On Fri, 24 Aug 2001, Anwer Khan wrote:

>AK: Hi
> I just read your paper defending the Real Bills doctrine. Your arguments
> seem very persuasive. Do you feel any momentum within the profession
> towards these ideas? Whose counterarguments do you consider
> important to investigate?
> Anwer
> (I'm a grad student at UCLA, but I'm not at all specialized in
> monetary questions)

MS: Thanks for you email! I'm always excited when someone reads the paper out
of the blue.

My best indication of momentum within the profession is from the IDEAS
database, which tracks the number of downloads. I had filed a copy of the
paper in 1994 as a working paper at UCLA, and lately it has been getting
something like 20-30 abstract views per month and 3-4 downloads. People
who have been sympathetic to the paper are Jack Hirshleifer and Charles
Calomiris. Everyone else that I know of, most notably Arnold Harberger,
has been unconvinced, their primary reason being that if a currency like
the dollar is not convertible, then it can't really be considered backed,
and so they reject my contention that fiat money doesn't exist.

I had a fairly long email exchange with David Laidler, who also remains
unconvinced for the same reason. I have copies of those emails if you'd
like to see them.

Also, there was an exchange on the web discussion group PKT
(post-keynesian thought) a year or two ago where I batted the idea around
with Warren Mosler. I might call him uncommitted to either side. Anyway,
those exchanges are still there on PKT, and easily found with google by
typing key words like fiat money, backed money, real bills doctrine, etc.

Oh yes, a few years back I got some passing interest from Daniel Vaz, a
former UCLA grad who at the time was head of research for the central bank
of Uraguay. He lost interest rather quickly.

Anyway, you can check out all my writings on the subject at

There's "Backed Money, Fiat Money, and the Real Bills Doctrine",  "Three
False Critiques of the Real Bills Doctrine", "The Quantity Theory vs the
Backing Theory in Colonial America", and "A Quick History of Paper Money".
None of them have baan published, and I have a pile of rejection letters
from journals--maybe 20-25 rejections.

If I'm right about backed money, (and I think I am) it will be a long time
before the profession realizes it.

Best regards,
Mike Sproul

AK: Mike:
> I got into this subject by wondering how demand for currency can be
> created without convertibility. Originally, I had a path-dependence thesis
> that there must have been a historical period where currency was backed
> and then legal tender laws could later substitute for convertibility. But you
> seem to claim that if backing is eliminated then the value will also be lost.
> What do the people at the Feds say? Why do they bother to hold assets?
> What will they do when the debt is paid off?

MS: As far as I know the Feds have never seen my paper, but the things I've seen on
the RBD are strictly on the critical side. They'll say they hold assets to impart
confidence, or as an archaic relic of the gold standard.
I'm not sure what "debt" you mean. When the US bonds held by the Fed reach
maturity, the Fed just trades them for new ones. If all US bonds were paid off
the Fed could just issue paper dollars for private bonds, foreign bonds, gold,
silver, land, lottery tickets, etc.

> >MS:  As for money substitutes: They are just derivative moneys, and won't
> > affect the value of base money for the same reason that the issue of
> > derivative stock doesn't affect the value of the base stock.
> AK: Then shouldn't it be fairly straightforward to judge between the two
> theories empirically?

MS: Except that the quantity theory is very slippery. Almost anything can be
explained as a result of a change in money demand or velocity. For example, when
credit cards were invented, the quantity theory would have predicted inflation,
and the RBD would have predicted none. One could gather data on base money,
credit card money, and the price level, but I don't think the data would make
quantity theorists abandon the QT in favor of the RBD. Thomas Cunningham examined
some data from Taiwan and Bruce Smith looked at data from the colonial period,
but neither of them were able to convince the skeptics.


Anwer Khan wrote:

> > MS: About bankruptcy and dividends: That's the same problem everyone has, and
> > I'm apparently not bright enough to make a persuasive counter-argument.
> AK: What if we say that a Central Bank has the _option_ of establishing
> convertibility at any point in the future?
> 1. At what ratios could it establish convertibility, without facing a run?

MS: If it has 100 oz of gold as backing for $100 paper bills, then a ratio of $1/oz
won't cause a run, but $0.99 will. Is that what you meant?

> AK: 2. How do we value this option? What is the optimal exercise?

MS: I usually think of the holders of paper dollars as having the option. A paper
dollar is a european-style call option with a zero strike price. The dollar
allows its holder to claim an uncertain payment at an uncertain time. Thus it's
just like most kinds of stock, and the value is determined just like stock.


          Sat, 25 Aug 2001 08:09:20 -0700

Anwer Khan wrote:

> > MS: If it has 100 oz of gold as backing for $100 paper bills, then a ratio of $1/oz
> > won't cause a run, but $0.99 will. Is that what you meant?
> AK: Yes. It seems to put an (upper?) bound on the value of the dollar. This in
> itself seems to give some value to the theory. It seems to be in line with the
> fact that countries that peg their exchange rate too high face speculative
> attacks ... is this an advantage over the Quantity Theory?
> If the assets backing the currency appreciate in value, then the currency also
> appreciates. I suppose that Quantity Theorists would explain this as an
> increase in the derived demand for the currency, no?

MS: Yes on both counts. For example, if the Brazilian Central Bank holds a lot of
Brazilian govt bonds, and if Brazil can't pay its bills, then the bonds depreciate.
Since the bonds backed the currency (the "real"), the real depreciates. If the bank
tries to support the real--that is, to maintain convertibility at too high a level,
then the bank faces a run and they eventually have to give up and devalue, or else
the bank loses all of its wealth. Seems so obvious that a loss of backing was the
problem with all those currency devaluations a few years ago, but economists were
content to explain it as a fall in money demand owing to a fall in confidence. I
remember a newspaper article by Rudiger Dornbusch saying exactly this.

> > >AK:  2. How do we value this option? What is the optimal exercise?
> >
> > MS: I usually think of the holders of paper dollars as having the option.
> AK: And the issue of these options doesn't change the value of the assets
> the bank is holding.

MS: Yes; as long as each option is issued on sufficient security. That's maybe a
confusing way of stating it though. I just think of a bank issuing new dollar bills
and taking in stuff worth an ounce of gold each time. This way, backing always moves
in step with the amount of money and the value of the money stays the same.

> > MS: A paper dollar is a european-style call option with a zero strike price. The dollar
> > allows its holder to claim an uncertain payment at an uncertain time. Thus it's
> > just like most kinds of stock, and the value is determined just like stock.
> AK: What I'm thinking is that if the value of the dollar drops to $1.50/ounce, then
> the bank would profit by buying back dollars, and could be expected to do
> this. In the extreme, it started with zero net worth and ends up with all dollars
> bought back and a net surplus of gold.

MS: That's certainly true for a private bank. Buying back the money would cause a
recession though, and a central bank would get a lot of heat, assuming people knew
what was going on. They didn't know in the great depression though.

> AK: Oh, but it would be disadvantageous to sell dollars at this price,

MS: You mean disadvantageuous for the people who have dollars, right?

> AK: so perhaps convertibility would not be the optimal strategy, or perhaps there
> could never be rational expectations that the bank will establish convertibility
> at $1.50/ounce - anything more than $1/ounce. I think that you describe this
> as issuance in exchange for sufficient security.

MS: Yes; certainly for private banks. Maybe less true of politically-motivated central

> AK: What I'm thinking is that it doesn't matter that there presently is not
> convertibility, as long as the bank is able to buy and sell at some point in
> the future, that possibility should be enough to support an equilibrium.

My thoughts exactly.

>AK:  Do you think that presenting this problem as a game would be more
> persuasive to anyone?

MS: Well, I think convincing people would require significant empirical work, an
intuitively appealing explanation (I think my paper might qualify) and a mathematical
explanation. So far I've done one of those three things, and yes, I think game theory
would convince game theorists.

>AK:  But I don't understand why financial economists will use no-arbitrage type
> thinking to price other assets but not this one.

MS: You and me both. People seem satisfied to think that paper bills have value for the
same reason that rare stamps have value--just the fact that they are limited in
supply and desired by people. It's very hard to make that leap from the fiat money
world we learn in school to the concept of backed money. I only thought of it in
1989. I happened to be pondering the fact that call options on GM stock don't reduce
the value of GM stock, when I realized that a checking account is actually a call
option on green dollars, and shouldn't affect the value of green dollars for the same
reason. I though about that problem for 2 years before I had it figured out, and
another 2 years before I could write it down sensibly.


          Sat, 25 Aug 2001 08:32:16 -0700
  Anwer Khan wrote:

> > > >MS:  About bankruptcy and dividends: That's the same problem everyone has, and
> > > > I'm apparently not bright enough to make a persuasive counter-argument.
> AK: Perhaps it's enough to say that sometimes it will be optimal for the bank to buy back
> the claims on its assets, just as firms do, and this is enough to determine a lower
> bound for the value of the money. The bank certainly values the money ... if we all
> give it back then the bank gets to keep its assets, otherwise not.

MS: Yes; though a politically-motivated central bank might not behave in this rational way.
One realistic scenario might be that a new kind of credit card drastically reduces out
demand for green cash, so people return it to the private banks. As cash piles up at the
private banks, the Fed finds that open market purchases of bonds are met with a chilly
reception--nobody wants the green cash. As the bonds held by the Fed mature, the green
cash returns to the Fed and is not re-issued since there is no public demand for it.
Eventually all the bonds are gone and only the Fed's gold is left. If the public still has
no desire for green cash, the Fed can buy back the cash with gold. Convertibility is
finally restored without the Fed having to really embark on any brave new policies.

> AK: If I have a lien on my car, that prevents me from selling it, then I should be willing
> to pay up to a maximum amount to have that lien discharged. This should be a
> lower bound on the value of such a claim. The fact that I value it is enough to
> support some value for it as a traded instrument.

MS: A debating opponent could always claim that the dollar is different--that convertibility
might never really be restored, and that backing without actual convertibility is a
nirvana-like, meaningless concept. Not that I believe that, but I haven't been able to
really defeat that argument so as to convince the skeptics.

> AK: Now people have treated Central Banks differently than commercial banks,
> since they don't face the legal restrictions that commercial banks do.

MS: Legal restrictions only apply to ordinary bank accounts--not to credit cards, eurodollars,
foreign currencies, etc. For this reason I don't think they have much real effect other
than to make it hard for banks to do business.

> AK: Still, they must value their money for other reasons if they are willing to conduct open
> market operations, but only under certain circumstances. The value that they
> place on their money should be able to determine or at least constrain an
> equilibrium value for it.

MS: You might be onto something. Opponents could always claim central banks don't operate
rationally. I don't quite see how this argument would be developed, or if it might be a
dead end.

> AK: It makes a lot of sense to me. I really don't understand why the reception has
> been so chilly.

MS: Some day I'll gather together my rejection letters and send them. Like I said:
"stubbornness and laziness"

>AK: Have you talked to Paul Krugman about it? He seems to be
> quite willing to take contrarian positions, and this would really be a scoop.

MS: I don't know him. I'd be glad to send him a copy, or maybe if you know him you could send
it from a source that he's more likely to respect.

>AK:  The only thing that concerns me is the historical associations that people
> might have with the Real Bills doctrine that might prevent them from judging
> this model objectively.

MS: David Laidler suggested I should call it the backing theory, to avoid unpleasaant
historical associations. I think that's a good idea.


          Sat, 25 Aug 2001 10:50:53 -0700
     Anwer Khan wrote:

> >MS: I've attached my correspondence with David Laidler. Sorry about the
> > chaotic look of the text, but it covers several important points.
> AK: I don't know a lot about monetary theory but David Laidler's responses
> seem utterly inadequate. To me, your arguments make perfect sense and
> are coherent. I may have the advantage that I've studied very little
> Monetary Economics so that makes me relatively unbiased.

MS: Amen!

> AK: This reminds me of the attempts to price derivative securities before Black and
>Scholes, with attempts to specifyemand etc. very messy, when ultimately the question was solved with
> no-arbitrage arguments. Somehow the profession did not take this
> reasoning to this derivative security. You give some arguments that are
> "no free lunch" in spirit, I don't see why they can't offer a better
> response than they have. Why is the availability of rival monies not
> taken as a no-arbitrage argument?
> How did Sargent and Wallace respond? Laidler says that they have
> their own spin on RBD.

MS: I sent them both (S&W) copies of the paper and got no reply. Same with about 20-30
of the most prominent economists I could think of. The only one to respond
favorably was Charles Calomiris. Laidler was probably the second-most
sympathetic--really the only prominent guy who gave me the time of day. I have
a nasty letter from Paul Samuelson that I thought I might frame someday.

>AK: It may be a political gaffe when you say sometimes that supply
> and demand don't matter - I think that you should keep
> emphasizing that Quantity Theorists have mis-specified these
> functions, yet take themselves as the defenders of supply and
> demand analysis. You do explain this, but it comes across as a
> form of damage control.

MS: I guess political gaffes are sort of what I'm all about.

> AK: The most compelling point to me is: why would people bother to
> pay seignorage when they can just use a rival money to satisfy their
> transactions demand? They could even use the subway tokens
> that Laidler mentions. The issuers of tokens aren't collecting
> seignorage in any way that's obvious to me.
> I agree that this will be extremely difficult to sell to the profession. I
> can only encourage you in what must be a lonely struggle ... maybe
> there is a Nobel in it for you some day. But it has definitely
> changed my perspective on the issue. If I ever become a journal
> editor ...

Cool; but at my present rate of progress I'll probably be in a nursing home
when it happens. Mendel, the father of genetics, had been dead 35 years when
he was discovered.

> AK: I'm glad I got a chance to look at this model before seriously
> trying to work through any monetary literature. This should give
> me a critical perspective. Your paper has been a valuable
> contribution, to me at least.

MS: Thanks; and your moral support has lifted my spirits quite a bit.

          Sat, 25 Aug 2001 11:13:18 -0700

Anwer Khan wrote:

> AK: What is wrong with the following story:
> Let's not look at the supply and demand for derivative monies and any type of monetary
> aggregate. Let's look at the supply and demand for high-powered money, the liability of
> the Central Bank.

MS: I'd start by saying that supply and demand is only an appropriate model for actual goods--not for
pieces of paper that are claims to goods (e.g. stock and money). For example, let's say that GM
has issued 1 million shares of stock and that GM's only asset is $60 mil in cash. Each share will
be worth $60. If they sold for $61, GM would sell infinite amounts and the public demand for them
would be zero; if they sold for $59, the public would want infinite amounts and GM would be
willing to supply nothing (or would buy back any shares outstanding, as you point out). Since
supply and demand are both horizontal at $60, I think it's safe to say that S&D is not what
determines value--backing is.
Now just switch to thinking about a bank that issues paper dollars backed by ounces of gold and I
think you'll have to conclude that the supply AND DEMAND for dollars is horizontal at 1 oz/$. You
already pointed out that S is horizontal. Were you also considering D as horizontal?

>AK: It has an inflation target, and is willing to conduct open market operations to defend it.
> So the supply curve is horizontal, reflecting this commitment to fix the value of the dollar.
> It will supply any amount from zero to as much as it can possibly produce.
> Now if it is not fully funded, there is no way that it can maintain a horizontal supply
> curve that reaches all the way to the axis (i.e. it cannot supply small amounts at that
> price, only large amounts). It will run out of assets if the demand drops too low. If it
> tries to support a value that it cannot, there will be a run on the currency (as has
> happened). So it can only support a price level where it can extend the supply
> curve all the way back to the axis. This determines the supply curve.

MS: This only applies under convertibility. If the currency is inconvertible the bank doesn't have to
defend anything. RBD opponents have to cave in on this one point at least: that when a currency
is convertible its value must be determined by backing. That's not the point they bring up in
arguing against me, although they will carelessly talk about convertible currencies as if the QT
applies and not the RBD. The best argument against the RBD concerns inconvertible currencies.

> AK: The demand curve could have any shape, it doesn't matter. The value of the dollar is
> determined by the supply curve, and the quantity demanded is determined by the
> intersection of supply and demand. The demand curve is shifting inwards with
> monetary innovations, but it doesn't matter, except if it shifts to a point where the
> Central Bank runs out of resources to support the dollar and must devalue. Of
> course we anticipate that this will happen, but perhaps that doesn't matter because
> a run could happen anytime it becomes obvious that the Central Bank won't be able
> to defend its inflation target.

MS: Again, only under convertibility.

> AK: So full asset-backing seems necessary to for the Central Bank to maintain the
> price level. Convertibility doesn't matter, the commitment to maintain the value of
> the dollar is what matters. And this requires that it be fully-backed with assets.
> Though I don't see how to eliminate the possibility of the Central Bank being
> over-capitalized.

MS: Not sure I follow this. Full asset backing clearly matters with convertibility, and even if
quantity theorists are careless about this point, they would ultimately admit that convertible
currencies must be valued according to backing. It's the case of an inconvertible currency
without backing, that still has value, that they have in mind.

By "overcapitalized" you mean that the bank has 101 oz backing $100 paper? Then the bank owners
are wealthier by 1 oz.

> AK: Is this a fair presentation of the story?

          Sat, 25 Aug 2001 18:00:43 -0700
     Anwer Khan wrote:

> > MS: I'd start by saying that supply and demand is only an appropriate model
> > for actual goods--not for
> > pieces of paper that are claims to goods (e.g. stock and money).
> AK: So the basic point is that Quantity Theorists are working like people
> trying to price derivatives using supply and demand before the arbitrage
> arguments of Black and Scholes.

MS: Yes.

> > Were you also considering D as horizontal?
>AK:  I was thinking that D might have a slight slope considering that rival
> monies are not perfect substitutes for cash as only cash is accepted in
> certain settings and there are transactions costs in converting rival
> monies into cash, plus government-issued money solves a coordination
> problem that presents itself if we try to avoid paying seignorage
> completely.

MS: OK. I can accept that.

> > If the currency is inconvertible the bank doesn't have to defend anything.
> AK: But it does have an interest in meeting its inflation target. As payments
> technology advances and our demand for base money decreases, what
> happens to the price level if they don't agree to buy it back?

MS: Good point. On the one hand the value of the dollar should fall because nobody
wants them. On the other hand the dollar shouldn't fall below its backing.
Whenever I run into a problem like this, I fall back on asking the same question
about GM stock. What if Merrill Lynch issued derivative shares of GM stock that
people preferred to actual GM stock, so nobody wants real GM stock anymore? What
if GM didn't buy it back? Well, as a partial answer, I'd say that as long as GM
has good use for investment capital, it will be able to issue stock regardless
of derivative shares, and the value of GM stock will be determined by its
backing no matter what. It seems that there still must be some force similar to
demand at work, since you and I are both comfortable with the idea that a
reduction of demand for GM must cause a slight reduction in the price of GM. But
if this is a demand curve, it certainly isn't a conventional one.


          Sun, 26 Aug 2001 08:12:10 -0700

Anwer Khan wrote:

> > >AK: But it does have an interest in meeting its inflation target. As payments
> > > technology advances and our demand for base money decreases, what
> > > happens to the price level if they don't agree to buy it back?
> >
> > MS: Good point. On the one hand the value of the dollar should fall because nobody
> > wants them. On the other hand the dollar shouldn't fall below its backing.
> > Whenever I run into a problem like this, I fall back on asking the same question
> > about GM stock. What if Merrill Lynch issued derivative shares of GM stock that
> > people preferred to actual GM stock, so nobody wants real GM stock anymore?
> AK: People _should_ want them if they fall below their backing because they could
> break up the company and sell its assets at a profit. Another arbitrage-type
> argument- there will always be buyers for GM stock out there somewhere, as long
> as GM has any assets, even if these are only receivables (think future taxes).

MS: Yes. A related point is that there is some equilibrium quantity of GM stock that
people want to hold, and likewise some equilibrium quantity of paper dollars that
people want to hold. Quantity theorists think that this desire to hold cash is what
gives the dollar its value, while the backing theory holds that value was determined
by backing, and the desire for cash only sets the quantity. I think a clear
explanation of this point might help to break the resistance of quantity theorists.

> AK: Of course, they can't do this with the Central Bank, but I think that earlier you
> were making a point that as our demand for base money decreases, our
> collective position in base money naturally unwinds because the government
> bonds that the Fed holds as backing will mature and that money will have to
> be retired without a bold new policy from the Fed.
> I'm starting to feel that what is holding up the dollar is the determination of
> the Fed to avoid inflation now or in the future. It is able to hold up the value
> of the currency, through open market operations if necessary. It doesn't
> need to constantly be making a market for the currency by establishing
> convertibility, other people will do that as long as they know that the Fed
> won't let prices shoot to infinity.

MS: Well put. I hadn't thought of phrasing it that way.

>AK: So we will ultimately be able to completely unwind our position in base
> money, and a day will come when we only hold assets that pay interest.
> The Fed's program for allowing us to unwind our position is fully-funded
> with the funds needed to allow us to close out without inflationary pressure
> presently stored in a "lock-box" labeled "collateral for base
> money". If the government raided the lock-box and spent those assets,
> then the unwinding operation would later need to be financed by taxes
> and we know that the political pressure on the Fed to avoid inflation
> is strong enough that we know that it will do this.

MS: A lot like the American colonial currency, which inflated when people became convinced
that the government could never collect enough taxes to redeem the currency it had
issued. One way of putting this is that the Fed just adds a step to what the colonies
did. The colonies backed their money directly with taxes. The Fed backs its money with
bonds, which are in turn backed by taxes.
Also: As far as empirical testing, people tried to test the QT vs. BT using colonial
currency, but the effort was doomed by the fact that "future taxes collectible" is
unmeasurable. The bonds owned by the Fed are at least measurable (in principle anyway,
but if you look at the Fed's balance sheet you'll see that it's impossible to get a
realistic value for those bonds. The data they collect is obviously collected by and
for quantity theorists--and any notion of backing is the furthest thing from their

One other random thought: I recently found a paper by James Madison (the founding
father) on a federal reserve website, where he wrote very convincingly that the
American paper money of his time was backed, and had depreciated from a loss of
backing. Remarkable that the paper was on a Fed website.

> AK: Again, if the Fed was relying on future taxes as backing, those future
> taxes cannot be used as security for anything else, and our overall
> debt-carrying capacity is reduced by that much. If we try to borrow
> more, bond traders will realize the incoherence of our policies and
> there would be a run on something, one of the policies would have
> to be repudiated, whether it was the commitment to price-stability or
> something else.
> It's like the debate over fully-funded Social Security vs that which is
> backed by future taxes and entails an implicit debt. If the assets are
> not put aside now, it is an implicit debt and thus an intergenerational
> transfer, not seignorage or net wealth any more than any other type
> of government debt is anything more than an intergenerational
> transfer.
> How does that sound?

MS: Like you've learned as much in 1 week as I learned in 5 years.

> > What if GM didn't buy it back?
> AK: Somebody will buy it because GM can't run away with its assets, there is a
> way of recovering them.
> > MS: But if this is a demand curve, it certainly isn't a conventional one.
>AK:  I think Andrei Shleifer wrote an article asking whether or not the
> demand curve for stocks sloped upward. I haven't read it so I don't
> know whether or not it has anything to do with this.
> But this is a nice theory - very clean, no epicycles, etc. Though I
> think that the problem of observational equivalence will be
> particularly severe in the case of the US because the asset-backing
> has already been set aside in a lock-box, and those assets are
> very stable in value. A real empirical test would seem to require a
> currency that is backed by unstable assets. You've pointed out
> examples where there was convertibility, I don't know if there are
> any examples without convertibility.

MS: The Asian and South American currency crises of a few years back would be a good
example. Very, VERY hard to find the right data.

> AK: BTW: why do we treat countries that attempt convertibility any
> differently than private banks in our theories? I remember that
> Paul Krugman described the Asian crisis as reminiscent of a
> bank run, and was really surprised that such a thing could
> happen in this age, despite all of the economic knowledge we
> have accumulated. But to a RBD adherent, such a run is not
> surprising at all.

MS: That's the carelessness of quantity theorists. The QT is based on the idea of fiat
money--money that is truly inconvertible and unbacked. When some country makes its
currency convertible (in international markets at least--often not to their own
people) they will still act as if the QT applies. When I've argued with them about how
the backing theory must be correct for a convertible currency, the thoughtful ones
will admit the point. Then they say "Well, that's nothing new" and then go back to
saying that the QT must be correct for inconvertible currencies. I suppose a few weeks
later they slip into the old habit of applying the QT to convertible currencies. Lloyd
Mints (Milton Friedman's Prof.), I think, explicitly claimed that the QT does apply
even to convertible currencies.


Anwer Khan wrote:

> > > AK: But it does have an interest in meeting its inflation target. As payments
> > > technology advances and our demand for base money decreases, what
> > > happens to the price level if they don't agree to buy it back?
> >
> > MS: Good point. On the one hand the value of the dollar should fall because nobody
> > wants them. On the other hand the dollar shouldn't fall below its backing.
> AK: If it will ever be optimal for the Central Bank to abandon its inflation target,

MS: Hard to see why that would happen.

> AK: we should be able to anticipate this (rational expectations) and price that into
> the present value of the dollar. It should only be worth as much as the assets
> that we believe are genuinely committed to buying back or retiring unwanted
> dollars. This might be less than, equal to, or more than the amount of money
> presently set aside as collateral e.g. it could also include that amount of
> future taxes that we believe are budgeted for this purpose.

MS: True--and one more reason that empirical testing is difficult. The bonds the central
bank holds might be a poor indicator of the wealth actually backing dollars.

> AK: In the case of a foreign currency, it would only be willing to take loans from
> the IMF to prop up its currency only up to the extent of the total amount of
> resources that it has budgeted as part of the asset backing,

MS: Unless, of course, the money is really a gift from the IMF.

>AK:  in this case part of the asset backing is in the reserves presently held, and part of it is in future
> taxes. What seems to matter is the total amount of assets that we believe that
> the Central Bank has committed to backing the currency. The present value
> should equal our valuation of the total stock of base money.

One random thought here: You know the "missing money" puzzle? The one that says the
Fed has issued enough paper to give every person in the U.S. $1500? Seems to me that
most of that money was just lost--in fires, trash cans, etc. If the average person has
lost just $10/year since 1913, that accounts for $880 of the $1500 in question. (per person anyway)
The point is that it's likely that the Fed's assets are much greater than necessary to
redeem the paper dollars that are still being held somewhere in the world.

Part (3)

Anwer Khan wrote:

> AK: I still don't understand why the Central Bank is considered irrational.

MS: Maybe "not profit-maximizing" would be a better way of putting it

> AK: We know its objective, we know its budget constrainst (the assets
> that it holds as collateral, which it can sell to hold up the value of the
> dollar). We can thus forecast its actions. That should be enough to
> support a rational expectations equilibrium in which money is valued.
> It is valued as much as the Central Bank values it, and this is
> reflected in the price that it will sell its assets for.

MS: I don't think people would argue when you put it this way.

> AK: It's like treasury bills. We could use them as money for large
> transactions. Suppose the government defaults on them. Would they
> still be worth anything? Would they then become an unbacked fiat
> currency? The only reason we hold treasury bills is that we have
> rational expectations that they will be honored. Does anyone say
> that US treasuries are "unbacked"?

> Have you sent a copy to David Colander? He doesn't mind being
> heterodox.

I don't know him. Do you know him or Krugman? If so I'd ask you the favor of sending
it from a source he recognizes. If not I'll take your suggestion and send him a copy.
I'd be sending the copies I keep on www.pricetheory.org

Backed Money, Fiat Money, and the Real Bills Doctrine
Three False Critiques of the Rela bills Doctrine

> http://community.middlebury.edu/~colander/

Anwer Khan wrote:
> AK: Yesterday I received three of the emails that you had sent last Wednesday.
> > MS: I remember I wrote to James Tobin thru the
> > Cowles foundation once. His reply to my paper was something like " We're
> > so far apart on money that it would take a long time, which I don't
> > have, to straighten out our differences". Anyway I know what you mean
> > about reading standard monetary theory. So messy that it's painful to
> > read.,
> AK: Does this have something to do with the importance of money in Keynesian
> theories of e.g. economic fluctuations? If I'm reading RBD correctly, since
> money is a derivative, it should not be identified as a primary cause of any
> fluctuations.
MS: In my "A Quick History of Paper Money" (www.pricetheory.org)
I discuss how money shortages cause recessions. The easiest
episode to understand is the French playing card money
story. One story I left out involved a run on the Bank of
England in 1800-something. The country was in a severe
recession, probably because of a money shortage, which was
probably caused by the bank. A clerk found a box of old
paper pounds in the back of the vault and began handing them
out to all the depositors who had been trying to get gold
for their deposits. The paper pounds relieved the money
shortage and the recession ended--that very day I think.
There are several similar historical episodes that I have
seen from time to time.

A money shortage is inconsistent with the QT, since any
"shortage" will just drive up the price of the remaining
money--hence the Keynesian rebellion against the QT. Money
shortages are no problem for the RBD. If the bank fails to
issue money to those who offer good security, then there
won't be enough money to conduct business. A recession
follows, but there's no deflation because there has been no
rise in backing. It's only been in the last year or two that
I've started to appreciate this myself, but I sketched it
out in "A quick History..."
>AK:  I'm reading some of Bruce Smith's articles on Colonial money. He seems to
> have much in common with you. Are there any key differences?

MS: The only difference, and it's a big one, is that he doesn't
make the jump to denying that fiat money exists. Also he's
not quite clear on exactly how paper money is "backed by
taxes". I've written him a couple of times and got no reply. I also remember
him saying that empirical testing of colonial currencies is simplified by the absence
of derivative moneys in that period. In my view this is no advantage, since derivative moneys (checking accounts & credit cards) don't affect the value of base money anyway.
> AK: He says that he is building on Sargent and Wallace, not the Quantity Theory,
> in his analysis of these episodes in monetary history.
MS: Yes; S&W also fail to make the leap to no fiat money.