Gold Standard Technical Operating
Discussions 2: More Variations
January 15, 2012
Last week, we were talking about some of the technical issues of
operating a gold standard system. There are actually quite a few
different ways you can do things.
January
8,
2012: Some Gold Standand Technical Operating Discussions
We looked at a few options: one was a standard currency board with
another currency. The second was a "100% reserve" type gold standard
system. The third was a gold standard system with a partial bullion
reserve, and using domestic high-quality bonds (government bonds) as
a reserve asset.
Our fourth example will be the operation of a gold standard system
without redeemability. No gold bullion is held, only government
bonds. (Actually, you can hold gold bullion as a reserve asset even
without redeemability.)
All of these systems have a target -- to maintain the value of their
currency at a certain parity with something else, namely another
currency or gold bullion -- and each has a mechanism of operation
based on increasing and decreasing the supply of base money. There
is no interest rate policy. The question is: how, when, and how much
do you increase or decrease the supply of base money? In both the
currency board and gold redeemability systems, the initiator of
action is some private market participant coming to the monetary
authority wishing to trade the currency for another currency or gold
at the specified parity price. Thus, both the timing and the size of
the transaction, and corresponding adjustment in the monetary base,
comes from the private market participant.
However, if we don't have a redeemability element (either the target
currency, in the case of a currency board, or gold bullion, in the
case of a gold standard system), then we must have some other means
of determining the timing and size of such adjustments.
For this, we can use the free market trading ratio of the currency
and the target, for example, the currency and gold bullion, aka the
"price of gold." Let's say we have a policy of maintaining the value
of a currency at $1000 per ounce of gold. In other words, we want
the currency to maintain a constant value of 1/1000th of an ounce of
gold. Let's say that, for whatever reason, there are more people
that want to sell the currency and buy gold (at the parity price of
$1000/oz.) than there are people who want to sell gold and buy the
currency. The value of the currency would naturally sag a bit, until
the market cleared. Thus, instead of being worth exactly 1/1000th of
an oz., it might have a market value of 1/1005th of an ounce. The
"price of gold" would be $1005.
At this point, the currency manager would act to reduce the supply
of base money in existence. They would sell assets, presumably bonds
(although it could be any asset, including gold bullion), receive
money in return, and extinguish the money. The total base money in
existence would shrink.
Exactly when would the currency manager act? At $1005/oz.? At $1010
oz? $1020? You could formalize this. You could say that the currency
manager would not act until the value of the currency deviated from
its target by at least 1%. In other words, either $1010 (reduce base
money) or $990 (increase base money).
Now we have defined some operating procedures. We have a HOW (sales
and purchases of bonds), a WHEN (at +-1% from the parity target),
but we still need a HOW MUCH.
In what quantity should the currency manager buy or sell assets
(bonds), thus altering the monetary base by an equivalent amount?
You could make some rules for this too, I suppose. Maybe you can say
that, for every day that the currency deviates by its target by 1%
or more in value, the supply of base money is adjusted by 1% in
quantity. Or, you could set up a ladder of increasing activity. For
example, if the currency drops 0.5% from parity, reduce the monetary
base by 1%. If this is not sufficient and the currency continues to
lose value, then at 1.0% from parity, reduce it by an additional 2%.
At 1.5%, reduce it by an additional 3%, and at 2.0%, reduce it by an
additional 4%, for a cumulative 10%, which is actually quite a lot.
So, you could make a lot of defined rules of this sort.
Historically, such rules were not laid out but rather left to the
discretion of the currency manager (central bank or commercial
bank), who probably has a feel for what would be appropriate.
The problem with offering discretion to the currency managers is
that they might be incompetent, as most central bankers are today.
They may not know what is appropriate action, and in fact they might
do exactly the wrong thing! Or, they might have some sort of
nefarious plan, doing the wrong thing on purpose. That happens too,
you know.
Although gold standard systems in the past have almost always had a
redeemability element, people who understood these things also knew
that this redeemability element was not strictly necessary.
“It is on this principle that paper
money circulates: the whole charge for paper money may be
considered seignorage. Though it has no intrinsic value, yet, by
limiting its quantity, its value in exchange is as great as an
equal denomination of [gold] coin, or of bullion in that coin …
It will be seen that it is not necessary that paper money should
be payable in specie to secure its value; it is only necessary
that its quantity should be regulated [adjusted] according to the
value of the metal which is declared to be the standard.”
David Ricardo, Principles of
Political Economy and Taxation, 1817.
"If, therefore, the issue of inconvertible paper were subjected to
strict rules, one rule being that whenever bullion rose above the
Mint price [gold parity], the issues should be contracted until
the market price of bullion and the Mint price were again in
accordance, such a currency would not be subject to any of the
evils usually deemed inherent in an inconvertible paper.”
John Stuart Mill, Principles of
Political Economy, 1848.
It would look something like this:

Maybe someone would say: what if people keep selling? What if the
currency keeps falling in value, even though the Gold Standard
Authority (GSA) is reducing the supply of money by selling assets?
This would indicate that the demand for money is shrinking even
faster than the monetary authority is reducing the supply. Nothing
wrong with that. Just keep selling assets, more aggressively if
necessary, and reducing the monetary base. Eventually, the monetary
base would go to zero, and the currency would no longer exist! It
would be impossible to sell more, because there's nothing left in
existence to sell.
"But," people then say, "what about bank deposits, bonds and so
forth? The total quantity of those things vastly exceeds your
reserves." As I've explained extensively, these things are not money. They are credit
contracts, between two counterparties, a borrower and a lender.
December
1, 2011: What is Money?
A credit contract is a commitment by the borrower (not the GSA!) to
deliver money (base money) at certain times and upon certain
conditions. Thus, if the borrower has to deliver money, to avoid
default, then obviously the borrower has a demand for money. The borrower
must acquire that money in some way. If they don't, for whatever
reason, that is a credit default. It has nothing to do with the
currency itself, and is not the responsibility of the currency
manager.
This is a rather important point. Do you see why it I insisted that
you understand what money actually is? These are not just arcane and
meaningless discussions.
Now we come to a fifth example, which is something of a hybrid
between the third example (gold redeemability with reserve assets
including both gold bullion and bonds) and the fourth example
(management of the monetary base via bond transactions, without gold
redeemability).
In this example, for HOW we have two things: 1) gold redeemability,
initiated by a private market participant; and 2) asset purchases
and sales, initiated by the GSA in reaction to the market value of
the currency compared to its parity value.
For WHEN, we have two things: 1) redeemability initiated by a
private market participant; and 2) asset purchases and sales
initiated by the GSA, either by discretion (in reaction to market
values compared to parity) or determined by some formalized system
such as the ones described earlier.
For HOW MUCH, we have two things: 1) redeemability in a quantity
determined by a private market participant; and 2) asset purchases
and sales initiated by the GSA (in reaction to market values
compared to parity) in a quantity determined by GSA discretion or as
determined by some formalized system such as the ones described
earlier.
This hybrid system is, actually, the system most commonly in use in
Britain and the U.S. during the 19th Century. It could be used on a
monopoly basis, in the case of the Bank of England, or by many
private commercial banks each issuing their own gold-linked
currency, as was the case in the United States.
Because buying and selling gold bullion was somewhat cumbersome --
you have to move the stuff around -- gold redeemability was the less
common (although perhaps more important) operating procedure. The
most common was adjustments in the monetary base by way of
discretionary purchases and sales of debt assets by the GSA.
Note that "debt assets" could include loans as well. Both the Bank
of England and the U.S. commercial banks were also regular lenders.
Thus, one way to reduce the monetary base would be to receive money
in payment of interest or principal, as happens every day with a
commercial bank, and take the money received and make it disappear,
thus shrinking the monetary base. The bank's assets would shrink,
just as if it had sold government bonds. One way to increase the
monetary base would be to make new loans, using "freshly printed
money," which would increase the monetary base and also the bank's
assets. The activities of currency management and the activities of
regular commercial banking were often intertwined. Does this make
things confusing? Oh, yes. That's why there has been a long effort
to separate the two actions. This happened with the Bank of England
in 1844, in which the Bank's activities were separated into an Issue
Department (for the currency) and a Banking Department (for regular
lending). In the U.S., responsibility for the currency was
ultimately taken from regular commercial banks, mostly during the
1920s, and concentrated in the Federal Reserve. This is why our
money today says "Federal Reserve Notes."
The activities of currency management and commercial banking have
been separate for quite a while now. But, you can see where we get
these stories about how "banks create money and then loan it to
you." That is, indeed, what they did in the old days, to some
extent. Not all loans were done this way, but some were. The funny
thing is how these things keep going for generation after
generation, long after they have become obsolete. It just goes to
show you how many people are actually thinking about these things
(0.01%) , and how many people (99.99%) are just repeating something
they heard somewhere.
So, as an example of a hybrid system, we could have gold
redeemability at +-2% from the parity value. For example, you could
have a parity of $1000/oz. The GSA would sell gold at $1020/oz. --
in other words, when the value of the A$ is sagging, and is 2% below
its parity, such that it now takes $1020 to buy an ounce of gold
instead of the promised $1000, the GSA would sell you the gold, buy
the A$, and make the A$ received in the sale disappear, thus
shrinking the monetary base.
At $980/oz., the GSA would buy gold, and give you A$ in return.
These A$ would be "freshly printed," and thus the purchase of gold
would increase the monetary base.
In practice, transport of gold has a small cost. Let's just say it
is 1%. In other words, it costs the private market buyer or seller
1% in expenses to move the gold to and from the GSA's vaults to
their own vaults. So, even if the GSA promises to buy and sell gold
at exactly $1000/oz., in practice the private market participant
wouldn't buy until the market value of the A$ drops to $1010 or
beyond, at which point it becomes profitable to buy gold from the
GSA. On the other side, the private market participant would sell
the GSA gold when the market value of the A$ was $990/oz. Thus, even
if there is not a formalized buy/sell spread or "trading band," the
difficulties of moving gold creates one naturally.
Personally, instead of having a buy/sell spread, I like to use a
"redeemption fee" or "monetization fee." In other words, the GSA
will buy or sell gold at exactly $1000/oz., but, if you want to go
to the GSA to buy or sell your gold instead of a private market
buyer/seller, you have to pay a little fee for the privilege. The
GSA might sell gold at $1000/oz. and then charge a $20/oz. (2%)
"redemption fee." This is exactly the same as selling for $1020
without a fee, but I think it helps cement the idea that the
government's parity for the A$ is exactly $1000/oz., not $980/$1020.
Within our hybrid system, we now also add purchases and sales of
assets (bonds) by the GSA, to adjust the monetary base. For example,
if the A$ was trading at $1005/oz. one day, i.e. a little weak but
not yet at the $1020/oz. gold bullion redeemability point, the GSA
would sell bonds, take A$ in return, and thus shrink the monetary
base. This would support the value of the A$ and perhaps move it
back to its $1000/oz. parity value. If the GSA constantly reacted to
these smaller movements with such actions, theoretically the value
of the A$ would never get to the point at which private market
participants would go to the GSA and demand bullion in return.
Likewise, if the A$ was trading at $995/oz. one day, the GSA would
buy bonds and pay for their purchase in "freshly printed" money,
thus increasing the monetary base and consequently lowering the
value of the A$. The value of the A$ should never get to the point
where people are taking gold to the GSA to get A$ in return.
As noted before, these smaller adjustments by the GSA could be done
either on a discretionary basis, or perhaps formalized in some way.
In the past, they were generally discretionary.
It would look something like this:
That's enough for this week. Actually, I haven't got to the original
topic I wanted to talk about yet. Eventually!