Money demand and banking – Some challenges for the “Free Bankers”

Mises in his library

Ludwig von Mises; photo by

Within the Austrian School of Economics there has long been disagreement and therefore occasionally fierce debate about the nature and consequences of fractional-reserve banking, from here on called simply FRB. FRB denotes the practice by banks of issuing, as part of their lending activities, claims against themselves, either in the form of banknotes or demand deposits (fiduciary media), that are instantly redeemable in money proper (such as gold or state fiat money, depending on the prevailing monetary system) but that are not fully backed by money proper. To the extent that the public accepts these claims and uses them side by side with money proper, gold or state fiat money, as has been the case throughout most of banking history, the banks add to the supply of what the public uses as money in the wider sense.

Very broadly speaking, and at the risk of oversimplifying things, we can identify two camps. There is the 100-percent reserve group, which considers FRB either outright fraud or at least some kind of scam, and tends to advocate its ban. As an outright ban is difficult for an otherwise libertarian group of intellectuals to advocate – who would ban it if there were no state? – certain ideas have taken hold among members of this group. There is the notion that without state support – which, at present, is everywhere substantial – the public would not participate in it, and therefore it would not exist, or that it constitutes a fundamental violation of property rights, and that it would thus be in conflict with libertarian law in a free society. This position is most strongly associated with Murray Rothbard, and has, to various degrees and with different shadings, been advocated by Hans-Hermann Hoppe, Jesus Huerto de Soto, and Jörg Guido Hülsmann.

The opposing view within the Austrian tradition is mainly associated with George Selgin and Larry White, although there are other notable members of this group, such as Steve Horwitz. This camp has assumed the label “free bankers” and it defends FRB against accusations of fraud and misrepresentation, maintains that FRB is a normal feature of a free society, and that no property rights violations occur in the normal conduct of it. But this group takes the defense of banking practices further, as it also maintains that FRB is not a disruptive influence on the economy, a position that may put the free bankers in conflict with the Austrian Business Cycle Theory, although the free bankers deny this. This point is different from saying that FRB is not fraudulent or suspect. We should always consider the possibility that otherwise perfectly legitimate activities could still be the cause of economic imbalances, even in a free market. If we did find that to be the case, we might still not follow from this that state intervention or bans are justified.

But the free bankers go even further than this. Not only is FRB not problematic, either on grounds of property rights nor on economic stability, FRB is even beneficial as it tends to maintain what the free bankers call short term monetary equilibrium, that is, through FRB the banks tend to adjust the supply of money (by issuing or withdrawing deposit money on the margin) in response to discretionary changes in money demand in such a way that the disruptions would not occur that otherwise seem unavoidable under inelastic forms of money when changes in money demand would have to be absorbed by changes in nominal prices. FRB is thus not just legitimate, it is highly beneficial.

Purpose of this essay

As I said before, this is an old debate. Why should we reheat it? – Before I answer this question, I should briefly state my position: I am not fully in agreement with either camp. I do believe that the free bankers’ defense of FRB is largely successful but that their claims as to it being entirely innocuous and certainly their claims as to its efficiency in flexibly meeting changes in money demand are overstated. In my view, their attempts to support these claims fail.

FRB is, in principle and usually, neither fraud nor a scam, and the question to what extent the depositing public fully grasps how FRB works is not even material in settling this issue. In their 1996-paper ‘In defense of fiduciary media’, Selgin and White argue that the type of money that FRB brings into circulation has to be distinguished from fiat money; they explain that FRB is not fraudulent and that it does not necessarily involve a violation of property rights; third party effects, that is any potentially adverse effects that FRB may have on those who do not participate in it, are not materially different from adverse effects that may emanate from other legitimate market activity, and thus provide no reason for banning FRB; furthermore, Selgin and White claim that FRB is popular and that it would occur in a free market. I agree with all these points. There is no basis for banning FRB, so it should not be banned. This position is, in my view, correct, and it also happens to be obviously libertarian. I may add that I believe it is also almost impossible to ban FRB, or something like FRB, completely. We could ban FRB as practiced by banks today but in a developed financial system it is still likely that other market participants may from time to time succeed in bringing highly liquid near-money instruments into circulation, and that may cause all the problems that the 100-percent-reserve crowd associates with traditional FRB. The question is now the following: do these problems with FRB exist? The free bankers say no. FRB, in a free market, is not only not a source of instability, it is a source of stability as it manages to satisfy changes in money demand smoothly. These positive claims as to the power of FRB are the topic of this essay. I do not believe that these claims hold up to scrutiny.

Why is this relevant?

At first it does not appear to be relevant. Selgin and White declare in their 1996 paper, and I assume their position on this has not changed, that they are opposed to state fiat money and central banking. This sounds similar to the conclusions that I develop in my book, Paper Money Collapse. I advocate the strict separation of money and state. No central bank and no state fiat money. I think it is extremely likely that an entirely uninhibited free market in money and banking would again chose some kind of inflexible commodity – a natural commodity with a long tradition as a medium of exchange, such as gold, or maybe a new, man-made but scarce commodity, such as the cryptographic commodity Bitcoin, or something similar – as the basis for the financial system, and even if the market were to continue with the established denominations of dollars, yen, and so forth, as the public is, for now at least, still comfortable using them, somehow link the issuance of these monetary units again to something inelastic that was not under anybody’s discretionary control. In any case, if we assume that some type of ‘market-gold-standard’ would again resurface, it is very clear that under such purely market-driven, voluntary arrangements and with essentially hard money at its core, any FRB activity would be strictly limited. FRB-practicing banks would not have lender-of-last resort central banks watching their backs. There would be no limitless well of new bank reserves to bail out overstretched banks and to restart new credit cycles whenever the old ones have run their course. There would be no state-administered and tax-payer-guaranteed deposit insurance, or any other arrangement by which the cost of failure in banking could be socialized. Lowering reserve ratios and issuing additional fiduciary media (substitute money, i.e. deposit money) would be legal (the state would abstain from any involvement in monetary affairs, including the banning of any such activities) but it would come with considerable business risk, as it should be. – Would there still be FRB? Certainly. And in my view, the remaining FRB activity, adding as it does to the elasticity of the money supply at the margin and thus potentially distorting interest rate signals, is going to lead to capital misallocations to some degree, and thus initiate the occasional business cycle. That, in my view, is the price we have to pay for having a developed monetary economy and entire freedom in money and banking with all the undeniable advantages such a system brings. Importantly, I believe that these costs are unavoidable. But they are minor due to the absence of FRB-boosting state policy. – No, an entirely free market would not fulfil any dreams of uninterrupted bliss or realize the macroeconomist’s fantasy of everlasting ‘equilibrium’, both notions that Ludwig von Mises frequently rejected and ridiculed, but it would for sure be considerably better, and much more stable, than anything our present elastic monetary system can produce.

In Paper Money Collapse, I argue that inelasticity of supply is a virtue in money. That is why gold is such an excellent monetary asset. Complete inelasticity is unattainable in the real world but something like a proper gold standard is close enough. But for the ‘free bankers’ the remaining elasticity under restricted FRB (restricted by a stable commodity base) would be a boon. It would further stabilize the economy and establish…equilibrium. In my view, these claims are unsupported. But, you may say, why should we argue about the specific features of the post-fiat-money world if we are in agreement that such a post-fiat money world is in any case preferable to the present one?

The reason is simply this: How do we evaluate current policies? On this question I thought that most Austrians, as advocates of gold or something similar, and as critics of fiat money, would still be in broad agreement. But to my initial shock and my lasting amazement I found that some Austrian free bankers frequently cannot bring themselves to reject ‘quantitative easing’ and other heavy-handed central bank intervention on principle, and that they are able to embrace monetarist policy proposals, such as nominal GDP targeting by central banks, as a kind of second-best-solution that will do for as long as our first choice of separation of money and state is not realized. I believe these positions to stand in fundamental conflict with key tenets of the Austrian School of Economics and, apart from that and more importantly, to be simply unjustifiable. I think they are misguided. But it seems to me that the occasional support for them among free bankers originates in certain expectations as to what the equilibrating forces of ‘free banking’ would bring about in a free market in terms of a stable nominal GDP, and the free bankers can thus advocate certain forms of central bank activism if these are bound to generate these same outcomes. Therefore, in order to refute the idea of nominal GDP targeting we have to show that the free bankers’ expectations as to ‘monetary equilibrium’ under free banking lack a convincing analytical foundation. In this essay I want to pose some challenges for the free bankers. In a later article I hope to address NGDP-targeting as such.

Money does not need a producer

Among all goods money has a special place. It is the most liquid good and the only one that is demanded only for its exchange value, that is, its price in other goods and services. Anybody who has demand for money has demand for real money balances, that is, for effective purchasing power in the form of money. Nobody has demand for a specific quantity of the monetary asset per se, like a certain number of paper notes or a particular quantity of gold, but always for the specific purchasing power that these monetary assets convey.

In contrast to all other goods and services, changes in money demand can in theory be met by either producing additional quantities or by withdrawing and eliminating existing quantities of the monetary asset (changing the physical quantity of money), or by allowing the price of money, money’s exchange value, to change in response to the buying and selling of money versus non-money goods by the public (changing money’s purchasing power). Furthermore, it can be argued, as I do in Paper Money Collapse, that the superior market process for bringing demand for and supply of money in balance is the latter, i.e. the market-driven adjustment of nominal prices in response to the public’s buying and selling of money for non-money goods according to money demand. Why? – Well, mainly because the process of adjusting the physical quantity of money does not work. 1) We lack a procedure by which we can detect changes in money demand before they have begun to affect prices, and if prices are already beginning to change then these price changes already constitute the very process that satisfies the new money demand. This makes changing the quantity of money superfluous. 2) We lack a procedure by which we can expand and contract the supply of money without affecting the supply of credit and without changing interest rates. This makes changing the quantity of money dangerous. Money demand and loan demand are different things. Our modern fiat money systems are, in any case, not really designed for occasionally reducing the supply of money but for a continuous expansion of the money supply. As the Austrian Business Cycle Theory explains, expanding the supply of money by expanding bank credit must distort interest rates (artificially depress them) and lead to mismatches between voluntary saving and investment and thus to capital misallocations.

To this analysis the free bankers appear to voice a few objections. Before we look at the differences, however, let’s first stress an important agreement: The free bankers agree that nominal prices can do the adjusting and bring demand for and supply of money in balance. But they introduce an important condition: in the long run. In the short run, they argue, the process is not quite as smooth as many hard-money Austrians portray it to be.

Selgin and White (‘Defence’, 1996):

 “In the long run, nominal prices will adjust to equate supply and demand for money balances, whatever the nominal quantity of money. It does not follow, however, that each and every change in the supply of or demand for money will lead at once to a new long-run equilibrium, because the required price adjustments take time. They take time because not all agents are instantly and perfectly aware of changes in the money stock or money demand, and because some prices are costly to adjust and therefore “sticky.” It follows that, in the short run (empirically, think “for a number of months”), less than fully anticipated changes to the supply of or demand for money can give rise to monetary disequilibrium.”

 Thus, the first objection of the free bankers is that the account of the hard-money Austrians about the smooth adjustment of prices in response to changes in money demand is a bit superficial and slick. In the real world, not all prices will respond so quickly. Not all goods and services are being priced and re-priced in a continuous auction process, and when the public reduces money-outlays at the margin in an attempt to increase money-holdings, not every producer of goods and services will quickly adjust the price tags of his ware.

I do think some of this criticism is valid, and I am not excluding myself from it. My own account of the process of adjustment of money’s purchasing power sometimes runs the risk of glossing over the real-life frictions involved. However, to my defense, I acknowledged some of these problems in Paper Money Collapse, although I do not treat them extensively. See page 144-145:

“In the absence of a flexible money supply, sudden changes in money demand will have to be fully absorbed by changes’ in money’s purchasing power. One could argue that this, too, has the potential to disrupt the otherwise smooth operation of the economy. Indeed, as we have seen, this phenomenon will also affect the prices of different goods differently. [This refers to the fact that when, for example, people try to raise their money holdings, they will reduce money-outlays on non-money goods or sell non-money goods for money, but they won’t cut every single expenditure item by an equal amount, or liquidate a tiny portion of each of their assets but will always cut the expenditure or sell the asset that is lowest on their present value scale. Downward pressure on prices from rising money demand will thus not be the same for all prices.]…A change in the demand for money will change overall prices but also relative prices and therefore the relative position of economic actors and the allocation of resources in the economy. All of this is true but it must lead to a different question: Is any of this avoidable….?”

Is ‘monetary disequilibrium’ a unique phenomenon?

The free bankers are correct to point to these problems but it is also true that every change in the preferences of economic agents leads to similar problems. If consumer tastes change and money-flows are being redirected from certain products to certain other products, this, too, means that nominal spending on some items is being reduced. Profitability will decline in some parts of the economy and increase in others. This, too, will ultimate redirect resources and change the economy but all of these processes “take time because not all agents are instantly and perfectly aware …” of what is going on, and also for other reasons, including the stickiness of some prices. I think agents are never “instantly and perfectly aware” of anything, and that the slickness of economic models is never matched by reality. Accordingly, the real world is constantly in disequilibrium, and as economists we can only explain the underlying processes that tend towards equilibrium without ever reaching it. I wonder, however, if the concerns of the free bankers, valid though they are, are not just examples of the frictions that always exist in the real world, in which tastes and preferences change constantly, and change in an instant, but prices, knowledge, and resource use always move more slowly.

Furthermore, the issue of stickiness of prices should not be overstated. These days many prices do appear rather flexible and tend to adjust rather quickly: not only those of financial assets but also industrial commodities, and even many consumer goods, from used cars to hotel stays to flight tickets to everything on eBay. Discounting in response to a drop in nominal spending is the first of line of defense for almost every entrepreneur, I would guess, and if what the entrepreneur faces is indeed a higher money demand among his clientele, rather than a genuine change in consumption preferences, then sales should stabilize quickly at the lower price.

But, I think, the main point is this: How can the banks do better? What do the free bankers say to my two points above that changing the quantity of money is not really a viable alternative to allowing changes in nominal prices? Let’s address the first point first:

Point 1) We lack a procedure by which we can detect changes in money demand before they have begun to affect prices, and if prices are already beginning to change then these price changes already constitute the very process that satisfies the new money demand. This makes changing the quantity of money superfluous.

How do banks detect a change in money demand – before it has affected prices?

Banks have no facility to create money and money alone (deposit money, fiduciary media). New money is always a byproduct of banks’ lending operations. Banks can only create money by expanding their balance sheets. Thus, they always create an asset (a new loan) at the same time they create a new liability (the demand deposit in which the bank pays out the loan to the borrower, and which is part of the money supply). Therefore, if you suddenly experience a rise in money demand, if you suddenly feel the urge to hold more of your wealth in the form of the most fungible object (money), the bank can’t help you. Of course, you could go to the bank and borrow the money and then keep it in cash. This is a possibility but I think we all agree – and the free bankers seem to agree as well – that this is very unusual, and that it must be rare. Banks meet loan demand, not money demand, and the two are not only different, they are the opposite of one another. Borrowers do not have a high marginal demand for money; quite to the contrary, they have a high marginal demand for goods and services, i.e. non-money items (that is why they are willing in incur interest expense). The loan is in form of money but the borrowers usually spend the money right away on whatever they really desire.

Banks are not in the money-creation business (or only in it by default – no pun intended); they are really in the lending business. The idea that rising money demand would articulate itself as higher loan demand at banks is wrong, and the free bankers do not usually make that mistake. They know (and some of them even stress) that money demand articulates itself in the markets for non-money goods and services (including, but not restricted to, financial assets). People reduce or increase spending in order to establish the desired money holdings.

To the extent that, when people experience a higher money demand, they sell financial assets to banks, the banks do indeed directly experience the heightened money demand, and if the banks increase their FRB activities in response and expand their balance sheets accordingly (the financial assets they buy enter the asset side of the balance sheet – they are the new loans – and the new demand deposits the banks issue to pay for them sit on the liability side of the balance sheet), the quantity of money is indeed being expanded in response to money demand. But to the extent that the public does not sell to FRB-practicing banks or that the public reduces other outlays or sells non-financial assets, the banks are not directly involved as counterparties. How can they still detect a rising money demand?

[As an aside, the free bankers sometimes speak of ‘the public having a higher demand for demand deposits or ‘inside money’ ’, and that the banks should be allowed to ‘accommodate’ this. I think these statements are confusing. Depositing physical cash in a bank, or conversely liquidating demand deposits to increase holdings of physical cash, are transactions between various forms of money. In a functioning FRB system, both forms of money, physical cash and bank-produced deposit money, are almost perfect surrogates. Both are used side by side, and both satisfy the demand for money. That is the precondition for FRB to work. The factors that occasionally determine preferences for a specific form of money are fundamentally different from those that affect the demand for money overall. If the public, for example, reduces demand deposits and accumulates physical cash, i.e. switches from ‘inside money’ to ‘outside money’, this may be because it is concerned about the health of the banks, and this is unrelated to the public’s demand for money, which in this case may be unchanged. As an example, in the recent crisis, the demand for physical cash increased in many countries, relative to the demand for bank deposits. At the same time, overall money demand also probably increased. But importantly, both phenomena are fundamentally different.]

The answer is this: If the public, in an attempt to raise money holdings, reduces money spending, this will slow the velocity of money, and to the banks this will be clearly visible. Money doesn’t change hands as quickly as before, and that includes transaction-ready deposit money at banks. Importantly, the slower velocity of money means a reduced risk of money outflows for each bank, in particular the likelihood of transfers to other banks that are a drain on existing bank reserves. Thus, the banks have now more scope to conduct FRB, that is, to reduce their reserve ratios, lower loan rates and issue more loans, and obviously to produce more deposit money in the process.

In the essay mentioned above, ‘In defence of fiduciary media’, this explanation apears in footnote 29, the emphasis here is different and so is the wording but the essence is the same, in my view. Banks increase FRB in response to a drop in money velocity. A rising money demand articulates itself in a lower velocity and thus a tendency for more FRB:

“But how can the banks manage to expand their demand deposits, if total bank reserves have not changed? The increased demand to hold demand deposits, relative to income [increased money demand, DS], means that fewer checks are written per year per dollar of account balances. The marginal deposit dollar poses less of a threat to a bank’s reserves. Thus a bank can safely increase its ratio of deposits to reserves, increasing the volume of its deposits to the point where the rising liquidity cost plus interest and other costs of the last dollar of deposits again equals the marginal revenue from a dollar of assets.”

 I think this explanation is exceedingly clever and accurate. I do not, because I cannot, object to the logic. But does it help us? I have two observations:

1)   Is it really probable that this process is faster and more efficient than the adjustment of nominal prices? The objection of the free bankers was that the adjustment of nominal prices takes time. But so does this process. The bankers will not be “instantly and perfectly aware” of what is happening anymore than the producers of goods and services. When the public reduces spending in order to preserve money balances the effect will be felt as soon by the producers of whatever the public now spends less money on, as by the bankers who see fewer checks being written. Why would we assume that the bankers respond faster? Sure, prices can be sticky, but does that mean that accelerated FRB will always beat nominal price changes in terms of speed? Will the bankers always expand their loan book faster than the affected producers discount their product? It is not clear to me why this would be the case.

2)   More importantly, the banks will, by definition, give the new deposit money first not to those who have a higher demand for money but to their loan clients who, we just established, have no demand for money but for goods and services, and who will quickly spend the money. From there, the money will circulate and may, finally, reach those who do indeed have a higher demand for money. But there is no escaping the fact that this is a roundabout process. For the very reason that banks can only produce money as a byproduct of their lending business, those who do demand higher money balances can only ever be reached via a detour through other markets, never directly. Bank-produced money has to go through the loan market first, and has to change hands a few times, before it can reach those who originally experienced a high money demand. There is no process as part of which we could ever hear a banker say to any of his customers: You have a higher money demand? Here, have some. – The question is now, what type of frictions or unintended consequences of this procedure of satisfying money demand do we encounter? Are these frictions likely to be smaller or even greater than the frictions inherent in allowing nominal prices to do the adjusting to meet changes in money demand?

 Before we address these frictions a few words on a related topic: The free bankers sometimes seem to imply that unwanted fiduciary media (demand deposits, inside money) would return to the banks. This is not correct, or rather, it would only be correct if people wanted to exchange the demand deposit for physical cash but this is a transaction that is, as we have seen, unrelated to money demand. Claims against any specific bank may be unwanted, or demand deposits may be wanted less than physical cash, but this is unrelated to overall money demand. If deposit money is seen as a viable money good, and this is the precondition for FRB to work, any excess holding of money, whether inside money or outside money, whether cash or demand deposit, will not be returned to a bank and exchanged but will be spent! If banks increase their FRB activities and bring new fiduciary media into circulation, this money will circulate until it reaches somebody with genuine money demand. Often – when money demand has not risen simultaneously – this process involves inflation as a lower purchasing power for each monetary unit is required to get the public to voluntarily hold the new monetary units.

Is money demand a form of desired saving?

According to the free bankers, banks respond to a drop in money velocity as a result of rising money demand by engaging in extra FRB. At lower velocity, the risks inherent in FRB are smaller and this encourages banks to reduce their reserve ratios marginally, create extra loans and produce extra money, i.e. new deposit money that is now satisfying at least some of the new money demand. But what about the extra bank credit that also comes into existence? Hasn’t Mises shown that bank credit expansion is a source of economic instability; that bank credit expansion sets off business cycles? If extra loans at lower interest rates are not the result of additional voluntary saving but simply of money printing, and these loans still encourage extra investment and capital spending, then these additional projects will ultimately lack the real resources, resources that only voluntary saving can free up and redirect towards investment, that are needed to see the projects through to conclusion and to sustain them. Extra bank credit is thus bound to upset the market’s process of coordination between saving and investment – coordination that is directed via market interest rates. Would the extra FRB not start a Misesian business cycle? Would the allegedly faster and smoother process of satisfying changed money demand via FRB, via the adjustment in the nominal quantity of money rather than nominal price changes, not create new instabilities as a result of the artificially lower interest rates and the extra bank credit that are the necessary mirror image of new deposit money?

In Austrian theory, desired savings are a function of time preference. A lower time preference means the public attaches a lower importance to consumption in the near future relative to consumption in the more distant future. The discount rate at which future goods are discounted is lowered and the propensity to save rises, i.e. the willingness to reallocate income from meeting present consumption needs to meeting future consumption needs rises. The extra savings are offered on the loan markets at marginally lower rates. This encourages a marginal increase in investment. The marginally lower rates on the loan market thus accurately reflect the marginally lower time preference of the public. But lower rates as a result of credit expansion and FRB can unhinge this process. That is the core message of the Austrian Business Cycle Theory. How can the free bankers get out of this dilemma?

The free bankers counter this point by claiming that an increased demand for money reflects a lower time preference. Holding more money is a form of saving.

Although in the already quoted “Defence of Fiduciary Media”, Selgin and White at some point state that

“We agree that time preference and money demand are distinct, and that a change in one does not imply a change in the other.”

 They also write, and this is more crucial to the case they are making, I believe,

“The argument for the equilibrating properties of free banking rests in part on recognizing that an increased demand to hold claims on intermediaries, including claims in the form of banknotes and demand deposits, at the expense of holding additional consumer goods, is equivalent to an increase in desired saving.”

 In any case, in the examples they provide later, time preference, desired saving, and money demand always move together.

 While I agree that accumulating money balances can be a form of saving (I say that much in Paper Money Collapse), it does not have to be the case, and I think it is more helpful to disentangle saving, consumption and money demand. Holding money is non-consuming, as Selgin and White point out, but it is equally non-investing.

 If I sell my laptop on e-Bay so I have more readily spendable money (demand deposits) in my bank account so that I can take advantage of any unforeseen spending opportunities during my holiday in Greece, would we say that my time preference has declined, and that this is an act of saving? This is a switch from a consumption good to money, and Selgin and White would label this an act of saving, at least as I understand them. But the laptop would have delivered its use-value to me over a long period of time. Now I hold instantly spendable demand deposits instead. Has my time preference really dropped?

 Here is a different example, one where we encounter a switch from investment goods to money, an example that Selgin and White put forward in their paper and where they argue that in such an operation total desired saving remains unchanged. Time preference remains the same. In the example given, the public sells bonds and accumulates cash or demand deposits instead. Both, money and bonds are non-consumption goods and thus saving-instruments in the Selgin and White definition. According to their theory, the banks would now acquire the bonds and issue deposit money against them. By doing this (increased FRB activity), the banks satisfy the demand for more money and keep interest rates from rising – which is appropriate as overall desired savings have not changed and time preference is still the same. – However, has the public’s time preference really not changed? Rather than holding a less liquid, long-term debt instrument the public now holds the most fungible asset (money). Is it fair to say, that when people liquidate their bond portfolios that their time preference remains unchanged? – Maybe the public does this precisely for the reason that time preference has increased. The public may spend the money soon on consumption goods, or the public considers market interest rates too low and as no longer representative of the public’s time preference, and a drop in bond prices (rise in yields) is thus warranted to reflect this, and should not be cancelled out by the banks’ accelerated FRB.

 The short run versus the long run

Furthermore, I suspect that there is an inconsistency in claiming that, in the long run, nominal price changes do bring the demand for and supply of money in line and then to argue that in the short run, money demand is best – and automatically -met by quantitative changes in the supply of money via FRB. The long run is evidently only a string of short runs, and if changes in money demand have been satisfied in the short run via FRB, how can these changes then still exercise up- or downward pressure on nominal prices in the long run?


The free bankers are correct to point to real-life frictions in the process of satisfying a changed money demand via an adjustment of nominal prices. The process is neither smooth nor instant, but then almost no market process is in reality. Their explanation that a rise in money demand will lead to a drop in money velocity and that this will, on the margin and under normal conditions, encourage additional FRB and thus an expansion of bank-produced money also strikes me as correct. Yet, the free bankers fail, in my view, to show convincingly why this process would be faster and smoother than the adjustment of nominal prices, and in particular, why the extra bank credit that also comes into existence through FRB would not generate the problems that the Austrian School under Mises has explained extensively.

 If only a subset of the population, rather than the entire public, experiences a higher money demand – and this must be the more likely scenario by far – and this subset than reduces nominal spending on those goods and services that are relevant to this group, and if this then leads to a marginal drop in the prices of these goods and services, the extra demand of this group for real money balances has been met with potentially fairly limited frictions and side-effects, I would argue. By comparison, FRB can never meet money demand of any group directly. Banks always have to inject the new money into the economy via the loan market, that is, at a point where money demand is low and demand for non-money goods is high. Money demand will always be met in a roundabout way. Furthermore, the lowering of interest rates through the additional FRB activity is only unproblematic if the additional demand for real money balances is identical with desired saving and reflects a reduce time preference. These are rather heroic assumptions indeed.

 Ludwig von Mises – The real free banker

The 100-percent-reserve Austrians have stuck – correctly in my view – with one of the most important insights of Austrian monetary theory as developed by the school’s most distinguished 20th century representative, Ludwig von Mises, namely the destabilizing force of credit expansion. Unfortunately, the 100-percent-reserve Austrians have taken the critique of banking too far. Claims of misrepresentation, deception, and fraud as being constituting elements of FRB go too far and remain ultimately unsupported. The self-styled ‘free bankers’ are correct to reject these claims but they are taking their defense of FRB too far as well. By claiming that FRB could smoothly and quickly satisfy any changes in money demand they assign equilibrating properties to FRB that are ultimately unsupportable. In the process, they risk ignoring some of the most relevant Misesian insights. In particular the free bankers, it seems to me, tend to ignore that in an established FRB system, bank-produced fiduciary media (such as demand deposits) will be seen as near-perfect surrogates for money proper (such as state fiat money or gold). In such an environment the banks can (within limits) expand FRB and thus create more fiduciary media regardless of present money demand. Unwanted money (deposit money) then leads to a rise in money velocity and an upward pressure on nominal prices – it does not lead to the public exchanging deposit money for physical cash, as that would be just a switch from one form of money to another. Therefore, the unwanted bank-produced money – that entered the economy via the bank loan market – does not return to the banks. In my view, the free bankers ignore some of the dangers in FRB and overstate its equilibrating powers.

Both camps refer to Mises as an authority, albeit the ‘free bankers’ generally less so. Selgin and White, in their 1996 paper, quote Mises as a champion of free banking. I do, however, believe that the quote, taken from Human Action, has to be read in the context of Mises’ life-long and unwavering commitment to a proper gold standard. Here is the quote:

 “Free banking is the only method for the prevention of the dangers inherent in credit expansion. It would, it is true, not hinder a slow credit expansion, kept within very narrow limits, on the part of cautious banks which provide the public with all the information required about their financial status. But under free banking it would have been impossible for credit expansion with all its inevitable consequences to have developed into a regular – one is tempted to say normal – feature of the economic system. Only free banking would have rendered the market economy secure against crises and depressions.”

Crises and depressions, in Misesian theory, do not come about because of short-term mismatches between money demand and money supply, or frictions in the adjustment of nominal prices, but because of credit expansion. In order to appreciate Mises’s concerns over credit expansion, one does not have to consider bankers fraudsters (or ‘banksters’), and I can see no evidence in Mises’ writing that he saw bankers that way. But in order to agree with him that banks should be as free as all other enterprises – which, importantly, includes the freedom to fail – you do not have to assign them mystical equilibrating powers, either.

Mises’ conclusions were consistent and his recommendations practical: Introduce inelastic, inflexible, apolitical money as the basis of the financial system, a hard monetary core, such as in a proper gold standard, and then allow banks the same freedom, under the same laws of corporation, that all other businesses enjoy – no special bans and no special privileges, such as ‘lenders of last resort’ or tax-payer-backed deposit insurance – and you can allow the market to operate. I believe that this should be the policy proposal under which all Austrians can and should unite.

Any deviation from the core Misesian message also occasionally gets ‘Austrians’ into some strange political company. With their damnation of FRB and allegations of fraud, the 100-percent-reserve Austrians seem at times to play into the hands of populist anti-bank fractions that have recently grown in influence since the financial crisis started, and to inadvertently be associated with the statist proposals of organizations such as the UK’s Positive Money or IMF economists Benes and Kumhof, all of whom consider money-creation by private banks – FRB- as the root of all evil and propose full control over the monetary sphere by the state – a proposal that could not be further from Mises’ ideals. On the other side, the free bankers are in such awe of the assumed equilibrating powers of FRB in a free market that they confidently predict a stable (or at least reasonably stable) nominal GDP – and if we do not have free banking and a free market yet, why not have today’s central banks target nominal GDP to get a similar result under today’s statist monetary infrastructure? Bizarrely, and completely indefensibly, in my view, these Austrians end up joining forces with aggregate-demand-managing Keynesians or money-supply-managing monetarists. This is not only in fundamental conflict with many tenets of the Misesian framework – it is simply misguided, even under considerations of monetary realpolitik, i.e. of what is politically practicable presently but better than the present system.

Banks should be free but can only ever be so within a proper capitalist monetary system, and that is a system with a market-chosen monetary commodity at its core, and most certainly a hard and inelastic one. No new ‘target’ for central bank policy can ever achieve results that mirror the outcome of a properly functioning monetary system and a free banking market. We do not have a gold standard and free banking at present, and under these conditions I would suggest that a central bank that imitates a gold standard as closely as possible – i.e. one that ultimately keeps the monetary base fairly stable – would be, under the circumstances, the second best’, or least worst, solution. But a full treatment of the NGDP-targeting proposal will have to wait for another blog.

In the meantime, the debasement of paper money continues.



Print Friendly
Share on LinkedInShare on TwitterSubmit to StumbleUpon to redditShare via email
Markets reject ‘forward guidance’ – for good reason
Forward Guidance? – Nonsense! Central bankers have no choice.


  1. George Selgin says

    “[t]he free bankers fail, in my view, to show convincingly why this process would be faster and smoother than the adjustment of nominal prices, and in particular, why the extra bank credit that also comes into existence through FRB would not generate the problems that the Austrian School under Mises has explained extensively.” What I find striking about this claim, Detlev, is that it seems to refer works by myself and Larry other than those specifically devoted to addressing the very points at issue. In particular, you don’t refer to my Theory of Free Banking, which (as it’s subtitle indicates) is specifically about “Money Supply under Competitive Note Issue,” and which explains at some length how the clearing system encourages free banks to expand their balances sheets rapidly in response to any excess demand for real balances. That book also contains a section (“Transfer Credit, Created Credit, and Forced Savings”) specifically explaining why such monetary expansion does not have the cyclical consequences of credit expansion that isn’t warranted by prior growth in real money demand. (The same section also points out certain inconsistencies in Mises’ position.) I further elaborate upon the behavior of money supply under free banking in papers on “Free Banking and Monetary Control” (Economic Journal 1994) and “In-Concert Overexpansion and the Precautionary Demand for Bank Reserves” (JMCB 2001). My and Larry’s paper “Against Fiduciary Media,” occupied as it was with the more broader questions concerning the legitimacy and consequences of fractional-reserve banking, couldn’t repeat all over again the details offered in these previous studies, and so refers readers to them. It is therefore to these expositions that any claim to the effect that free bankers have failed to adequately defend their claims concerning the rapidity and stabilizing nature of nominal money stock changes under free banking must address themselves.

    On a more minor point, your characterization of Mises’ views is inconsistent with abundant evidence of his having himself often–not to say consistently–recognized the advantages of (free banking based) fiduciary money issues. (One can quote Mises selectively to claim otherwise, but doing so proves nothing.) Moreover the implicit appeal to Mises’ authority in your last few paragraphs, though well-calculated to win the approval of readers of what is, after all, the “Mises Daily,” doesn’t advance the economics debate, which ought to concern, not whether such-and-such a view is or isn’t in agreement with what Mises had to say, but whether it is or is not both logical and consistent with available evidence.

  2. George Selgin says

    “I would suggest that a central bank that imitates a gold standard as closely as possible – i.e. one that ultimately keeps the monetary base fairly stable – would be, under the circumstances, the second best’, or least worst, solution.” This begs the question, for in my own works cited above and elsewhere, I also argue the merits of having a “stable” monetary base; indeed, in TFB I endorse the idea of absolutely freezing the monetary base. But I then proceed to point out how, with such a frozen base, free banking tends to lead to a stable level of NGDP. The same follows if, as under a gold standard, the monetary base tends to grow slowly and steadily. If this view is correct, a central bank that “imitates” a (free-banking supplemented) “gold standard as closely as possible” is in fact one that seeks to stabilize NGDP, and _not_ one that would allow the sort of dramatic collapse of NGDP such as occurred in late 2008.

    Despite repeated suggestions to the contrary from the Mises Institute crowd, I’ve never been a champion of the Fed’s various attempts to rescue the U.S. economy from the recession to which it was driven partly by the Fed’s own previous policy mistakes. But neither have I ever resisted pointing to the collapse of spending in 2008-9 as an instance of a major blunder of monetary central planning. To argue against the view, taken by myself and some other free bankers, that the Fed ought to have acted to avoid the collapse in question, is not to take a firmer stand against monetary central planning than we ourselves take. It is merely to apologize for what was in fact a particularly flawed monetary central plan!

  3. George Selgin says

    “The free bankers counter this point by claiming that an increased demand for money reflects a lower time preference.” You make this claim, yet go on to cite two passages by Larry and me in which we specifically deny that changes in the demand for bank money necessarily involve changes in time preference! The second of the two passages, which treats the case in which the two things change together, differs from the first in that it specifically says that a change in time preference is involved only if the increase in demand for bank money comes at the expense of demand for consumer goods. Of course (as we also say), if people switch from holding bonds to holding bank money, a change in time preference is not necessarily in play. We then give an example of such a case. But by doing so we never meant to suggest that there cannot be a switch from bonds to money that also involves a change in time preference! So, if we offer a case in which changes in money demand involve changes in time preference, we are accused of asserting that changes in money demand are _always_ associated with changes in time preference; whereas if we go out of our way to explain by way of a counterexample that this was not our position, we are accused of overlooking the possibility of a change in time preference connected to a change in money demand! Of course there is no end to such attempts to play “gotcha” by misrepresenting the purpose of specific illustrations!

    • says

      Dear professor Selgin,

      the main problem with your position I see in assuming that the desire to hold inside money equals bank credit. Or narrow the issue down, that bank credit is automatically a part of the money supply. But this is an empirical issue. It’s inductive reasoning, not deductive. In order to make a praxeological argument, you have to prove that only bank’s lending activities affect the decision of market participants to hold inside money. And this is contradicted by practically all economic schools I’m familiar with. Even you yourself, in The Theory of Free Banking, write that transaction costs affect the choice whether to hold inside or outside money. Which again is an empirical issue (and recognised as such by Lawrence White).

      The chart you use to show how loans and money supply equilibrate is thus wrong, because it only has two axes.

      I submit a contrarian deductive approach. Bank credit acts as a part of the money supply when it decreases transaction costs. Whether people want to hold bank credit in their portfolio and whether the view it as money are two different issues, and need need to be answered separately. They might hold it in their portfolio in order to earn interest, but they only use it as a medium of exchange when it decreases transaction costs. Even though you appear to understand this, you never connect the dots together.

  4. says

    I don’t see how FRB could be non-fraudulent. Lending out more money than one is in possession of is fraudulent on its face. The common rejoinder to this is that if the borrower is aware that the loan is not covered 100%, he is simply taking a calculated risk. But then I think the fraud is passed over to a third party. Every increase in the money supply means that some people receive the new money before prices have risen, while others receive it only after prices have risen. Those latter are defrauded. And this has to be true of FRB under free banking as well. Is there a flaw in my reasoning here? Or is it the case that the inflationary effects of FRB are so small as to be negligible? And that the beneficial effects far outweigh this small inflationary effect?

      • Hugh Pendleton says

        Sorry, not an economist so your statement that causing inflation in not synonymous with fraud confuses me.

        What do you call it when you steal from people with out them being aware that you did it? In financial accounting circles that is referred to as fraud. Of course economists call it seigniorage when the government does it by printing money. A rose by any other name…

  5. says

    “The great problem of fiduciary media is that they set up money and debt like a house of cards or row of dominoes that any breeze can knock over.” – George Reisman, “Capitalism: A Treatise on Economics”, p. 513.

    Were there no such breezes knocking down the card house in the “good old days”? There were, and governments stepped in to save the fractional banks:

    “Again and again, when banks did fail, the government stepped in and allowed them to suspend payment in specie, in flagrant violation of their agreement to pay their depositors specie on demand. This prevented the wiping out of fractional-reserve banks and enabled such banks to return to issue still more fiduciary media.” – Ibid., p. 515.

    I think it follows from this that if we had fully free banking – i.e. no government interference and no governments saving the banks by suspension of specie payment – banks that are reckless with issuing fractional loans would quickly disappear, and the banks that are most cautious about issuing fractional loans would stand a much better chance to survive.

  6. says

    As a lay person I too have difficulty in the concept of an institution lending money it does not have by way of a “money substitute”, but nevertheless the practice took hold not by government decree but through the acceptence of all parties involved at the time. The problems really began did they not when the central banks were invented to act as “liquidity backstops”, “lenders of last resort” or whatever quaint phrase they want to use to describe themselves. It seems to me that it is this that has created the moral hazard. We are now at the point where the FED are bailing out trading losses! I believe more and more people are now starting to question the role of the central banks in the world.

  7. says

    I find the first several paragraphs to succinctly describe why free-banking is most consistent with an Austrian and Libertarian view (and why FRB, at least as practiced today, isn’t fraud).

    Beyond this, I offer a couple of comments:

    “[Free bankers claim] FRB, in a free market, is not only not a source of instability, it is a source of stability as it manages to satisfy changes in money demand smoothly.”

    I will break this into two parts: as to the statement of why free bankers claim it is a source of stability (“as it manages to satisfy changes in money demand smoothly”), I offer no comment. However, to the first part – that FRB in a free market is a source of stability seems undeniable. Or more precisely, free banking (and if FRB develops, so be it) will offer the most stability – even more than a 100% commodity standard.

    Why? The reason is found in the essay:

    “There is no basis for banning FRB, so it should not be banned. This position is, in my view, correct, and it also happens to be obviously libertarian.”

    Any action taken to counter free banking can only happen by the interjection of a non-market force – certainly to have a 100% commodity standard would require interjection of a non-market force.

    How can force in the market produce more stability (to the extent “stability” is achievable) than market actions would otherwise produce? The best way to achieve stability is to allow prices to communicate freely, allow profit and loss to allocate resources to the most efficient, and to allow actors to act – meaning contract as they like (consistent with NAP, of course).

    As a 100% standard can only come about via force, it cannot be more stable than the market.

    As to what Mises did or didn’t say about 100% commodity money or gold, I find both Mises (Human Action, Ch. 17, Sec. 12) and Rothbard (The Mystery of Banking, Ch. VIII) to make very eloquent statements about the ability of the market to regulate banking practices.

    Whatever else either of them have written or said on this subject (and certainly quotes on both sides can be found), if the market is an effective regulator, I find no reason to butt in.

  8. says

    The government certainly should not decree a gold standard, or any standard. In a truly free market, gold and silver would be used as money, regardless of any government decree. The government’s only role should be to enforce a law against counterfeiting.

  9. waramess says

    Both the article and the comments are very interesting, if a bit high church but they all miss one great fundamental: why should I be obliged to lend my money to a bank if I do not wish to?

    Of course one might say I could put it under the matress, but that is a bit impracticle as the money laundering regulations would make it difficult to put back into the system should I want to.

    The truth is that should I wish to have a cheque account and cash card I have no option than to lend it to a bank who give me no security,and whose creditworthyness is completely unknown except that its portfolio is almost certainly stuffed full of dodgy assets.

    Presently I have no option but, a 100 percent reserve system would give me the option I wish for whilst those who wish to earn interest on there money would be free to lend it to the bank

  10. daniloux libertarian says

    Dear Detlev,
    Fractional reserve banking is in this monetary contest is a scam because even if indirectily, causes all the problems we are aknowledged with. It could be that in a monetary system with sound foundation there would be a chance that a perfect noticed frb would exist without very alarming implications.. But in that contest the risks wouldn’t be socialized so extensively as it is now.
    Dear Doctor Selgin,
    What i found very alarming in your point of view is the lack of consideration for the debt ratios of the economies in the present monetary system. Do you really think the monetary equilibrium can always be met with some frb because this is the best emulation of TFB? Do you really think the next financial burst could ever be halted by that? No, it can’ t. Us economy is leveraged 3.5 times. What the market really wants is lowering that ratio and so whatever mechanics you run the implosion of Us economy cannot be stopped. it can only be delayed by some other money printing, through direct bond purchases, because at the present status even ultralow interest rates are not sufficient to stimulate frb. Can you please explain what is your view on the sustanibility of this system? How much can, this second best solution, last?

  11. karol zimmer says

    Dear Detlev,
    thanks for presenting the main arguments of the FRB. Yet, it helped me confirm the FRB position is wrong.

    As Per-Olof plainly states, if a man lends more money than he posseses, he commits a fraud. It makes no difference, if it is “only” in a form of mere claims to money proper. I cannot see how a bank could/would compensate the third parties (depositors, holders of money claims, previous borrowers) for the decline in their claims’ purchasing power. Without such a compensation, the lending contract involves costs for third parties. The operation is, in its effects, no different to direct counterfeit of money proper.

    So FRB is actually a ongoing free ride on ownership of third parties. And therefore has nothing in common with austrian philosophy, for which private property is holy.

    FRB would indeed give banks a different treatment than other businesses have. Grain silos cannot sell or lend grain deposited with them. It would be treated as fraud. I believe no one would be advocating such an act because “it would smooth grain markets”. Yet, it would.

    It is symptomatic that FRB proponents need to abolish the real, microview of a contract’s flaws and get to the macrolevel so as to find its supposed benefits and defend FRB.

    As Per-Olof writes, this is not a new discussion. It is in a way a repetition of banking vs. currency schools fight in 19th century England. Since then, we live with the winners’ unintended consequences.

  12. OneTinSoldier66 says

    I really enjoyed this article. Very intellectually stimulating and thought provoking in my opinion. I will state that I am not for any “ban” on FRB for reasons stated pretty clearly by Detlev and posts by others on here. I would like to interject a critique at Detlev and George Selgin though.


    “But to my initial shock and my lasting amazement I found that some Austrian free bankers frequently cannot bring themselves to reject ‘quantitative easing’ and other heavy-handed central bank intervention on principle, and that they are able to embrace monetarist policy proposals, such as nominal GDP targeting by central banks, as a kind of second-best-solution that will do for as long as our first choice of separation of money and state is not realized.”

    You used the word ‘monetarist’, but you still seem to cling to some sort of belief that these people are Austrians. You stopped just short of calling them out for what they really are. If you embrace monetarists policies then you are a monetarist. If a person really has principles, then there is no excuses. To me, a real Austrian wouldn’t try to claim it would be “the end of the world” or some such thing as an excuse to be a monetarist because it happens to suit them at the moment.

    @George Selgin

    “To argue against the view, taken by myself and some other free bankers, that the Fed ought to have acted to avoid the collapse in question, is not to take a firmer stand against monetary central planning than we ourselves take.”

    Well, what would it take to make a firmer stance against monetarist central planning then? Although I’m not to sure I should be asking this question of someone who claims to take a stand against the very thing that they also believe should have been, or should be done.

    “It is merely to apologize for what was in fact a particularly flawed monetary central plan!”

    It is no apology when you endorse the maintaining of the particularly flawed established status quo by having the Fed act to maintain it!

  13. QuickNote says

    The following contradicts with your definition of demand for money!

    “…The demand for money, properly understood, refers to the desire to hold money as part of a financial portfolio….” in Theory of Free Banking by Selgin

    “…It is only when people who receive money income elect to hold it rather than spend it on other assets or consumer goods that they may properly be said to have a demand for money…” in Theory of Free Banking by Selgin


Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>