is 2% of GDP you get hyperinflation).

I am waiting for your next post on this topic. Possible extensions are "call options on cash when CPI declines below 2%", "does the NGDP futures targeting peg the price of american put options on base money" and "is the prepayment option in the ECB LTRO 3 year auction intended to replicate this put option".

Mike,

I don't think money as stock works. Suppose that the central bank's assets had, for some reason, appreciated so that they were worth substantially more than its outstanding money issue. Would this be deflationary? I doubt it.

RebelEconomist:

Start with a bank that holds assets worth 100 oz. of silver as backing for $100, so that each dollar is worth 1 oz., and the bank maintains convertibility at that level. If the assets appreciated to 110 oz, the bank would have no reason to pay more than 1 oz per $, so the dollar would stay at 1 oz, and no deflation. But suppose that over the weekend, when the bank was closed, its assets fell in value to 90 oz, and everyone knew it. Speculators would value the dollar at .9 oz/$. If the assets later rose to 94 oz, then the dollar would rise to .94 oz/$, and you would get deflation.

But isn't that the point, Mike? You write "If the assets appreciated to 110 oz, the bank would have no reason to pay more than 1 oz per $, so the dollar would stay at 1 oz". If banknotes were like GM stock, the dollar should rise to 1.1 oz, because the banknote holders would own their share of the extra 10oz. Unless you have something other than ordinary shares in mind when you write "stock"?

Actually, it would be less confusing to stick to a central bank issuing money backed by nominal assets, as it is hard to see how silver could appreciate in oz!

RebelEconomist

I was assuming that the bank's assets were worth 100 oz of silver, not that they consisted of physical silver. Thus the assets could rise in value against silver.

It would be better to say that money is a liability of its issuer. This means it shares characteristics with stocks, bonds, options, etc. without being legally equivalent to stock. For example, if the bank has committed to maintaining convertibility at 1 oz./$, and if the bank's assets rose to 110 oz., the bank would take the extra 10 oz as a gain in its net worth, and would not normally raise the value of the dollar to 1.10 oz.

Thanks Mike, that's what I thought - you are not being dogmatic about money being more like equity or stock than anything else, unlike some MMT fans, or C.J.Cook, with whom I have discussed this with before, and never seemed to get to the bottom of why they are so sure that money should be regarded not as debt but as equity / preference shares or whatever. I would say it takes more than being perpetual to make equity, such as returns that vary with the fortunes of the enterprise and/or control rights etc.

The comments to this entry are closed.

Search this site

  • Google

    WWW
    worthwhile.typepad.com
Blog powered by Typepad