Fractional reserve banking and boom-bust cycles

In his various writings, the famous Austrian economist Murray Rothbard argued that in a free market economy that operates on a gold standard the creation of credit that is not fully backed up by gold (fractional reserve banking) sets in motion the menace of the boom-bust cycle. In his The Case for 100 Percent Gold Dollar Rothbard wrote,

I therefore advocate as the soundest monetary system and the only one fully compatible with the free market and with the absence of force or fraud from any source a 100 percent gold standard. This is the only system compatible with the fullest preservation of the rights of property. It is the only system that assures the end of inflation and, with it, of the business cycle.[1]

Prominent Austrian School of economics economists George Selgin and Lawrence White have contested this view. In his article in The Independent Review, Summer 2000 George Selgin argued that it is not true that fractional reserve banking must always set in motion the menace of the boom-bust cycle.

According to Selgin,

In truth, whether an addition to the money stock will aggravate the business cycle depends entirely on whether or not the addition is warranted by a pre-existing increase in the public’s demand for money balances. If an expansion of the supply of bank money creates an overall excess of money, people will spend the excess. Borrowers’ increased spending will, in other words, not be offset by any corresponding decline in spending by other persons. The resulting stimulus to the overall level of demand for goods, services, and factors of production, together with changes in the pattern of spending prompted by an artificial lowering of interest rates, will have the adverse business-cycle consequences described by the Austrian theory.[2]

However, argues Selgin, no business-cycle will emerge if the increase in the money supply is in response to a previous increase in the demand for money.

Such an expansion, instead of adding to the flow of spending, merely keeps that flow from shrinking, thereby sustaining normal profits for the “average” firm. The expansion therefore serves not to trigger a boom but to avoid a bust. As far as business-cycle consequences are concerned, it makes no difference whether the new money is or is not backed by gold.[3]

Likewise in their joint article Selgin and White wrote,

We deny that an increase in fiduciary media matched by an increased demand to hold fiduciary media is disequilibrating or set in motion the Austrian business cycle.[4]

Note that Selgin and White raise several issues here. First, for them the business cycle emerges only if the increase in the supply of money exceeds the increase in the demand for money.

Second, a bust is set in motion if an increase in the demand for money is not matched by a corresponding increase in the supply of money.

Finally Selgin and White imply that an increase in the supply of money, which is fully backed up by gold, in excess of the demand for money, will also trigger the menace of a boom-bust cycle.

Money out of “thin air” and boom-bust cycle

According to Selgin and White, it would appear that if counterfeit money enters the economy in response to an increase in the demand for money, no harm will be done. In other words, the increase in the supply of money is neutralised, so to speak, by an increase in the demand or the willingness to hold a greater amount of money than before. As a result the counterfeiter’s newly pumped money won’t have any effect on spending and therefore no boom-bust cycle will be set in motion. But does it make sense? What do we mean by demand for money? And how does this demand differ from demand for goods and services?

Now, demand for a good is not a demand for a particular good as such, but a demand for the services that the good offers. For instance, individuals’ demand for food is on account of the fact that food provides the necessary elements that sustain an individual’s life and well being. Demand here means that people want to consume the food in order to secure the necessary elements that sustain life and well being.

Also, the demand for money arises on account of the services that money provides. However, instead of consuming money, people demand money in order to exchange it for goods and services. With the help of money, various goods become more marketable – people can secure more goods than in the barter economy. What enables this is the fact that money is the most marketable commodity.

An increase in the general demand for money, let us say on account of a general increase in the production of goods, doesn’t imply that individuals sit on the money and do nothing with it. As we have seen, the reason an individual has a demand for money is in order to be able to exchange money for other goods and services.

In the process of exercising their demand for money, some individuals lower their demand by exchanging their money for goods and services, whilst other individuals raise their demand for money by exchanging goods and services for money. Note that whilst overall demand did not change, individuals’ demand did change. We will show below that it is individuals’ demand and not the overall demand for money that matters in setting boom bust cycles.

Some holders of money may lend the money to some other individuals in return for an IOU. By accepting the IOU, the lenders are relinquishing their claims on final consumer goods and services for the duration of the loan to borrowers. The borrowers can now exchange the money for goods and services they require. (Note that the existence of banks helps to match between lenders and borrowers).

Now let us assume that for some reason some individual’s demand for money has risen. One way to accommodate this demand is for banks to find willing lenders of money. In short, with the help of the mediation of banks, willing lenders can transfer their gold money to borrowers. Obviously such a transaction is not harmful to any one.

Another way to accommodate the demand is that instead of finding willing lenders, the bank can create fictitious money – money unbacked by gold – and lend it out.

Note that the increase in the supply of newly-created money is given to certain individuals. There must always be a first recipient of the money freshly created by the banks.

This money, which was created out of “thin air”, is going to be employed in an exchange for goods and services, i.e. it will facilitate an exchange of nothing for something. The exchange of nothing for something amounts to the diversion of real wealth from wealth-generating to non-wealth-generating activities, which masquerades as economic prosperity. In the process, genuine wealth generators are left with fewer resources at their disposal, which in turn weakens the wealth generators’ ability to grow the economy.

Once banks curtail their supply of credit out of “thin air”, this slows down the process of an exchange of nothing for something. This in turn undermines the existence of various false activities that sprang up on the back of the previous expansion in credit out of “thin” air, and an economic bust emerges.

We can thus conclude that what sets in motion the boom-bust cycle is the expansion of credit out of “thin air” regardless of the state of the general demand for money. Again, irrespective of whether the total demand for money is rising or falling, what matters is that individuals employ money in their transactions. As we have seen, once money out of ‘thin air’ is introduced into the process of exchange, this lays the foundation for the boom bust cycle.

Contrary to Selgin and White, we can further infer that it is not the failure to accommodate the increase in general demand for money that causes an economic bust, but actually the accommodation by means of money out of “thin air” that does it.

Does an increase in commodity money in relation to demand cause boom-bust cycles?

The introduction of money made it possible for individuals to specialise and engage in trade on a much wider scale than the barter economy would have permitted.

In the early stages of the emergence of money it was an ordinary commodity that people demanded because it contributed some tangible benefits to their life and well being. In other words, people already attached some importance to this commodity. In addition to offering benefits pertinent to this commodity, people also discovered that this commodity, let us call it X, had some other features that made it more marketable than other commodities. For instance, commodity X is durable and it is also portable. The various producers of perishable goods found that it was to their benefit to exchange their produce for commodity X and then use commodity X in exchange for other goods.

Would an increase in the supply of X, in response to an increase in the demand for X, undermine the process of real wealth formation? The answer is no. Since X is a commodity it implies that individuals attach importance to it on account of the benefits it offers. So the fact that producers of this commodity derive a much greater benefit than otherwise on account of the fact that X is also demanded as a medium of exchange is no different from any other commodity which for some reason suddenly experiences much stronger demand than before.

Now, if all of a sudden the supply of X were to increase sharply in excess of demand, people would find that its purchasing power would fall and this in turn would diminish its marketability. Should this persist, the demand for X as a medium of exchange would decline and people would seek the services of another commodity as a medium of exchange. Once a commodity loses its appeal as the medium of the exchange, it remains in demand for its other attributes. However, all this is not going to set the boom-bust menace in motion.

Now, the introduction of paper money, which is fully redeemable into commodity X, doesn’t alter anything we have said so far. Paper money should be seen as a receipt or a claim on the commodity X. So whenever this certificate is exchanged for goods and services the seller of these goods acquires a claim on X, while the seller of the claim acquires goods and services. Note that in the process of the exchange useful goods have been traded.

This is, however, not so when a bank prints a certificate which is unbacked by X. The bank then lends this unbacked certificate to some individual. What we have here is a claim on money that was created out of “thin air”. Note that in the case of a fully backed certificate an exchange of useful goods takes place, i.e. something useful is exchanged in return for something useful. In the case of unbacked certificate, we have a situation that once this certificate is employed in an exchange it leads to an exchange of nothing for something useful. We have shown above that the exchange of nothing for something is what sets in motion the menace of the boom-bust cycle.

We can therefore conclude that in contrast to money out of “thin air”, a market chosen money can never be harmful to individuals well being – it cannot set in motion the menace of boom bust cycle. An increase in the supply of fully backed money in relation to demand will only lead to a fall in the purchasing power of money. This, however, will not give rise to a misallocation of resources and to the boom-bust cycle. Again, an increase in the excess supply of proper money doesn’t set in motion an exchange of nothing for something. (We still retain here the act of an exchange of some useful goods for some other useful goods). Contrary to Selgin and White, then, as far as the business cycle is concerned of course it matters whether the new money is or is not fully backed up by gold.

Selgin also maintains that fractional reserve banking (the creation of money out of “thin air”) was responsible for the industrialization of developed countries.

According to many scholars, including Adam Smith, the industrialization of the West and of developed countries elsewhere was crucially dependent on funds mobilized by fractional reserve banks. Other nations’ failure to industrialize has to a significant extent been due to their repressive financial legislation, including laws (typically aimed at enhancing central bank profits) that forced banks to maintain needlessly high reserve ratios.[5]

This does not make much sense once it is realized that fractional reserve banking (the creation of money out of “thin air”) is actually instrumental in creating the dilution of real wealth formation and boom-bust cycles. After all, if fractional reserve banking is an important source of wealth formation, surely world poverty should have been eliminated a long time ago.

It seems that Selgin is confusing funding with money. What gives rise to the expansion of real wealth is the expansion in the pool of real savings. It is real savings that funds the production of various capital goods, i.e. tools and machinery. In short, it is real savings that sustain various individuals that are engaged in various stages of production. All that money does in all of this is to provide the facility of the medium of the exchange. It makes it possible for individuals to exchange goods and services.

The services of money are not enhanced on account of its greater supply. If anything, the increase in the supply undermines the services of money. After all when people’s demand for money rises they don’t want more money as such, but rather more purchasing power – it is the increase in the purchasing power of money that makes goods and services more marketable. The increase in the supply of money only prevents an increase in the purchasing power of money from taking place.

According to Mises,

The services money renders are conditioned by the height of its purchasing power. Nobody wants to have in his cash holding a definite number of pieces of money or a definite weight of money; he wants to keep a cash holding of a definite amount of purchasing power.[6]


Also, on a gold standard – contrary to Selgin and White – fractional reserve banking will always set the platform for boom-bust cycles. The main problem in Selgin and White’s analysis is that they look at the demand for money from a macro perspective rather than from the perspective of the individual. In short, Selgin and White’s macro-analysis forces them to ignore the misallocation of resources that unbacked credit expansion produces.

This article was previously published on 19 March 2007 by

[1] Murray N. Rothbard – The Case For A 100 Percent Gold Dollar, Cobden Press 1984.

[2] George Selgin – Should We Let Banks Create Money?, The Independent Review, Summer 2000 p 93-100.

[3] Ibid.

[4] George Selgin and Lawrence White. In Defense of Fiduciary Media; or, We Are Not Devolutionists, We Are Misesians! Review of Austrian Economics 1996, 9:83-107.

[5] George Selgin – Should We Let Banks Create Money?, The Independent Review, Summer 2000 p 93-100.

[6] Ludwig von Mises, Human Action, 3rd rev.ed. (Chicago: Contemporary Books, 1966) p 421.


  • mrg says:

    Too repetitive for my tastes. It feels like proof by repeated assertion.

  • mrg says:

    One issue I’ve never been clear on is why gold miners in a world with gold as money are considered righteous, while counterfeiters (or government money printers) are rightly denounced.

    Frank writes:

    “Would an increase in the supply of X, in response to an increase in the demand for X, undermine the process of real wealth formation? The answer is no. Since X is a commodity it implies that individuals attach importance to it on account of the benefits it offers. So the fact that producers of this commodity derive a much greater benefit than otherwise on account of the fact that X is also demanded as a medium of exchange is no different from any other commodity which for some reason suddenly experiences much stronger demand than before.”

    Suppose the economy has 1000 eggs worth of real goods and services in it. Suppose the industrial value of an ounce of X is 10 eggs, while its monetary value is 100 eggs.

    If a producer of X puts another ounce of it into the economy, the real wealth of the community has increased to 1010 eggs (ignoring costs of production), but the producer ‘magically’, ‘fraudulently’ now also has a claim on 90 eggs by virtue of the difference between the industrial value of X and its monetary value. This claim on 90 eggs has appeared “from thin air”, and when exchanging an ounce of X for 100 eggs, only 10 eggs are genuinely exchanged in a something-for-something fashion; 90 of the eggs are exchanged as something-for-nothing, according to the logic of this article.

    While the industrial value of X is inherent (at a given point in time), the monetary value of X is a function of the scarcity of X. When an extra ounce of X is produced, only the industrial value of that new ounce should be considered a blessing. To the extent that the new ounce reduces the scarcity of X, it undermines its value as currency, robbing those who hold X as money just as surely as the government or a counterfeiter does in a fiat system.

    Where have I gone wrong?

  • Doth not compute.

    Every time a customer buys goods on credit, new “money” is created and when he pays up, that new “money” is destroyed again. Anybody can create money by issuing an IOU and destroy it by redeeming the IOU. Credit is not a bad thing in and of itself.

    These credit bubble only happen with land price bubbles (the 1929 share price bubble in the USA was merely the after shock of a land price bubble which popped a few years earlier), when banks lend money to willing/duped buyers to buy overpriced land, safe in the knowledge that the seller will deposit the purchase cheque straight back in the bank again.

    So in the grander scheme of things, the seller and buyer could circumvent the bank and agree to leave the purchase price outstanding; it’s just that the bank spreads the risk of any individual buyer not repaying the loan.

    So ultimately it’s the land price bubbles you ought to be looking at, not the banking system (although yes, a bit of Glass-Steagall wouldn;t go amiss).

  • chef says:

    There’s no great mystery to all this, really!

    The money supply is our means of financial communication, the banking system records transactions and individuals are either creditors or debtors depending upon recent consumption. If they’re debtors they need to enter the real economy to pay down their debts.

    This would be fine if we were all trading freely, but some of us aren’t. The banking system is used by real estate speculators to bid up the price of housing, as the underlying housing asset -land- is in fixed supply it’s very easy to monopolise it and extract above market payment for it’s use. The game Monopoly demonstrates this principle clearly, the winner wins by purchasing real estate and bankrupting the other players.

    No monetary system imagineable would be able to solve this single handedly, it’s a fiscal failure.

  • Current says:

    I disagree with Shostak about several things in this article. I’m in the process of writing an article that’s focused on the dynamic effects of money creation and destruction which I hope will clear up some of these problems.

    Apart from the more macro-economic issues I disagree with Shostak about the “fraud” aspect. Shostak assumes that there is no form of honest money other than gold. Shostak writes:

    Another way to accommodate the demand is that instead of finding willing lenders, the bank can create fictitious money – money unbacked by gold – and lend it out.

    Why is such money “fictitious”? Shostak hasn’t established that it is. The problem here is that money is whatever market participants accept as money. How can Shostak know what market participants will accept? The Gold standard era isn’t evidence in his favour because in that era almost all notes and fiduciary media were fractionally reserved. Under the historical gold standard a banknote was a debt. The amount on it represented an amount of gold that the bank owed the note holder. That didn’t mean that a bank could create as much money as it wanted, it meant that it was limited by the extent of its assets – not its reserves.

    To draw an analogy…. Sometime in the 80s the CEO of a company that made mainframe computers and mini-computers said something to the effect that a “real computer” sits on the floor not on a desk. He could have gone on to claim that his competitors in the microcomputer industry were unjustly damaging his business. They were providing computers that weren’t “real” which reduced his profits through competition and the increasing cost of inputs. This is all wrong because this CEO can’t say what “real” is. It’s the market that decides what it is.

    The same is true of the market for money. If the market accepts only gold or warehouse bailments for gold as money then I’ll shut up. But, if they don’t (and history strongly indicates the market wouldn’t) then proponents of 100% reserve banking can’t say that fiduciary media isn’t money.

    (As a sidenote, in a previous discussion Chef was concerned about competition in money causing confusion. I assume he was worried about the situation arising as we have today in foreign-exchange between the different fiat monies of the world. Currency competition wouldn’t really be like that. What would happen is that businesses would offer different monies, but they would likely price them in terms everyone could understand. Suppose gold were used as the price – which would be likely. I could have a fractionally-reserved banknote for 50 ounces of gold from one bank and a fully-reserved banknote for 50 ounces of gold from another. The competition would come in two ways. Firstly the security of the money, how likely it was the bank would go bust. Secondly, the services the bank provides to holders of it’s fiduciary media.)

    • Chef says:

      Current, there are huge advantages to modern fiat currencies that Austrians et al consistently overlook. The most obvious one being that they’re universally acceptable, by definition local money would be less ‘moneyish’ than state fiat because there’s a greater risk of it beng rejected.

      A loose example of this is the gift card, stores issue their own and often liquidity becomes a problem (most people would have preferred the cash!), so they end up an the back of the sock drawer. This in inconvenient when you get them once a year but would quickly turn into a nightmare (imo) if we adopted this model on a day to day basis.

      Of course there are ways round this, markets would be set up so the various currencies were interchangeable with one another and shops could price their good in the main currencies, but from the macro perspective what have we really gained from this?

      It just seems like a lot of effort for no reward.

      I don’t understand the multi-currency dream!

      • Chef says:

        Edit to add, some money reformers believe that if the state repealed the legal tenders laws etc they would be free, free to use the currency of their choice. This view is misguided, the buyer never decides on the method of payment, the seller does, so this form of ‘coercion’ will always be present in the modern marketplace.

        Eventually society would settle on one or possibly two currencies and by weight of numbers we’d all have to use those currencies whether we liked in or not.

        The icing on the cake would be when the state deemed those currencies suitable for tax payments, meaning we’d gone full circle.

        • Current says:

          > This view is misguided, the buyer never decides on the method of payment, the seller does

          Really? Where is your evidence for this?

          Take credit cards for example. Not every shop takes them, but many do, and they do so because it’s demanded by customers.

          Also, where is your evidence that society would settle on two currencies? If you read about the historic occurrences of free-banking that never actually happened. There were always many different banknotes issued by different banks. Though banks did standardise the commodity used as base money, normally to gold or silver. That said, this standardisation was often driven by the state to some degree not by the banks. That’s not a great problem in my opinion though, the important thing is that that there is competition in money issuing.

      • Current says:


        A bank need not operate only in one place. The spread of a type of money priced in gold (or silver) can be very large. In the time of the international gold standard in the 19th century gold was acceptable money throughout all the western world and most of the rest of the world too. That no longer occurs today.

        Perhaps you are right and fiat currencies would be accepted over a larger area than those created by banks. Even if that is right why should we accept the problems that come with fiat currencies? Why should we accept that money be centrally-planned by an inevitably incompetent central-planner?

  • joebhed says:

    On the issue of whether so-called ‘counterfeit’ money can serve the real needs of the economy, it seems this question was answered historically by those first “gold-tender” bankers that introduced fractional-reserve banking by issuing gold-receipts in excess of their holdings.
    Worked just fine until the owners wanted their gold back.
    The natural economy wants the right quantity of money required to exchange goods and services between producers and consumers.
    That right quantity will preserve the money’s buying power.

    The problem with fractional reserve banking is that all money is created as debt, requiring interest payments in money that is never created. Dr. Bernd Senf of the Berlin School lays out the problem of compounding debt interest here –

  • Current says:

    To other folks who’ve argued with me here…

    I have a sort of priority list of unorthodox economic theories. I want to write about why I think the 100% reserve Austrians are wrong first. That’s because I think they’re closest to being right. Then I’m going to have a go at the MMTers like “joebhed” and Mark Wadsworth. Then the “liquidity-trap” Keynesians like Krugman and DeLong. Then when I’ve finished with that I’ll get onto the Georgists.

    • Current says:

      I see I just accused Mark Wadsworth of being an MMTer, I’m sorry. The MMTers around here are “Joebhed” and Ralph Musgrave. And of course, the Georgists are Mark Wadsworth and ChefDave.

  • George Selgin says:

    Chef’s assumption that competitively supplied paper currency has to be “local” and therefore of limited circulation is belied by numerous episodes of competitively-issued banknote currency throughout history. Branch banking suffices to allow any banks’ notes to have a national circulation.

    Joebhed seems to imagine that failure is the usual fate of fractional-reserve banks. Here, too, history seems not to count for anything. Indeed, even some of the goldsmith bankers to whom he specifically refers are in fact still in business!

    I don’t know why it is so, but the sad fact is that the world seems to team with people whose opinions about money and banks are both stated with perfect conviction and perfectly ill-informed.

  • Chef says:

    Current, a simple question. If you woke up tomorrow and the amount of business that accepted your money was cut in half would your net have increased or decreased?

    At the moment the public are inconvenienced by having to use one type of currency, I’m sure they’d appreciate having to take 10 or 12 different currencies out them whenever they visited the high street!

    You’re not a wallet manufacturer are you?

    • Chef says:

      *net freedom that’s supposed to read.

      I’m a slave to the edit option.

    • Current says:

      As George Selgin says, it wouldn’t happen that way. Historically it didn’t happen that way. When there were 10 different currencies from different banks they were all denominated the same way and all generally accepted.

      If you can come up with any reason why it would happen this way then we can have a discussion.

      • Chef says:

        Selgin’s point is a non sequitur, even if these news banks operated nationally there’s no guarantee that their currencies would be accepted universally.

        If they were all acceptable as well as totally compatible it would be identical to the system we have now with bank credit. Technically they’re private IOUs but due to their homogenous nature the public accept them as cash.

  • George Selgin says:

    “If they were all acceptable as well as totally compatible it would be identical to the system we have now with bank credit. Technically they’re private IOUs but due to their homogenous nature the public accept them as cash.”

    Well, it would be identical except to the extent that it isn’t! For one thing, it would no longer force the public to rely on a monopoly for its circulating currency. That’s currency competition, no?

    • Chef says:

      Well, it would be identical except to the extent that it isn’t! For one thing, it would no longer force the public to rely on a monopoly for its circulating currency. That’s currency competition, no?

      Except that the public wouldn’t get to choose the currency they use, by dint of their purchasing power Tesco’s would, or Philip Green. So once again we’d have it decided for us as you’d be ‘forced’ to use currency that everyone recognised and were happy to use. Don’t forget that ideologically most people have no affinity with ‘currency competition’ so instead of trying to throw a spanner in the works would use whatever was convenient.

      • Current says:

        You could argue that about any sort of good or service. Logically that argument leads to Marxism.

        If you’re not a Marxist then why do you think any other sort of good is different?

  • Gary says:

    The problem lies in inflation, the money supply grows faster than GDP. The boom is monetary inflation and the fractional reserve system has inflation built into it.New money is always created when a loan is made in this system, and since not all loans will perform and grow GDP, money will grow faster than GDP. This is mathematically certain.Some loans will be repaid and shrink the money supply but more loans will not be repaid at the rate that new loans are created. This is especially true when the loans are for the purchase of existing homes.Nothing is contributed to GDP, but the money supply has grown. What should happen is the loan should be made 100% out of reserves and AFTER the loan performs the principal and interest should be repaid out of profits. Demand for money is increased by the resulting growth of GDP, all else equal, and the money supply can be increased AFTER the the growth in GDP drives the demand for new money, not vice versa. If the loan fails then the lender loses the principal plus interest, while the borrower loses his collateral and access to future loans. No new money has been created, no boom, no bust.

Comments are closed.