Free Banking Theory versus the Real Bills Doctrine: Reply to Larry White

Publicado el 20 septiembre 2015 por Juan Ramón Rallo

Last July, I held a very interesting conversation with Larry White about free banking and macroeconomic coordination. I was already very thankful to be able to talk to such an internationally recognized authority in free banking issues and, obviously, I am even much more thankful now after Larry has decided to write some new comments and criticism about our dialogue.

On the one hand, Larry criticizes the Real Bills Doctrine, as it is traditionally understood, i.e., as a norm for money issuing. On the other hand, Larry sees some sense (without embracing it) in the Real Bills Doctrine as a prudent banking norm: banks must certainly control their asset profile and therefore real bills may be a good reserve asset due to their short duration, low default risk and high liquidity.

Since my defence of the Real Bills Doctrine was done quite unsystematically in our conversation, let me summarize the main points of my thesis so that I can reply to Larry’s comments.

It’s not only about bank management, but about economic coordination

One of the merits of the Austrian Business Cycle Theory is that it links financial theory to the coordination of savers and investors through the structure of production. Savings are the intertemporal demand for economic goods; investments are the intertemporal supply of economic goods. Economic coordination requires that the intertemporal demand for goods matches the intertemporal supply of goods, and this can be achieved thanks to financial markets: demand for capital (investment) is channelled through the issuance of liabilities with a given time, risk and liquidity profile, while the supply of capital (saving) is provided by the purchase of those liabilities with a given time, risk and liquidity profile. Savers’ intertemporal preferences impose some quantitative and qualitative restrictions to investors’ intertemporal projects, so that intertemporal economic coordination becomes possible.

Banks are pure financial intermediaries: they intermediate between savers and investors by selling their liabilities to savers and by purchasing investors’ liabilities. Financial intermediation can be extremely useful to foster coordination as it allows a much larger amount of exchanges. However, financial intermediation can also lead to discoordination: if financial intermediaries do not match savers’ intertemporal demand with investors’ intertemporal supply, we will have a qualitative discoordination (supply of future goods will not satisfy demand for future goods in some respect: time, risk or liquidity). Let’s have a closer look at a particular kind of intertemporal discoordination: the supply of goods is not available when the demand desires it.

The banks that transform the maturities of their assets and liabilities are causing a discoordination between savers and investors. They are promising savers to redeem their liabilities much sooner than the moment when their assets will be paid by investors, i.e., they are promising savers the availability of some future goods before they are provided by the investors’ projects they are financing.

Of course, ex ante discoordination may be solved by ex post recoordination. Maturity mismatching does not necessarily imply that savers’ future demands will be totally frustrated: savers may postpone their demand for future goods or other savers might be willing to surrogate into their position (thereby rolling-over banks liabilities). Something similar happens with traditional Austrian Business Cycle theory: if new fiduciary media created by banks were completely saved by factors of productions (thereby turning forced saving into voluntary saving and preventing the Ricardo Effect from taking place), or if a substantial increase in productivity occurred during the process of credit expansion, an ex ante artificial boom would not turn into an ex post hard bust.  Something similar happens with the free banking “monetary equilibrium”: if there is an increase in money demand after banks have increased inside money supply above previous money demand, there will be no monetary disequilibrium. But in both cases we would say that we faced an ex ante macroeconomic uncoordinated process that was exogenously solved ex post. The same is true for maturity transformation, especially when many financial intermediaries perform it simultaneously.

In order to safeguard banks’ own liquidity and in order to coordinate savers and investors, banks should followed what Mises (1912) called the Golden Rule of Banking: “For the activity of the banks as negotiators of credit the golden rule holds, that an organic connexion must be created between the credit transactions and the debit transactions. The credit that the bank grants must correspond quantitatively and qualitatively to the credit that it takes up. More exactly expressed, ‘the date on which the bank’s obligations fall due must not precede the date on which its corresponding claims can be realized.’  Only thus can the danger of insolvency be avoided”. Mises thought that the Golden Rule should only be applied by those banks who acted as financial intermediaries, not by the ones who performed the function of issuers of fiduciary media. However, this is a mistake: banks that act as issuers of fiduciary media are also financial intermediaries. Banks’ demand liabilities are a particular type of financial assets with some specific characteristics —duration, risk and liquidity—that make them interesting for some savers. Therefore, the Golden Rule should be applied to them as much as to other financial liabilities.

But then, the question becomes the following: is there any kind of asset that allows banks to preserve their liquidity while issuing demand liabilities? Note that this question is totally relevant for free banking theorists. They usually argue that an increase in the demand for money (money in the broad sense) should be counterbalanced by an increase in inside money on the part of banks, i.e., by an increase in banks’ demand liabilities. In other words, the increase preference of savers for a particular type of financial asset must be attended by an increased supply of that financial asset. That is correct, but we should not forget that banks are negotiators of credit, not creators of credit: hence, the analysis of monetary equilibrium (how savers’ preferences are satisfied by banks) should always go hand in hand with the analysis of the bank asset portfolio (which investments are funded by banks). To state the contrary is to state that we may analyse savers’ intertemporal demand disregarding investors’ intertemporal supply. If we are trying to figure out the conditions for intertemporal equilibrium, we obviously cannot.

Therefore, we do have to analyse what kind of investments are appropriate when banks issue demand liabilities, and this is where the Real Bills Doctrine fits.

Real Bills as the most liquid asset after money

What is the proper kind of asset to have against a demand liability? One may think that it is a demand asset, but as Mises (1912) explains: “If such banks made loans only on the condition that they had the right to demand repayment at any time, then the problem of liquidity would of course be solved for them in a simple manner. But from the point of view of the community as a whole, this is of course no solution, but only a shelving, of the problem. The status of the bank could only apparently be kept liquid at the expense of the status of those who borrowed from it, for these would be faced with precisely the same insurmountable difficulty. The banks’ debtors would not have kept the borrowed sums in their safes, but would have put them into productive investments from which they certainly could not withdraw them without delay. The problem is thus in no way altered; it remains insoluble”. Mises is right in highlighting that the liquidity of the creditor depends on the liquidity of the debtor, but is wrong in thinking that therefore there is no way to achieve aggregate liquidity for the whole financial community (In this case Mises is endorsing Keynes’ dictum that “there is no such thing as liquidity of investment for the community as a whole”).

The great monetary economist Melchior Palyi (1936) described liquidity in the following way: “Perfect liquidity means that, for any length of time, all financial obligations are fulfilled without net liquidation of capital. A liquid society has adjusted its obligations to the flow of its income, both in amounts and in maturity dates, so that forced sales should not occur (disregarding war, or other extra-economic factors)”. In other words, liquidity of the financial system is tantamount to coordination between the profile of aggregate intertemporal demand and the profile of the aggregate intertemporal supply.

Savers holding a demand liability against a bank are holding what they expect to be a highly liquid asset: a financial asset almost as liquid as outside money. High liquidity property of demand liabilities is continuously being checked by the fulfilment of the banking promise to keep their convertibility in outside money at par: banks are expected to manage their assets so that all their liabilities are fully paid at maturity. Some economists claim that this can only be achieved by a banking system with a 100% reserve in outside money. But that view forgets that there are other ways of managing banks’ assets and liabilities that ensure convertibility without net liquidation of capital, i.e., that ensure continuous convertibility of demand liabilities without forcing changes in the intertemporal demand of savers. No ex ante discoordination needed of ex post recoordination.

How can this be done? According to Real Bills Doctrine, banks should only discount real bills, i.e., they should only issue demand liabilities against one particular kind of financial asset: real bills. A real bill is just a commercial debt that arises out of the sale of a bundle of economic goods (technically, the drawer of the bill orders the drawee to pay a certain amount of money to the payee: the payee is the seller of the bundle of goods, while the drawee is the buyer of those goods). If these goods are already produced and are already in high final demand, their “objective exchange value” (using the misesian expression) will be stable until their imminent sale to final consumers: there is not technical uncertainty involved (since the goods are already produced) and it is quite doubtful that final consumer demand for some highly demanded goods and services will shift dramatically in the very short term. Stable supply and stable demand implies stable value, at least while those conditions remain unaltered. This should not come as a surprise. According to Menger, outside money is the most highly liquid good of an entire economic system: but the most highly liquid is not equal to the only liquid good; other goods may exhibit stable value and consequently may be used in indirect exchange.

Stable exchange value of some goods in high final demand (i.e. consumer demand for use, not for speculation) means two things.

The first one: bills in the bank portfolio will either be turned into outside money (if those goods have been sold against outside money) or they will be paid in the reflowing demand liabilities of the bank (if those goods have been sold against inside money): in the first case, the bank gains outside money that makes it easier for the bank to keep convertibility; in the second case, real bills and demand liabilities are mutually cancelled out, so that convertibility of the remaining demand liabilities improves.

The second one: holders of demand liabilities can change without cost their intertemporal demand for goods whenever they desire (as if they were hoarding outside money). Holding a bank demand liability (or a real bill) is tantamount to holding a share of a very stable turnover of economic goods whose realization just depends on the fulfilment of the strongest present demands for consumption goods. In other words, holding a bank demand liability is quite similar to holding a credit balance against a clearinghouse where just present consumptions goods are admitted as means of payment (so that no promises to deliver future consumption goods are admitted). Let us assume for the sake of simplicity that agent A sells a bundle of consumption goods to agent B and agent B uses them not for his own consumption but either to invest or to sell. In the first case, if agent B uses the bundle of consumption goods as a way to finance the short run production of more consumption goods (which will be used later to finance the short run production of more consumption goods), we can say that agent A has a claim against agent B which can be repaid in consumption goods at any time. In the second case, if agent B sells the bundle of goods for a real bill against C, agent A will retain an ultimate claim against the bundle of goods of agent C (or against the agent who has purchased agent C’s goods for money or real bills). Therefore, one can use their share of a continuous flow of liquid consumption goods as a bearer of options to consume or to invest in the future (one could look for parallels between that stable turnover of consumption goods that provide liquid means for investing and the classical wage-fund or the Striglean subsistence fund: real bills are backed by free capital in the form of highly demanded consumption goods).

Regarding these two properties, let’s compare the bank demand liability issued by discounting real bills with the bank demand liability issued by discounting mortgages.

The first one: a mortgage will not be paid in full neither in outside money nor in inside money in the short run, so banks will not be able to recover any liquidity but by selling the mortgage to some third investor. In this case, the problem clearly becomes that the liquidity of the bank improves at the expense of the liquidity of that third investor: there is no way both can increase their liquidity in the short run.

Secondly, holding a demand liability issued against a mortgage does not allow you to change your intertemporal demand without cost: you need to change other people’s interpersonal demands (in the previous case, third party investor had to renounce to his liquid assets in order to start saving long term) and therefore you’ll need to compensate them. If you wish to consume or to invest in other assets different to the mortgage, you would need to find a surrogate that takes on your “forced saver” position (normally, at a loss on your part). For instance, if agent A sells a bundle of consumption goods to agent B and agent B uses them to invest long term, it is clear that agent A does not retain any option to shift his plans until the long term investment frees up through the production of the same amount of capital which has been fixed in that long term investment. Similarly, if agent B sells the bundle of goods not against money or a real bill, but to an agent who will only be able to pay with production in the distant future, agent A does not retain any option to shift his plans meanwhile.

In other words: an economy where every saver is holding real bills is a liquid society, both from a financial and from a productive point of view (invested capital is fully recovered in the short run, which allows paying down debts and reinvesting capital in other business lines). An economy where every saver is holding a mortgage is a very illiquid society, both from a financial and from a productive point of view (savers will only recover their “free capital” in the very long run).

I could accept, although not fully agree, with some economists’ statement that the maturity transformation of banks (specially performed by neglecting the Real Bills Doctrine) is socially useful because individual rationality would lead to a too high demand for liquid assets, therefore diminishing our capabilities to invest in a long term productive process. But I cannot concur with other economists’ statement that the maturity transformation of banks provides both liquidity to savers and long term funding for investors. Devoting savers capital to long term investments is not a way to preserve their liquidity: it is a way of leading them to illiquidity under the expectation that there will be enough liquidity remaining in the rest of the economy (which may be doubtful specially in those moments in which liquidity is more highly valued: i.e., where there is a crisis of liquidity). In the quite clear words of Henry Parker Willis:

A difference between this kind of credit operation on one hand and commercial credit on the other may be found in the fact that whereas reimbursement or liquidation of commercial credit comes from the interchange of existing goods or goods which are on the point of being rendered consumable, investment credit may eventually be liquidated as a result of savings made through the increase of productive power. Both interest and principal will be repaid to those who have advanced the current funds only as a result of a real increase in wealth resulting from the productive operation. It is evident from this analysis that the operations which are covered by the term investment banking rest upon a very different basis from that which is usually included under the term commercial credit. The function of the investment bank or credit institution is that of accumulating units of savings or current funds whose owners are wilhng to allow them to be converted into investments — productive or income-yielding opportunities. The difference between such credit and that furnished by the commercial bank is ordinarily spoken of as being one of time, but, as has already been frequently indicated, the essential difference is not that of time, but rather of the use that is made of time. It is a difference in the character of the enterprise that is undertaken. The characteristic enterprise undertaken with the use of commercial credit is that of bringing together consumers and producers, while the characteristic enterprise undertaken by the investment institution is that of bringing together producers and those who desire income rather than immediate enjoyment of capital. This is a difference which, as will be seen, involves broad and fundamental differences of method m banking and far-reaching differences in canons of judgment as to banking soundness or liquidity [Emphasis is mine].

To summarize: the Real Bills Doctrine just states that the proper financial asset for preserving savers liquidity is real bills, or in more general terms, IOU’s that are collateralized for goods in process of being sold to final consumers. The tradition of the Real Bills Doctrine (or, more broadly, the tradition of liquidity theorist) started with Adam Smith, but has included such renowned economists as Jean Baptiste Say, William Huskisson, James Wilson, James Laughlin, Henry Parker Willis, Felix Somary, Heinrich Rittershausen, Charles Rist, Jacques Rueff or Antal Fekete. I would also include Carl Menger in this tradition, attending to the final pages of his book Money(1909) and to enthusiastic defence of the Real Bills Doctrine by his student Eugen von Philippovich, but since Menger never developed a theory of credit, this inclusion is rather risky.

Let me include some quotes to illustrate my point:

The notes issued by a bank of circulation, even if it have no funds of its own, are never issued gratuitously; and, therefore, of course, imply the existence, in the coffers of the bank, of a value of like amount, either in the shape of specie, or of securities, bearing interest; upon which latter only the whole real advance of the bank is made; and this advance can never be made upon securities that have a long time to run; for the securities are the fund, that is to provide for the discharge of another class of securities, in the hands of the public at large, payable at the shortest of all possible notice, namely, on demand. Strictly speaking, a bank can not be at all times in a condition to face the calls upon it, and deserve the entire confidence of the public, unless the private paper it has discounted, be all, like its own notes, payable on demand; but, as it is no easy matter to find substantial assets, that shall bear interest, and at the same time be redeemable at sight, the next best course is to confine its issues to bills of very short dates; and, indeed, well-conducted banks have always rigidly adhered to this principle.

From the preceding considerations may be deduced a conclusion, fatal to abundance of systems and projects, viz. that credit-paper can supplant, and that but partially, nothing more than that portion of the national capital performing the functions of money, which circulates from hand to hand, as an agent for the facility of transfer; consequently, that no bank of circulation, or credit-paper of any denomination whatever, can supply to agricultural, manufacturing, or commercial enterprise, any funds for the construction of ships or machinery, for the digging of mines or canals, for the bringing of waste land into cultivation, or the commencement of long-winded speculations; any funds, in short, to be employed as vested capital (Jean Baptiste Say, 1803).

Though these transactions are performed on the mere personal credit of the parties connected with them, yet the whole basis on which they are built, and without which they could not be sound, is the various commodities which they represent A bill is good, and is certain to be paid, because it is supposed to represent goods delivered to the acceptor; which goods, not being consumed, but used for the purposes of reproduction, will be repaid out of the current income of the community who purchase them from the dealers for consumption. The current income of the country is thus pledged in reality for the repayment of loans made on bills of exchange, representing commodities sold to the acceptor; and it is the certainty of being thus provided for from current income in the usual course of business, that bills of exchange constitute a sounder banking security, though really only dependant on personal credit, than landed property, railway shares, or any other object of fixed capital, the interest of which is only paid out of income. A loan based upon the security of commodities, which are entirely replaced from the income of the country, presents a certainty of repayment at a stated time, that a loan based upon fixed property does not. The former is the best temporary investment, although contingent only on personal security; the latter would be the best security for a lengthened and permanent loan. The former is a fitting security for a banker who wants temporary employment for his deposits; and the latter is the most fitting security for the mortgagee who wants a permanent investment for his capital. The former securities represent floating capital, while investment in the latter converts floating into fixed capital, and is, therefore, wholly unsuitable for bankers (James Wilson, 1847).

A man’s purchasing power consisted of all his cash, and of all his immediately saleable goods; or, of all his cash plus all his credit. The general purchasing power of a community, therefore, directed against all goods, is composed of all the cash, plus all the immediately saleable, or bankable goods. This, however, is only a statement of the machinery by which all goods, — all supply and all demand — are exchanged against each other. In truth, normal credit, by coining salable goods into present means of payment, merely sets more goods into circulatory exchange against each other than would be possible without the use of credit (James Laughlin, 1905).

The preceding analysis shows that in all cases the demand for liquidity implies a demand for wealth of equal value. This wealth can, according to circumstances, be metal or credits, themselves representatives of goods stored or sold on credit. We are therefore not entitled to conclude that “liquidity preference” diminishes proportionately the purchasing power impinging on the market. This always remains determined, everything remaining the same, by the value of the production offered there. The demand for liquidity —like every demand, whatever its nature— simply sets forces in motion which tend to stimulate in the productive apparatus the adaptation capable of satisfying it (Jacques Rueff, 1947).

Can the Real Bills Doctrine solely prevent the over issuing of credit?

Larry White is particularly critical with the Real Bills Doctrine when understood as a monetary norm for monetary issue. According to some interpretations of the Real Bills Doctrine (in particular, the Antibullionist interpretation during the suspension of payments of the Bank of England), if banks discount real bills, over issuance is not possible because banknotes in excess supply will reflow in the short run to his issuer.

As it may be apparent in the previous lines, I am an advocate of John Fullarton’s law of reflux although not under the constraints Fullarton conceived. The Real Bills Doctrine is a necessary condition for the reflux to operate, but it is not a sufficient condition. Just as Mises or Larry pointed, discounting real bills is no guarantee that defaulted bills are not redrawn at maturity. Quoting Mises (1912): “Even if it is true, as Fullarton insists, that banknotes issued as loans automatically flow back to the bank after the term of the loan has passed, still this does not tell us anything about the question whether the bank is able to maintain them in circulation by repeated prolongation of the loan”.

But the Antibullionist version of the Real Bills Doctrine was not the one described by Adam Smith. Adam Smith described the self-regulating process of money creation by a bank (which has been known as the Real Bills Doctrine) under three conditions: free money, free banking and prudent banking, i.e., gold and silver standard, Scottish competitive banking, and real bills as a banking prime asset. One without the other does not work: free money without free banking (for instance, a 100% reserve gold standard) would provoke a secular deflation that would distort the coordination of economic agents; free banking without money would not really be considered free banking due to the fact that the government would be in charge of the final liquidation of debt. Prudent banking without free money and without free banking would not really be prudent because there would be no incentive to achieve the goal of prudence through self regulation, therefore prudent banking would be achieved through state regulations which would either be too restrictive or too lax; free money and free banking without prudent banking would condemn the financial system to massive  illiquidity and business cycles.

Upon arriving at this point, Larry White and other free bankers may ask: if free money and free banking can provide the optimal amount of prudent banking on its own, so there is no need to theorize about what prudent banking really means (i.e., there is no need to theorize about the Real Bills Doctrine and we can just stay comfortable with the free banking theory). For me, this is a mistake for two reasons. First, in order to check whether free banking provides the optimal amount of prudent banking we need to have a theory about what prudent banking is: prudent banking is not whatever free banking does (otherwise we are falling into a petitio principii). Secondly, even if free banking always achieved the optimal amount of prudent banking, economic theory should try to describe what prudent banking is and how it relates to macroeconomic coordination: up to now, free banking theory has focused almost exclusively on the macroeconomic coordination between banks’ liabilities and bank creditors’ needs (monetary equilibrium) but it has not tried to study the consistency between that equilibrium and the equilibrium between banks’ assets and bank debtors’ cash flows arising out of their investments.

If I had to answer what kind of institutional framework is more akin to produce prudent banking and macroeconomic coordination, I would say that it is free money and free banking without any need for regulating prudent banking (as the heavily-invested-in-real-bills Scottish free banking system shows). But that is not tantamount to say that free money and free banking always provides the optimal amount of prudent banking thus completely avoiding business cycles: it just means that free money and free banking tend to minimize both. I believe, it is too strong a statement to say that we can completely get rid of imprudent banking and business cycles through free money and free banking. The task of the monetary economist is to answer that question independently of our libertarian political convictions: and to properly answer that question we need to rely on the tradition of the Real Bills Doctrine as I have tried to demonstrate.