Home > Macro/Monetary Theory > Why is Fiat Money Valuable

Why is Fiat Money Valuable

I have been developing a theory of money and banking for a while and I noticed that Scott Sumner did a post on why fiat money has value in which he listed five theories explaining this and mine wasn’t among them.  This made me think it might be kind of novel so I want to try to get it out there and see what people think about it.  First I will make a few points relating to Sumner’s post, then develop a model of banking under a specie standard that leads to an explanation of fiat money having value. 

First, what I am trying to explain is why “the dollar” has value, not just paper currency.  This explanation should apply to all dollar denominated assets including bonds, demand deposits etc.  To see what I mean by this consider explanation #5 on Sumner’s list which admittedly, he doesn’t think is important but puts up only to appease a regular commenter.  This explanation says that the currency has value because the central bank holds assets which it stands ready to sell in order to keep that value from falling.  But the assets the Fed holds are (mainly) dollar denominated bonds.  If people suddenly lost confidence in the dollar, how could confidence be restored by the Fed offering to trade in exchange for your dollars, a piece of paper promising to pay you in the future….in dollars.  This explanation dodges the real issue which is what keeps the value of a dollar, either today or in the future, from falling to nothing.  Similarly #1 on the list tries to explain why currency could have value relative to T-bills but this does nothing to explain why a $100 T-bill would be valuable in the first place.

#2 on Sumner’s list is I think the most commonly held belief.  “There is a sort of social contract/network effects/focal point thing going on.  People accept it as money because others accept it as money.”  I won’t try to prove that this isn’t possible but to me it seems like a fallback explanation we resort to because we can’t think of something better.  I am, and think others should be, highly skeptical about this explanation.

Chartalism (#4 on the list) is actually very similar to what I describe here except I think that they are missing the main issue which is not taxes but debt.

Start with an economy where some good, which has “intrinsic value” (value for some reason other than exchange and storing value), is being used as the primary medium of exchange. Call this good currency. Assume that the quantity of this good is fixed by nature. In this case, if people know and trust one another, then it will be convenient to carry out some of their trade with credit rather than trading goods (including currency) directly. Currency can be used to settle imbalances in credit which arise between two people over time. Here is an example
A brewer makes beer which he sells to the local tavern. Each week, the tavern buys some quantity of beer from the brewer and writes him an I.O.U. for some amount of gold which he can redeem from the brewery at any time. Several nights per week, the brewer has dinner at the tavern and each time pays with I.O.U.s. These may be the tavern’s I.O.U.s which he pays back, or they could be I.O.U.s which promise the brewer will pay gold to the tavern keeper upon presenting them. Over the course of a few months, a balance may grow on one side or the other. If the tavern buys a lot of beer from the brewer and the brewer doesn’t eat at the tavern very often, then the brewer will accumulate credit from the tavern keeper. At some point he will wish to collect the gold but doing this, let’s say once per month, is more convenient than carrying and counting out coins every time an exchange of goods or services occurs between them.
This use of credit increases the velocity of money. It allows more trade to be carried out with a fixed quantity of money at a lower cost of transacting. There are problems with this arrangement though. Primarily, there may be a deficit of trust. In order for someone to trust someone else’s notes will be good, they must know that person pretty well. This will make credit unsuitable for anonymous trade.
Similarly, credit may not be convenient between people who trade infrequently. If everyone knew and trusted everyone else in the economy, then I.O.U.s would likely be transferable and this would allow credit to be used for one-time transactions. John could write an I.O.U. to Jane for a one-time purchase and Jane could simply trade that I.O.U. to someone else for something she wanted. This third-party would accept John’s I.O.U. because he knows John and knows other people know John and he will expect to be able to trade it to others or redeem it with John. As long as people believe that John has enough assets to cover all of the I.O.U.s he is floating, they will feel no need to rush to redeem them. Every once in a while someone will trade these back to John for whatever goods/services John produces and every once in a while, someone who happens to be holding his I.O.U.s will desire currency and will come to him to redeem them.
The problem with this situation is simply that everyone in the economy is not likely to know and trust everyone else. The people Jane trades with may not know/trust John, in which case they will be unlikely to accept his I.O.U.s. This fact will likely make Jane unwilling to take John’s I.O.U.s in the first place and all infrequent trade will be conducted with currency.  This results in a situation in which people would prefer to conduct trade with paper but find the risk and hassle associated with doing so in the way described above prohibitively high. This allows for the possibility of profits if someone can find a way to provide low-risk, well-recognized paper credit. This is the primary function of a traditional bank.
A traditional bank accepted deposits of currency (let’s say gold) and issues bank notes which are I.O.U.s from the bank redeemable in currency. There seem to be two predominant theories on the main function of banks. One theory says that banks originated as warehouses for currency and the banks covertly issued more credit than they had currency in a dishonest pursuit of profits and in so doing, began a progression of crimes against the public that is still being perpetrated today. Another theory says that the primary function of a traditional bank was to funnel funds into profitable investments making a profit on the spread between the rate at which they borrow and the rate at which they loan which is derived from their ability to identify good investments.
What I am proposing here is that the primary function of a traditional bank is to provide a convenient and stable medium of exchange. People deposit their currency in the bank and receive bank notes because these are easier to store, transport, and trade with. In short, using these bank notes is just like using one’s own credit except that the bank guarantees them. This means that the person you trade with doesn’t need to know and trust you, they only need to know and trust the bank whose notes you are using. In other words, the bank specializes in maintaining a credible reputation for solvency. This allows for anonymous and infrequent trade to be conducted using credit rather than currency. It is worth noting here that a checking account functions in a similar way but with a few differences that I will discuss later.
Since conducting trade using credit is more convenient than using currency, people will be willing to pay some premium for holding some of their assets in the form of bank credit rather than currency up to a point. Let us assume that the rate of conversion between currency and bank notes is fixed at 1:1. In other words, assume that you can exchange 1 oz. of gold (or whatever) for a bank note redeemable for 1 oz. of gold, or in still other words, assume there is no fee (either positive or negative) for converting one into the other. This means that people will choose to hold them in quantities such that they are indifferent between them on the margin.
Note that this would not be possible if banks were forced (either by law or by market forces) to hold 100% reserves, since this would not allow them to cover expenses or make a profit. However, if banks can float additional notes above and beyond the quantity of currency in their vaults, then they can generate income by printing additional notes and lending them at some positive interest rate. (Note further that this does not mean the conversion rate must be 1:1, it just means it could be. A more detailed model would include a supply and demand for currency/bank notes and an endogenous conversion rate but starting down that path quickly complicates things and is not necessary to make my main point.)
As an example consider a bank who trades $100 in bank notes for $100 in currency (let a dollar be defined as some fixed quantity of gold, it doesn’t matter how much). The bank, noticing that very few of these notes come back to them quickly decides it can issue some additional notes and lend them at the market rate i. Let’s say that i=.10 and the bank lends an additional $200 worth of notes. If there is sufficient demand for circulating bank notes, then these will remain in circulation for the duration of the loans and the bank (assuming the loans are repaid) will have revenue of $20.
Now here is the important part that I think most people are missing. In the interim, between the bank making the loans and the loans being repaid, there are $300 worth of bank notes circulating from this bank which are all supposedly redeemable for gold but the bank only has $100 of gold. This does not mean that the $300 of bank notes are only backed by $100 of gold. When the bank loaned the additional $200, it received an asset. This asset goes on the books of the bank as “accounts receivable” which everyone realizes. What most people overlook is that in most cases there are real assets behind that accounting asset. Usually these are the assets which the borrowers purchased with the $200. So if the loan is for mortgages to buy houses, then the houses are backing those bank notes. If it is a business loan, then the assets of the businesses are backing them.
This does not rule out a bank run. If the bank issues too many notes, then there will not be sufficient demand for them as a medium of exchange and they will come back to the bank and be redeemed for currency. This will drain the reserves of the bank which will result in a lower reserve ratio and could lead to increased concern about the solvency of the bank which could cause more people to redeem their notes and so on. If this goes too far, the bank could run out of reserves and be forced to close its doors because it can’t redeem its notes for the accounts receivable which are not yet due. This prospect is widely understood.
What I don’t think is widely understood is that the assets which are collateralizing the loans which increased the supply of notes beyond the quantity of gold in the vault create their own demand for “money” in the form of either gold or bank notes because presumably, the two are perfect substitutes for the purpose of paying back those loans. To see what I mean, imagine that you were caught holding some of these notes after a bank run occurred and the bank has run out of gold and closed its doors. Does this mean your notes are worthless? Probably not.
At this point you are unable to redeem your notes for gold and nobody else is able to redeem them for gold but this does not mean that nobody would want them. Somewhere out there are people who signed a contract saying that they had to pay back $200 of gold or notes or else lose their houses/businesses/etc. Presumably, those people took out the loans under the expectation that in the future they would produce something valuable that they could trade for money (gold or bank notes) in order to pay off their loans (if not, then they were just bad loans which is a different story). For those people, the notes are redeemable for real assets (houses/businesses/etc.) at a fixed rate. Therefore those people will be willing to trade real assets for those notes even though there is no gold still backing them.
Of course, exactly what would happen in this case depends on the law and the nature of the contracts involved. One can imagine trading the notes to those people who had the ability to redeem them (the borrowers). What’s more likely is that the bank would go into some sort of receivership and the assets (accounts receivable) would be sold for present value and the proceeds distributed to the holders of the notes. Whatever the process, annoying though it may be, the important thing to notice is that there are real assets still involved so long as the collateral still exists. These real assets, along with the gold in the vault, all “back” the notes in the sense that the notes represent a claim on them. It just happens to be the case that the claim on the houses/businesses/etc. is not as direct—anyone can’t just show up, present the notes and demand them—and therefore not as obvious. Nonetheless, the claim on these assets does provide support for the value of the notes.
The intended (for lack of a better term) and most likely (and typically observed) effect of this is that people will not be inclined to run on the bank unless the bank’s loans (and the collateral backing those loans) start to look iffy. This is, of course, assuming that the bank does not issue “too much” credit. When the market is in equilibrium, the interest rate will adjust so that the amount of circulating bank credit is equal to what people demand at that interest rate given the reserve ratios of the banks (along with the conversion rate between currency and bank notes which we have assumed to be 1:1). If a bank issues “too much” credit, it will simply come back to them and drain their reserves which they will have an interest in avoiding.
But in the absence of a bank run, when the above loans are paid off, the borrowers will return $220 to the bank, either in the form of bank notes, which will reduce the amount of outstanding notes to $80, or in gold which will increase the amount of reserves to $320, or some combination of the two. Either way, at that point the gold in the bank’s vault will be $20 more than the quantity of outstanding notes. This $20 represents the revenue to the bank which they can devote to any of the purposes that any other business could devote their profits. They could keep it in the vaults and issue even more credit in the form of loans or take it out and spend it on consumption or building new branches thereby putting it back into circulation. This choice is not important here as it has little bearing on the aggregate demand for circulating bank notes. The important thing is that, if banking is competitive, the expected rate of profit in that sector would be in line with other industries with similar risk profiles, the industry would expand (both reserves and branches and employees etc.) to the point where the marginal product of investment was equal to the real interest rate and the revenue and profits would circulate back into the economy either through investment or wages to employees/owners and corresponding consumption expenditures.
On the other hand, imagine that there is one big bank with a monopoly on the issue of bank notes. In this case, it may be able to generate some monopoly profits in excess of what is necessary to pay the expenses of operating. This means that some of those profits need not be injected back into circulation but could instead be used to reduce the liabilities of the bank. So imagine that the $20 operating profit above does not go to paying expenses and dividends but just becomes $20 worth of gold in the vault that the bank owns free and clear of any claim from circulating notes. So the bank has $80 in notes outstanding and $100 of gold in the vault. Now, in order to keep the quantity of notes in circulation constant, the bank issues $220 of new loans at the same rate i=.10. Then in the next period, when the loans are due to be repaid, they issue $242 in new loans and so on, keeping the quantity of circulating notes at $300. After 3 more periods (so 5 periods total) the amount of new loans necessary to maintain a quantity of $300 of circulating notes will be $322.1. These loans will presumably be backed by $322.1 of real assets (houses/businesses/etc.) making the accumulated profits of the bank equal to $122.1 which is greater than the total quantity of gold in their vault. At this point, the gold is no longer necessary to “back” these notes because the collateral backing the debt which created them is large enough on its own.
At this point, if the bank declares that it will no longer redeem the notes for gold, some would imagine that the notes would become worthless. However, this would not be the case. Some people having accumulated notes with no debt might decide they don’t want to hold the notes if they are no longer redeemable for gold but these people would look around and find that there were plenty of people willing to trade real goods for them because there would be plenty of people holding debt and not enough notes to repay it. These people would be willing to trade goods and labor for the notes because they still have the ability to “redeem” the notes for real goods, namely their houses/businesses/etc. These goods will maintain a floor for the value of the notes so that even if you are not one of the ones in debt, you will be able to remain confident that the notes will command the labor and assets of those who are in debt. In this way the debt (and real assets backing it) backs the notes. So there is no reason for people, indebted or otherwise, to abandon use of the notes as both a store of value and medium of exchange.
This is the situation I believe we were in in 1968 when we officially went off the gold standard and remain in to this day. The debt load of society became large enough to keep the value of a dollar (a Federal Reserve note) propped up without the need to redeem dollars for gold. This is why the value of a dollar doesn’t suddenly fall through the floor. It isn’t just that everyone expects everyone else to take them so we just go along with it out of convention even though they aren’t really worth anything. It’s because most of us need to get dollars to pay off our houses, boats, cars, businesses, credit cards, student loans, and on and on. In fact, this is not only why the value of a dollar doesn’t fall through the floor, it is why we have to keep expanding the money supply and the debt load exponentially to keep the value of a dollar from going through the roof. But that is an issue for another post.

P.S. I will try to devise a test of this theory and dig up some data to refute/support it in the near future.

Categories: Macro/Monetary Theory Tags: ,
  1. April 5, 2013 at 1:18 am

    I think this is a pretty compelling account of fiat money. Looking back at Sumner’s post you can find a comment by Mike Sproul mentioning the “backing” theory of fiat money. in particular he states that it is a fallacy to say inconvertible = unbacked. Seems this is a “backing” theory, is it not?

    One of my first thoughts in reading this is the importance of law in making sure that people trust the backing. We need laws to make sure that people trust that if a bank ever went under, the assets their claims would be distributed to people who hold bank notes.

    When we extend this idea to a whole nation, it seems that people can have faith in fiat currency because it is the law of the land that debts can be repaid in this currency. These are the same laws that uphold a whole host of property rights.If people start to doubt that a government protects property rights, fiat currency is at risk.

  2. Free Radical
    April 5, 2013 at 2:51 am

    I agree that it is a “backing” theory but it differs from Mike Sproul’s theory (at least as I understand it) in that he seems to assert that the assets backing the dollar are the assets of the Federal Reserve which are financial assets, meaning they are dollar denominated themselves. As I briefly mentioned in the intro, this does not get us out of the fiat money paradox because if the value of a dollar collapses in real terms (the quantity of real goods people are willing to trade for a dollar falls dramatically), then the value of these dollar denominated financial assets will fall equally dramatically. A sound backing theory has got to rely on real assets. Luckily, there are a ton of real assets backing the dollar as there is no shortage of loans which are “backed” by real collateral which dollars can be “converted” into (so to speak).

    It does rely on the law allowing payment of debts with the fiat money but it’s worth mentioning that this could be accomplished through mutually voluntary contracts between banks and borrows (obviously enforced by law), it doesn’t necessarily have to be by a general decree that the dollar must accepted for payment of all debts. Of course, we have such a decree now anyway, but I’m trying to sort of connect the dots between an organic, decentralized banking system and the one we have today. I’m a Fed. detractor but I think a lot of the other anti-fiat money theories floating around out there get a little carried away and act like fiat money works only because the government forces people to take it. I think it is a little more subtle than that. I do think that a central-bank monopoly is necessary to push us over the threshold where there is so much debt that the money can stand alone and I think that is undesirable in the long run.

  3. Free Radical
    April 5, 2013 at 3:01 am

    P.S. If you have ever noticed that the major crashes we’ve had since the Fed was instituted like in 1929 and 2008 were deflationary, meaning asset values pretty universally fell in terms of dollars, and wondered how it can happen that real assets across the board can just all become worth a lot less over night, it’s exactly like the inflation alarmists tell us…only backward. We don’t try to get rid of our worthless paper dollars. Since we owe more money that there is in circulation we rely on a constantly expanding steam of credit to keep injecting more money into the system driving inflation. If we get the idea that that inflation won’t be there, we try to get out of those assets and get our hands on the insufficient quantity of dollars so we can keep our houses and cars and businesses. This makes stock, real estate, etc. values plummet.

  4. April 5, 2013 at 4:50 am

    I agree that this is a major part of the story, even that we don’t need a universal law stating that all debts can be repaid in dollars. But I don’t think it is the whole story. One of the comments in Sumner’s post that struck me as interesting pointed out the case of Bitcoin. I’m not sure exactly why bitcoins are valuable, but I don’t think it is because they are backed by assetts. I think people just like them as currency + some kind of network effect. Maybe after a while some assett backing gets thrown into the mix.

    Yes, I have noticed that major recessions are deflationary. But are you saying that the deflation is a cause or an effect of the recession? or both? In any case, it seems that deflation as a cause is consistent with the aggregate demand story.

  5. Free Radical
    April 5, 2013 at 5:05 pm

    I think I’m saying that it’s a cause but it’s sort of a matter of how you look at it. Basically I’m saying that the cause is that the money supply doesn’t grow as fast as the debt and/or people for some reason don’t expect it to do so in the future. I think this is actually pretty similar to what Sumner says (NGDP falls below trend) and what Keynesians say (insufficient aggregate demand). I would say that the fall in output and employment is caused by the deflation. For the record, I think moderate and persistent deflation would be natural and not undesirable in a decentralized economy. It’s just that we’ve set up a system where deflation is catastrophic because we aren’t expecting it and we are leveraged to the hilt due to our expectations of perpetual inflation.

    It’s interesting you bring up Bitcoin, I was actually just trying to figure out what that is. As I understand it you are correct that it is not backed by any real asset which is fascinating. I think this does make it a test of the “network” theory which admittedly it is passing so far. I am skeptical that it will be sustainable in the long run but I could be wrong. If the value of it collapses, they will call it an elaborate pyramid scheme and we know pyramid schemes can go on for a while, just not forever. Still, I wish I had bought some a few weeks ago.


  6. April 23, 2013 at 10:11 pm

    “it differs from Mike Sproul’s theory (at least as I understand it) in that he seems to assert that the assets backing the dollar are the assets of the Federal Reserve which are financial assets, meaning they are dollar denominated themselves. As I briefly mentioned in the intro, this does not get us out of the fiat money paradox because if the value of a dollar collapses in real terms (the quantity of real goods people are willing to trade for a dollar falls dramatically), then the value of these dollar denominated financial assets will fall equally dramatically.”

    The backing theory does require that at least some ‘real’ assets back the dollar, but it’s OK if some of the assets are dollar-denominated. For example, a bank issues 100 paper dollars, backed by 100 oz of silver (real backing). Each dollar is worth 1 oz. Then the bank issues another $200 in exchange for bonds worth $200 (nominal backing, since it’s dollar-denominated). Define E as the value of the dollar (oz/$). Setting assets (100 oz + $200 of bonds, which are worth 200E oz.) equal to liabilities ($300 worth E oz each) yields:

    100+200E=300E, or E=1 oz/$

    Now suppose the bank loses 30 oz of silver, which is a loss of 10% of its assets. The above equation becomes

    70+200E=300E, or E=0.7 oz/$ (a 30% loss of value)

    Note that the 10% loss of assets caused 30% inflation, because of inflationary feedback. This is the process whereby a loss of assets makes the dollar lose value, which makes dollar-denominated bonds lose value, which makes the dollar fall more, etc.

  7. Free Radical
    April 25, 2013 at 6:15 am

    Hey Mike Sproul, good to hear from you on this, that illuminates the backing theory somewhat better. I wonder if anyone has tested this empirically. It seems like it would be fairly straightforward to do so. I’m not sure how you would define the assets and liabilities. If you put all assets on the left and all liabilities on the right and just separate out the gold on the Fed’s balance sheet, then you will necessarily get E=42.22 (the price at which they carry it on their balance sheet). In order for this theory to hold up, you should be able to define things in such a way that you can solve for that E at any time and get something close to the dollar price of gold.

    I think what I am saying can be described in the context of this theory as a situation in which the financial assets of the central bank become larger than the liabilities so that E becomes negative for any (positive) value of the real assets. In that case the real assets become unnecessary. I seem to remember you saying a while back something like “the dollar has value because the Federal Reserve stands ready to sell assets to keep its value up and people believe they will do so however much is necessary to keep it up.” So with what you describe above, they have to essentially be ready to trade the silver to keep the value of the dollar up. If they came out and said “you know what, we’re keeping all the silver, we’re never giving it back” the value of the dollar should fall to zero since there are more than enough of them in circulation to pay back the loans to the Fed.

    I’m saying that the Fed can say “we’re never giving the gold back” and dollars will remain scarce because there would not be enough to pay off all the debt (assuming they stopped printing more). This is also turning out to be more difficult to test than I imagined for reasons I won’t get into here.

  8. April 25, 2013 at 3:04 pm

    Empirical testing has been done by Sargent, Thomas Cunningham, Bomberger and Makinen, and others that are referenced in my paper titled “There’s No Such Thing as Fiat Money”. Unfortunately, empirical testing is difficult. Just to give you some idea, I’m told that finance professors have a very hard time empirically confirming that stock prices are determined by the assets and liabilities of the issuing firm. Doing the same thing for money is even harder.

    For one thing, suppose the Fed dumped all its assets in the ocean. Most of us would expect the government to bail out the fed, so anyone who thought he could just compare the Fed’s assets with its liabilities will be surprised to learn that he should have also looked at the government’s assets and liabilities. There’s also the problem that bank notes are a first and paramount lien on the issuer’s assets, so even if the fed, or the government, became insolvent, paper dollars would be first in line for the government’s assets, so they might hold their value even as the government defaulted on most of its other debts.

    Not sure what you mean by E becoming negative. As long as the money-issuer is able to buy back the money it has issued, the value of the dollar (what I call E) will be positive.

  9. Free Radical
    April 25, 2013 at 8:46 pm


    This can’t be accomplished just by issuing more paper but it can be accomplished over time since they are able to charge a positive interest rate on their loans. So start with your scenario where the bank has 100 oz. of silver matched by $100 of bank notes and issues $200 of notes which they lend at i=.05 with a term of 1 year and imagine that they want to keep the money supply at $300. After a year, the original loans come due for the amount of $210 so they have to issue $210 of new loans to keep the money supply at $300. After another year (assuming the interest rate is constant) they will have to issue 220.5. After eight years, if they do not spend any of the interest proceeds on anything (including remitting it back to the treasury to spend) they will have accounts receivable of $310.26 and $300 of currency outstanding. At this point E from your equation would be negative. Then they could dump all the silver in the ocean and the dollar would still maintain value purely for its ability to repay the loans.

    Furthermore, once you were in this situation, if the bank stopped issuing enough new credit, there would be a rush for the exit of credit not of the dollar so you would be prone to deflationary (and not inflationary) panics. The monetary authority would have to keep growing the money supply exponentially to hold this off.

    • April 25, 2013 at 10:58 pm

      I had actually never thought of that case before, where E=-1. At first glance, it looks like it can’t be right because if the bonds had initially been worth $200, the equation would have been 100+200E=300E, for E=1 oz/$. Then the money issuer gets $200 MORE assets, so now it’s 100+400E=300E, and somehow the dollar becomes worth LESS.

      I’m not an accountant, so excuse any lameness in my explanation, but the problem is that you are omitting the firm’s net worth (or, equivalently, stockholder’s equity) As the money issuer gets the extra $200 of bonds on the asset side, there will also appear another $200 worth of net worth on the liability side, so you get 100+400E=300E + 200E, or E=1 oz./$.

      Note that this means that the extra assets will not increase the dollar’s value, because, after all, the dollar is just a claim to 1 oz. If net worth became negative, then the dollar could fall below 1 oz., so it looks a little different in that case.

  10. Free Radical
    April 26, 2013 at 1:43 am

    Actually I have an undergraduate degree in accounting (though I’m not one either) and basically accounting forces you to balance debits and credits so when you earn interest income (a debit) you have to credit stockholder’s equity or retained earnings or something like that but that is not a liability, it’s a separate category of accounts, this means your assets and liabilities do not have to match so you can accumulate more assets than liabilities by generating income. Indeed, this is basically the purpose of all businesses.

    You seem to be assuming that the only changes in the balance sheet come from issuing new credit, I am talking about something different which is the accumulation of assets over time via interest payments. Now theoretically the Fed pays all of this income back to the treasury and the member banks and if they did that and that money all went back into circulation, then what I’m talking about (E<0) wouldn't happen. However, I don't think it is clear that this actually happens especially when you think in terms of the banking system as a whole and not just the Fed. It is a little difficult to disentangle these things though and I'm not very good with data. Also when you try to build a big model that accounts for expectations, it gets pretty complicated pretty quickly.

  11. Free Radical
    April 26, 2013 at 1:46 am

    correction: I think income is actually a credit and equity is a debit. It’s been a while….

  12. Free Radical
    April 26, 2013 at 1:51 am

    Oh and just to be clear, I don’t think that equation actually defines the value of a dollar. This is an alternative theory. They are somewhat similar in explaining why a dollar is valuable but they aren’t compatible when it comes to determining what that value would be at any given time so for instance, I’m not predicting the value of a dollar would be negative if E were negative, I’m saying that the value is positive even if there is no real asset there because E is negative.

  13. April 26, 2013 at 3:52 am

    Imagine we are in a world where the value of X note is maintained because there is a lot of debt outstanding that can be paid in X.

    It is not clear to me that the ratio of outstanding debt to X will definitively determine the value of X. As you mentioned, if someone with a bunch of X has no debts to pay they can negotiate a trade with someone who does. But how do they determine the terms of that trade? It doesn’t depend entirely on the value of the collateral that the buyer of X put up against his loan.

    It also depends on bargaining power. Of course, the amount of X outstanding and the amount of debt denominated in X have a lot to do with determining the relative bargaining power of the parties, but it is not everything. The parties also may take into account things peculiar to their circumstances, and more importantly, their expectations about the future ratio of X to X denominated debt.

    An extreme case would happen if somehow it happened that there was no debt denominated in X at a certain point in time. But everyone expects some to arise in the future (maybe the government will require taxes to be paid in X soon). The value of X won’t go down to zero.

    I suppose my main point is that the value of currency doesn’t just depend on accounting. It depends on aggregated individual expectations about the future of the currency. It seems to me that there is a large number of reasons that people might think a particular kind of note is valuable. So I can’t really bring myself to believe that there is a single dominating explanation.

  14. Free Radical
    April 26, 2013 at 5:40 pm

    Mike C. I agree completely, that’s basically what I was trying to get at in my “just to be clear” comment. I don’t think you can look at the Fed’s balance sheet and determine what the value of a dollar should be. Expectations about the future are key. I basically see two situations, the “normal” state that usually prevails in which people believe that the Fed can keep the money supply one step ahead of the debt load and in which the value of a dollar is determined by liquidity preference inflation expectations, interest rates and the usual things that economists imagine determine the price level. But this model is sort of detached from an actual explanation of why a dollar has value.

    The other situation is one in which for some reason people lose “faith” about the effectiveness of future monetary policy. This case sort of forms the rock on which the other case is built. So inflation alarmists think that in such a case there would be no reason for the dollar to have value so its value would quickly go to zero or close to it. I don’t think it’s possible to build the house described above on that rock. I think it is actually the other way around. If, for instance, the Fed just stopped printing new money, the value of the dollar would be determined by the existing debt relative to the dollars. This doesn’t mean you can just write a simple equation with those two things and solve for the value of the dollar because it is really the assets backing the debt and peoples’ subjective values of them that matter.

    For instance, if someone has a mortgage on their house of X dollars and they have a value of the house which represents other goods and services they would be willing to give up to keep the house, then they will be willing to trade that many (however you measure them) goods and services for X dollars. If you have millions of people in this situation, you have normal supply curves for the various goods and services that they are capable of providing and you have demand curves derived by the people holding surplus dollars and these supply and demand curves will determine some set of prices.

    If the quantity of dollars is low relative to the quantity of debt, then I predict those prices in scenario two would be considered low relative to scenario 1 as people scramble to get the limited amount of dollars.

  15. May 5, 2013 at 10:32 pm

    Go back to the case of a bank that holds 100 oz of silver as backing for 100 paper IOU’s (dollars) that it has issued. Then the bank prints up another $200 in paper and lends them to someone who reliably promises to pay back $200 plus interest. It’s not really correct to say that these dollars have value because there is a lot of debt out there that must be paid in dollars. It’s really the case that the silver and the $200 IOU are the assets, and the dollars are the liability, and the assets back the liabilities.

    If some third party makes a loan which specifies repayment in dollars, that would have no effect on the assets and liabilities of the issuing bank, and so it would not affect the value of the dollar.

    Or suppose the issuing bank accepted loan repayments in silver, tobacco, chickens, etc., rather than insisting that its loan must be repaid in its own dollars. That would also have no effect on the assets and liabilities of the issuing bank, and so no effect on the value of the dollar.

    • May 7, 2013 at 5:37 am

      “that would have no effect on the assets and liabilities of the issuing bank, and so it would not affect the value of the dollar.”

      I think this is the key point where we disagree. I think accounting identities only go so far in explaining really complicated phenomena.

      Imagine a society that mostly relies on barter, and silver is the traditional medium of account and exchange. Then one silversmith issues notes that can be redeemed for silver. People really trust this silversmith, and they find that the new paper is way more convenient than actually owning silver. Soon another person starts making loans that are payable in these notes rather than silver, and people start putting their notes into accounts that can only be redeemed in notes. Fractional banking practices multiply the number of notes and because they are so convenient everyone starts accepting them as payment.

      I believe that eventually it is possible for the value of the notes to become independent of the assets and liabilities of the original silversmith. If an asteroid destroys the silversmith, it is not necessarily true that all of the notes would become worthless. I might not be able to redeem my notes for silver with that silversmith, but I might be able to pay off my mortgage, pay my employees salary, or even pay taxes because these contracts are still denominated in these notes. Sure, the equilibrium might be disrupted, but this is not an easy question to resolve.

      • Free Radical
        May 9, 2013 at 4:40 am

        That would be true only if the notes issued by the silversmith were indistinguishable from other notes, and of course if this were the case, there would be nothing stopping him from printing as many as he wanted, so this would probably not be the case without a central bank. If the notes were a claim on specific assets of a specific person, I suspect that they would become worthless when those assets were wiped off the map. The difficulty in verifying the assets backing a lot of random notes from random people is, I think, the main original explanation for the existence of banks, since a traditional bank essentially specialized in having a reputation which makes floating notes possible. I thought I had a post about that but I can’t seem to find it. Maybe I didn’t publish it here? Perhaps I will write something on that in the near future. (I haven’t kept my promise to find data on this yet, it’s harder than I thought.)
        Instead, here is a previous attempt at explaining the “backed by debt” theory which I found looking for it.


    • Free Radical
      May 9, 2013 at 4:13 am

      You’re right in saying that if a third party makes a loan which specifies repayment in dollars it will have no effect on the value of the dollar. That’s not the debt that I’m talking about backing the dollar it is the debt to the original issuing bank, the $200 of “IOUs” (bank notes) create a debt from the person who borrows them to the bank.

      You can’t see the effect on the bank’s balance sheet from any one-time static transaction. In order to make the balance sheet of the bank such that their accounts receivable are greater than or equal to their notes outstanding (having started with the notes backed by some real asset) is to have them accumulate retained earnings over time from interest.

  16. May 7, 2013 at 2:29 pm

    If 100 notes existed, and if there were 100 contracts, each specifying that 1 note is to be paid for 1 bushel of wheat, then each note will be worth 1 bushel until the bushels are delivered. After that, nobody will want any notes and they will be worthless.

    The trouble with unbacked money having value is that the issuer of that money gets a free lunch as he issues and spends the money. This free lunch attracts rival money issuers, and the value of the original money falls to zero.

    • Free Radical
      May 9, 2013 at 4:47 am

      I agree with the first paragraph. The second paragraph would be true if people were really just printing money and buying stuff with it. This is why I don’t think that has ever happened. If the government just tried to print money and buy stuff without having an existing debt base to support the money, I suspect nobody would be willing to take it (unless they forced them to, or forced them to do something else with it like pay taxes).

      Note: I think Bitcoin is an exception to my above rule so I could be wrong but I don’t expect it to work in the medium-long run (whatever that means). If it does, we will see if others come in and compete with it.

  17. May 9, 2013 at 4:39 am

    In the example you gave, all 100 contracts were fulfilled without any new contracts being created. If new contracts are being created in the meantime it might take a really long time (or forever) for the value to drop to zero.

    Obviously the balance sheet of the issuing bank has an effect, but do we have reason to believe the effect is immediate? Or that it is the only effect?

    Also, in the example I gave, a bunch of contracts were created and then the issuing bank ceased to exist entirely. If there were still a bunch of unfulfilled (but enforceable) contracts out there the price might even go up if people thought there wouldn’t be enough supply left to fill the need.

  1. June 1, 2015 at 5:06 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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s

%d bloggers like this: