Home > Macro/Monetary Theory > On the Convertibility of Fiat Money

On the Convertibility of Fiat Money

Okay, I got some good comments from J.P. Koning and others, for which I am grateful, on my “debt chartalism” theory explaining the value of fiat money.  I set out to answer them all in one post but as I dig into it I realize that there is a lot there that I take for granted that probably deserves a more careful explanation so I will begin here with a more thorough discussion of the fundamental nature of fiat money relative to a convertible currency explicitly backed by a hard asset.  Then in a future post, I will try to delve into monetary policy, inflation, determining the specific value of a dollar etc.

The important thing to notice here is that the nature of fiat money is actually not that different from “hard” money.  To see what I mean, let me start with a bank issuing gold notes and go through some minor alterations that I think most people would agree would not destroy the value of those notes.  In what follows I will represent the banking sector with a single bank.  The relationship between the CB and individual banks raises a bunch of issues which I don’t think are necessary to make my point (and are distracting) so I will abstract from them for now.

We start with a bank that has no assets and some people who have gold and want to exchange it for bank notes because the notes are more liquid (easier to use for transacting).  People take their gold to the bank and the bank creates notes “out of thin air” which they, or anyone else, can return at any time and exchange for “their” gold.  Most people seem (rightly) to have no difficulty seeing why these notes would be valuable.

Now imagine that instead of taking your gold to the bank and getting notes, the bank comes to your house, verifies that you have the gold, issues you the notes and tells you that in one year, you have to “redeem” the notes or else they will take your gold.  These notes may be convertible by anyone at any time at the bank or they may not be.  In the first case, of course, the bank will have to have some other gold in their vaults to deal with these redemptions but as long as there is sufficient demand for liquidity relative to the quantity of notes in circulation, most of the notes will float and redemptions will be limited.  But the important point is that you have not turned a perfectly legitimate enterprise into a Ponzi scheme by letting people hold the gold in their individual vaults rather than the bank’s vault.  The gold is still backing the notes because of the nature of the contract that created them.

Looking at it this way, hopefully we can see that it is not the universal convertibility that gives the notes value.  This, of course, may give them somewhat more value but mainly it is a mechanism which imposes discipline on the bank and inspires confidence in the holders of their notes.  This is a discussion for another time.  But consider whether these notes would still be valuable if the bank withdrew the universal convertibility.  By this I mean that you still have to have 100 oz. of notes in one year to keep your gold (let us call this individual convertibility) but if you trade the notes to someone else, they can’t just take them to the bank and cash them in for gold.

Will these notes still be valuable?  The answer I think is yes because they are still convertible.  It is just that they are only convertible at a given time by a specific individual.  But because there are some individuals who are able to “convert” the notes to gold at a fixed rate, those individuals will always be willing to trade real goods and services for those notes.  In this way the gold is still “backing” the value of the notes, it is just that the convertibility has been changed from a general obligation to a specific obligation between the bank and certain individuals.

Finally, imagine that the bank accepts other goods besides gold to back their notes.  This raises the issue of denomination.  You could have “silver notes” and “ruby notes” and “corn notes” because these are essentially commodities but these notes would make general exchange somewhat more complicated (everyone would have to keep track of the relative values of all of these things) and you could not create specific notes for idiosyncratic goods like “house notes” and “car notes” and “boat notes” because one man’s house is not worth the same as another’s.

One solution to this problem is to denominate all notes in a single good.  So if you want to convert your house into notes, instead of the bank issuing you “house notes” they issue you gold notes for a quantity of gold equal to the value of the house (or somewhat less).  Then you have the right to “redeem” some quantity of “gold notes” for your house at some point (or points) in the future at a specific, predetermined rate.  In this case, even though the notes represent a value denominated in gold, it is your house that is actually “backing” those notes.

But if the notes are not universally convertible into gold, the denomination is arbitrary.  The bank can just as easily issue you notes denominated in “quatloos” and say that you need to return 1,762 quatloo notes in order to keep your house.  People can put up whatever assets they want, the bank can assess their values in quatloos and issue notes which are redeemable at various rates for various asset.  As long as these notes are interchangeable and denominated in the same units (even if that unit is completely arbitrary), they will be able to trade them amongst themselves and there will be a somewhat stable demand for them because there are always people who can “redeem” them for real assets at various rates.

In this way, the quatloo notes are still backed by real goods, it is just that those goods are not as obvious because the type of goods and the rate of convertibility vary from person to person.  To get a better feel for the similarity between the creation of universally convertible notes and individually convertible notes, note (haha) that under a classical gold standard with fractional reserve banking, the two actually exist side by side.

Let us return to our bank which starts with no assets.  The bank attracts some “depositors” who deposit 100 oz. of gold and are issued universally convertible gold notes.  The gold goes on the bank’s balance sheet as an asset (obviously) and the notes go on as a liability because the holders of the notes can bring them in at any time and redeem them for gold.  The balance sheet then looks like this:

Assets                                          Liabilities

Gold: 100 oz.                                 Notes outstanding: 100 oz.

Now imagine that somebody else comes into the bank and wants a loan worth 100 oz. to buy a house.  The bank creates “out of thin air” another 100 oz. of gold notes and lends them to this person and accepts the house as collateral.  The bank’s balance sheet then looks like this.

Assets                                          Liabilities

Gold: 100 oz.                                 Notes outstanding: 200 oz.

Acc. Receivable: 100 oz.

The accounting is the same except the asset which goes on the books is an account receivable instead of gold.  But there is a real asset behind that account receivable, namely the house.  The house is backing the notes of the bank in the same way that the gold from the first guy is.  The fact that the bank does not have enough gold to redeem all of its notes does not make it insolvent.  This is where Rothbard’s people go off the rails.  Fractional reserve banking is not based on deceiving people into thinking that there is more gold in the vault than there actually is.  It is just a method of turning various types of assets into a more liquid form.  The total assets backing that money grow along with the supply of money.

If the house burns down and the borrower defaults, then the bank will become insolvent.  There will be too many notes outstanding for the total assets on the bank’s balance sheet.  Then people may rush to redeem their notes and find out that there is not enough gold.  But this only happens because the demand for notes from the borrower becomes removed from the market without reducing the number of notes (or delivering the house, which could be sold for notes, to the bank).  (Of course, a bank could become illiquid without being insolvent but this is a matter of demand for notes, liquidity preference, etc. that is well-understood, and not worth getting into here.)

In this case, you have a situation where anyone can redeem notes for gold and one person can redeem notes for gold or for a house.  The gold-redeemability only keeps the value of the notes anchored to the value of gold.  This prevents the bank from issuing so many notes that the value of them falls.  If they issue enough notes that the risk/liquidity premium on them relative to gold becomes negative, then people will start redeeming them for gold.  If this is not the case, the notes will float and the bank will be able to keep issuing more.  But the gold is not the sole source of backing.

Of course, the more loans a bank makes in this way, the more other assets will make up the “backing” of their notes.  If at some point, they drop the universal gold redeemability, the value of a note will no longer be anchored to the value of gold but it will not  just evaporate into thin air because there will still be a large amount of other assets “backing” it.  People will still have individual convertibility in various assets.  In fact–especially if it is a monopoly–the bank may be able to accumulate profits over time which can allow it to basically absorb the value of the gold on its balance sheet.

So this change from convertible gold notes to fiat money is not as dramatic a shift as many make it out to be.  Once it is accomplished, the value of a dollar comes down to flows of credit–how much new credit is being created relative to how much needs to be paid back (and of course expectations about the future matter a lot).  The CB can influence this in various ways.  I will try to dive into that discussion soon, though I admit up front that I don’t have everything perfectly pinned down yet.


  1. J Thomas
    March 4, 2014 at 2:36 am

    This is a clear, lucid explanation. You make it obvious that things could go just as you say.

    Your explanation implies some failure modes too. Like, if a man gets a loan based on the value of stock market shares he owns, the actual value could fluctuate rapidly. Far faster than the value of his house. If he uses the loan money to buy more shares and then uses them as collateral for a new loan, it just gets worse. This is one of the classical explanations for some of the events leading to the Great Depression.

    Similarly, if people keep buying each other’s houses for more money, the official value of the houses keeps going up. They might imagine they can live off their profits. If they assume that the prices of their new houses will go up fast enough that they can sell before they need to make many payments, and the bank goes along and lends to them…. And then at some point it must stop lending them the money to pay the interest on their new loans….

    Banks need to be conservative about assigning value to volatile products. But maybe there are complexities involved in how to be conservative when in the short and medium run the radical bankers outcompete others….

  2. Free Radical
    March 4, 2014 at 3:36 am

    Thanks Thomas, that’s basically right about margin and the 1929 stock market crash. The connection to the economy at large is not as obvious but I think it is there (similarly with the housing market in 2009). The important thing to notice is that the value of these loans is fixed and doesn’t fluctuate when the market values of the assets (or the currecy) change. This is how they serve as an anchor for the value of the currency relative to real goods. Economists are always talking about “sticky prices/wages” but they rarely mention the “stickiness” of debt which I think is the much more important issue.

  3. J Thomas
    March 4, 2014 at 6:33 am

    “The important thing to notice is that the value of these loans is fixed and doesn’t fluctuate when the market values of the assets (or the currecy) change.”

    That’s clearly important, and doesn’t get enough notice.

    “This is how they serve as an anchor for the value of the currency relative to real goods.”

    I’m not as clear about that. If the value of the real goods fluctuates but the loans don’t, that would seem to distort the value of the currency more than anchor it.

    “Economists are always talking about “sticky prices/wages” but they rarely mention the “stickiness” of debt which I think is the much more important issue.”

    Yes! But to some extend debt can be refinanced. Doesn’t that make it less sticky when interest rates head downward, but fully sticky when they go up? (Except for ARMs.)

  4. March 5, 2014 at 1:45 am

    No major criticisms, except maybe the last paragraph. I’ll wait till your next.

  1. March 12, 2014 at 12:26 am
  2. April 9, 2014 at 4:13 am

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: