Saving is no fun. Americans have very nearly given up the habit — what good reason could there be for reducing consumption? The desire to avoid this painful choice has motivated a near endless search by monetary cranks and inflationists for alchemy: if the means to turn paper into real wealth could be found, material progress could be greatly accelerated without the pain of saving.
Antal Fekete claims to have discovered just such a mechanism: clearing. Fekete claims that clearing (implemented through his marvelous Bills of Exchange mechanism) enables production to be funded without corresponding savings.
We will show that, while there is nothing wrong with the issue of Bills, they do not perform the same work as savings. While clearing reduces demand for cash, it does not reduce the amount of savings required to fund investment.
Fekete’s error is confusing the financing of production with the funding of production. Any amount of business plans can be financed through the issue of more paper. But the funding of these plans is limited by the capacity of the economy to produce final goods. Expanding the quantity of money, credit masquerading as money, near money, or money substitutes cannot increase this capacity.
Mises and Rothbard on Clearing
In order to address Fekete’s errors on the topic of clearing, the economics of clearing and netting will be reviewed from the perspective of Mises and Rothbard. I will make a comparison to the writings of Fekete where relevant.
Rothbard gives a short explanation of clearing:
Clearing is a device by which money is economized and performs the function of a medium of exchange without being physically present in the exchange.
A simplified form of clearing may occur between two people. For example, A may buy a watch from B for three gold ounces; at the same time, B buys a pair of shoes from A for one gold ounce. Instead of two transfers of money being made, and a total of four gold ounces changing hands, they decide to perform a clearing operation. A pays B two ounces of money, and they exchange the watch and the shoes. Thus, when a clearing is made, and only the net amount of money is actually transferred, all parties can engage in the same transactions at the same prices, but using far less cash. Their demand for cash tends to fall.
The above passage applies only to cases where the transactions to be cleared occurred at the same time. In the following passage Mises discusses the extension of clearing to transactions that occur at different times. This is done through a combination of clearing and credit.
When all exchanges have to be settled in ready cash, then the possibility of performing them by means of cancellation is limited to the case exemplified by the butcher and baker and only then on the assumption, which of course only occasionally holds good, that the demands of both parties are simultaneous. At the most, it is possible to imagine that several other persons might join in and so a small circle be built up within which drafts could be used for the settlement of transactions without the actual use of money. But even in this case simultaneity would still be necessary, and, several persons being involved, would be still seldomer achieved.
These difficulties could not be overcome until credit set business free from dependence on the simultaneous occurrence of demand and supply. This, in fact, is where the importance of credit for the monetary system lies. But this could not have its full effect so long as all exchange was still direct exchange, so long even as money had not established itself as a common medium of exchange. The instrumentality of credit permits transactions between two persons to be treated as simultaneous for purposes of settlement even if they actually take place at different times. If the baker sells bread to the cobbler daily throughout the year and buys from him a pair of shoes on one occasion only, say at the end of the year, then the payment on the part of the baker, and naturally on that of the cobbler also, would have to be made in cash, if credit did not provide a means first for delaying the one party’s liability and then for settling it by cancellation instead of by cash payment.
Mises provided a further analysis of the transfer of claims. Bills that are not yet settled can be transferred within a network in place of cash payment. In this case, claims attain the status of money substitutes.
exchanges made with the help of money can also be settled in part by offsetting if claims are transferred within a group until claims and counterclaims come into being between the same persons, these being then canceled against each other, or until the claims are acquired by the debtors themselves and so extinguished. In interlocal and international dealing in bills, which has been developed in recent years by the addition of the use of checks and in other ways which have not fundamentally changed its nature, the same sort of thing is carried out on an enormous scale. And here again credit increases in a quite extraordinary fashion the number of cases in which such offsetting is feasible.
The use of credit in a clearing transaction requires the payment of interest to the party who accepts a claim that will not settle until some time in the future. The payment of interest on bills is accomplished by trading the bill at a nominal value less than its full principal value. This is called "discounting." The discount is computed from the short-term interest rate and the amount of time from the payment date to the settlement date. Mises explained how discounting enables the problem of non-simultaneity of transactions to be solved:
Since it was the general custom to make payments in this way, anybody could accept a bill that still had some time to run even when he wanted cash immediately; for it was possible to reckon with a fair amount of certainty that those to whom payments had to be made would also accept a bill not yet mature in place of ready money. It is perhaps hardly necessary to add that in all such transactions the element of time was of course taken into consideration, and discount consequently allowed for.
Fekete’s discussion of the process parallels that of Mises.
Yet the supplier can use the bill to pay his own suppliers. Endorsed on the back, the bill can be passed along a number of times, the endorsement indicating that title to the proceeds has thereby been transferred from payer to payee. This transaction is also called “discounting” as the payee applies an appropriate discount, calculated at the current discount rate, to the face value of the bill proportional to the number of days remaining to maturity. Upon maturity the last payee presents the bill for payment to the producer on whom the bill is drawn.
Mises wrote in 1912 on the origin of clearing:
The modern organization of the payment system makes use of institutions for systematically arranging the settlement of claims by offsetting processes. There were beginnings of this as early as the Middle Ages, but the enormous development of the clearinghouse belongs to the last century. In the clearinghouse, the claims continuously arising between members are subtracted from one another and only the balances remain for settlement by the transfer of money or fiduciary media. The clearing system is the most important institution for diminishing the demand for money in the broader sense.
Fekete has also written on early clearinghouses:
Let us look at another instance of clearing and self-liquidating credit that was vitally important in the Middle Ages: the institution of city-fairs. Among the most notable ones were the fairs of Lyon in France, and those of Seville in Spain. They were annual events lasting up to a month. They attracted fair-goers from places as far as 500 miles away who brought their merchandise to sell, as well as their shopping-list of merchandise to buy.
A significant proportion of Fekete’s writings concern the explication of clearing arrangements. Mises and Rothbard also have provided a full explanation of clearing, netting, settlement, and discounting. These mechanisms are well understood by economists of the Austrian School. And, there is no problem with clearing. As will be shown, clearing is nowhere near the miracle that Fekete claims.
Clearing and Transaction Costs
We now examine the economic effects of clearing. The two most important effects are the reduction of transaction costs and the reduction of money demand.
First we examine the reduction of transaction costs. Consider the following example. Suppose that there are two banks, Bank F and Bank H. Customers from one bank frequently deposit checks in their accounts drawn upon the other bank. Each bank must settle these checks against their bank of issue. The banks are in the custom of settling inter-bank balances in the following manner:
- During a business day, 1000 oz of checks drawn upon Bank H are deposited in Bank F.
- Acme armored car service transports 1000 oz of gold bars from Bank H to Bank F.
- The same day 900 oz of checks on Bank F are deposited in Bank H.
- Ajax armored car service transports 900 oz of bars from Bank F to Bank H.
There are obvious efficiencies that could be realized by netting. On the day used in the example, with netting, only 100 oz would have to be transferred, and only in one direction. This step alone would reduce the value of the cargo in the trucks, and consequently the insurance premiums by about 95%. Wage and vehicle costs and would be reduced by around one half because the truck would only make one trip rather than two. On days when the clearing balances happened to be equal, no transport at all would be required.
Further efficiencies could be gained if Bank H and Bank F loaned each other the net amount from day to day, on the assumption that a daily net clearing balances in one direction on one day would tend, over the course of a month, to approximately cancel out. Settling for the net amount (including the interest on the daily loans) once per month would reduce costs additionally, compared to daily netting, by a factor of about 30-to-1.
Further cost savings could be realized by including other banks. Suppose that there are N banks within a clearing network. If each bank settled with each other bank, there would be around 2*N2 exchanges without netting in either direction. If the net position of each bank relative to all other banks were calculated each day, then each bank could make a single transfer of its net clearing balance, for a total of N exchanges.
Once started, there will tend be a competitive process driving the adoption of clearing systems. When at first a few firms start to use a clearing system, they will be able to reduce their money demand and correspondingly the reduction of money demand enables those firms to offer higher money prices for factors. If other firms in their industry did not also adopt a clearing system, they would find that they were being outbid for factors by the firms using the system. In most cases, a rapid readjustment of factor prices will occur as the remaining firms join the clearing system.
There will be changes in wealth distribution from this shift. The first movers will have made some gains at the expense of the late adopters because they will have reduced their money demand and thus been able to purchase scarce factors at the original, lower prices. But overall the changes in factor prices reflect the reduction in money demand — factors do not become cheaper in real terms when the purchasing power of money changes. Only because of the reduction of gold bar transport will overall costs be slightly reduced.
Clearing and the Demand for Money
There are two secular influences on the long-term trend in the demand for money. Economic growth and clearing. They have opposite effects, with economic growth tending to increase money demand because more goods are produced so more transactions take place.
Clearing tends to reduce money demand because less money must be held for the settlement of transactions. Rothbard notes, the "major long-run factor counteracting this tendency and tending toward a fall in the demand for money is the growth of the clearing system." Mises explains how this occurs:
The reduction of the demand for money in the broader sense which is brought about by the use of offsetting processes for settling exchanges made with the help of money, without affecting the function performed by money as a medium of exchange, is based upon the reciprocal cancellation of claims to money. The use of money is avoided because claims to money are transferred instead of actual money. This process is continued until claim and debt come together, until creditor and debtor are united in the same person. Then the claim to money is extinguished, since nobody can be his own creditor or his own debtor
A reduction in money demand, as for any other good, shows up as a lower price for that good (assuming that supply does not change at the same time). What does it mean for money to have a lower price? The concept of "a lower price for money" is more difficult to explain than for a (non-money) good because money does not have a price as such — it has many prices. The prices of all goods, expressed in terms of money are the inverse prices of money expressed in terms of goods. If a loaf of bread sells for $2, then the price of dollars in terms of bread is . A lower price for money means higher money prices for goods.
But does this matter? Fekete suggests that it does. He proposes that a limited quantity of money per se is a constraint on production:
To put the matter differently, [under the RBD] the gold standard [i.e. the relatively fixed supply of money] is no longer a fetter upon technological progress and further division of labor, as it would be in the absence of the bill of exchange. …The bill of exchange has opened up new avenues for progress, leading to great improvements in the condition of human life on earth. Technological progress will never again be obstructed by a dearth of gold.
[Explications added - Blumen]
On the contrary, the nominal purchasing power of a single money unit (a coin, gram, or ounce) does not matter where production is concerned. Here, we join with Charles Carroll in "denouncing the idea that an increasing trade necessarily requires an increase of money, as an error and a delusion."
Economic calculation deals with ratios, nominal quantities. Ratios are formed between nominal quantities, tending to cancel out proportional variations. Workers, for example, are concerned with real wages — the ratio of their nominal wages to the nominal prices of goods that they wish to purchase. Investors are concerned not with nominal profits, but with return on equity, yields, and other dimensionless quantities.
Moreover, any quantity of money can perform any volume of transactions because the same real transaction can be performed at any money price. That is to say, there is no monetary benefit to the additional gold. (This conclusion did not come from Rothbard; it was already well known to classical economists such as David Hume).
Given a quantity of money, the same coin can turn over more, or less, frequently depending on the volume of transactions. If the money supply remains roughly constant while more transactions occur, the same coin will turn over more often. If clearing systems were not adopted in a growing economy, more turnover of each money unit would be necessary to settle the increased number of transactions. But the transactions could be performed just as well without clearing.
While it is true that clearing makes it unnecessary to use gold coin as intensively, this does not amount to any significant reduction in the consumption of scarce factors (except for the cost of loading more gold bars on trucks), only a slower turnover of the given stock of coins, whatever that is.
In the end, the nominal price changes resulting from clearing arrangements don’t make scarce productive factors cheaper in real terms. For capital to become cheaper in real terms, there must be more of it. Capital can only be created by the diversion of more final goods from consumption to savings.
Clearing and Savings
Fekete claims that savings alone are insufficient to fund capital investment, while the appearance of more Bills of Exchange provides a means of funding investment without savings. While this claim might seem incredible, I will present several direct quotations from Fekete’s own writings to establish it. Here, for example he states that savings are insufficient to finance capital investment:
Let me suggest it to you that no conceivable economy can generate savings so prodigiously as to move all the indispensable items to the consumer. I conclude that the division of labor could have never been refined, and the “roundaboutness” of the production process could have never been lengthened, beyond the level reached by the cottage industries of the medieval manors, wherein every family had to produce not only its own food and fuel, but also its clothes and shelter.
And here, Fekete writes,
…the real bill will do the miracle of financing production and distribution spontaneously, without taking one penny out of the piggy-banks of the savers, and without legal tender coercion.
As an inflationist in good standing, Fekete’s theory is firmly anchored in the confusion between money and wealth. Fekete starts with the true premise that clearing increases the efficiency in the use of cash, to the false conclusion that it allows production to be funded by a bill alone.
While the premise is true, the conclusion is false. Clearing has economic benefits, but it has nowhere near the magical properties that Fekete would have us think. Fekete’s extravagant claim regarding the ability of bills to substitute for actual savings is entirely erroneous.
Financing is not funding. Economizing the use of cash is not the same as economizing scarce real factors. Land, labor, and fixed capital do not come into being through the establishment of clearing systems. Economizing cash only enables the existing supply of factors to trade at higher money prices.
Final goods are used up in the process of producing other goods. Savings consists of the goods that are made available to producers for their consumption while they are not producing any final goods themselves. The saved goods are consumed in the service of funding greater production in the future. Mill used the term reproductive consumption to emphasize the two aspects of savings: consumption and production.
If money were savings, then more money (or more bills) would be the equivalent of more savings. But money is not savings: savings is in essence a non-monetary phenomenon. As E.A. Goldenweiser explains (quoted by Kurt Richebächer) "Saving means the withdrawal of sufficient resources from the production of consumption and services to have enough for maintenance, expansion and improvement of the plant."
Then, he adds a remark that could have been aimed at our contemporary RBD inflationists:
ever since Wesley Mitchell’s Business Cycles there has been a tendency to concentrate too much on the monetary expression of economic developments, and it has become reactionary to think in physical terms.
If a farmer were to consume an apple as a snack while on vacation, then no new production would have come about as a result. But a farmer who sets aside some apples from the apple harvest, then eats them to sustain himself while planting some apple trees that will bear more fruit in the future has reproductively consumed the apples.
It may come as a surprise that money is not savings. Living as we do in a monetary economy, we often think of savings as saved money because our saving is done with money. The difference between monetary savings and in-kind savings, as in the apple example above, is that with monetary savings, the transfer of money from the saver to the producer confers on the producer the ability to purchase goods on the market with the saved money. With monetary savings, the saver and the producer may be different people. The producer makes the decision of what kind of goods to save.
Fekete’s case for the fallacious Real Bills Doctrine relies on the alchemical properties of clearing systems. Clearing systems are said to overcome the savings deficiency that will inevitably appear in a growing economy. This conclusion is based on a serious misunderstanding of the nature of savings.
There is really nothing wrong with clearing, and Austrian economists have no issue with clearing systems, protests by Nelson Hultberg that Austrians wish to prohibit it notwithstanding. However, clearing has nothing to do with savings. More clearing does not mean more savings.
No quantity of bills of exchange could enable a single barrel of oil to be refined into twice as many gallons of gas, or a single loaf of bread to feed twice as many shoe makers. The amount of rubber, oil, bricks, computers, accountants, office buildings, or other factors of production that go into the manufacture of a car or a house would be unchanged, even if all intermediate transactions were settled in cash.
While savings are scarce, clearing is no substitute for them. The issue of Bills of Exchange, then, is a non-solution to a non-problem. Once it is understood that the fable that Fekete spins about clearing is another tall tale, there is no motivation for the rest of his doctrine.
For a bill to replace actual savings, the bill would have to be one and the same thing as a saved good, so that it could be reproductively consumed. In reality it is only a claim to that good. As André Dorais wrote to me in an email, "if you consider a Real Bill as a good and add one each time you produce a real good you would obtain two goods. But you did not produce two goods, only one."
To end, we can do no better than did Charles Holt Carroll when he wrote, "We cannot eat our cake and have it too; this truth was settled to the satisfaction of each one of us in the nursery."
Robert Blumen [send him mail] is an independent software developer based in San Francisco.