By Dr. Walter Zander

[Taken from John Zube’s peace plans #9]

The subjoined study is the outcome of a lecture delivered by the author on 8 march 1935 before the National Institute Of Geneva (Institut National Genevois).

The devaluation of the be belga, which supervened in April 1935, is taken account of here. On the other hand, the May 1935 crisis of the French franc took place after this study had gone to press. It forms an additional link in the lengthening chain of monetary difficulties and offers one more illustration of the dangers to which every monetary system, even when supported by an enormous gold reserve, is exposed under the rule of a forced rate for the notes of the central bank. The proposals so far made known for ending the French monetary crisis do not suggest anything beyond what is implicit in the views generally current today.

—-Berlin,  2 June 1935, Dr. Walter Zander

(This reprint is made from the English edition of the “Annals of Public and Cooperative Economy,” the volume “Organized Compensatory Trading”, edited by Prof, Edgar Milhaud and published by Williams and Norgate, 1938. ‑ Compare page 75.

I possess only a photocopy of it and judging by the style to of the translation I assume it has been undertaken, like the other translations above, by G. Spiller, London. I compared this translation with the German version, a pamphlet: “Ein Ausweg aus dem Waehrungs‑Chaos”, a special reprint from: Deutsche Sparer‑Zeitung, Berlin, No. 18‑20, 1935, and found the translation to be considerably better than those of plans 190 and 192. I undertook therefore only very slight changes.

Ulrich von Beckerath, made the following annotation to one of the German copies in my possession:

“Between philosophy and economics there exists an intimate relationship. Proudhon said rightly: ‘L’économie c’est la métaphysique en action’.

At present the philosophy of subordination prevails, being originally already developed by Plato and ‘perfected’ with regard to subordination by some scholastics (not by all). Hegel, later on, crowned this system. The subordination of individuals under the central banks of issue is related to the prevailing philosophy of subordination. Quos ego!” – – The Ed, John Zube)


The States participating today in the world economy may, roughly said, be divided into two groups. One of these has abandoned the gold standard and has thus removed the foundations of its currency. The other has introduced foreign exchange legislation and has thereby abolished the freedom of settlement operations. The first group includes more particularly England, the sterling bloc countries, the United States, and Japan; the second group, Austria, Germany, Russia and the great majority of the remaining countries. Placed between these two groups, is the steadily disintegrating gold bloc. Italy left it for all intents only recently and yesterday as it were Belgium followed suit. If we bear in mind that since the War France and Poland had already devaluated their currencies to a fraction of their pre‑war parity, virtually Holland and Switzerland alone may be said to uphold the pre‑war monetary system. However, even in these two small countries, which represent but a tiny fraction of the world economy, certain restrictions concerning the convertibility of banknotes and the gold market have been introduced, and the general uncertainty as to the future of currencies exerts a baleful influence in these countries, also.


Whether the abandonment of the gold standard is advantageous to an economy is decidedly problematic. In most cases the object aimed at has not been attained. So far as devaluation is intended to stimulate exports, we should not forget that for most countries the magnitude of their export trade compared to that of their total trade is comparatively insignificant. This alone renders it questionable to embark, for the sake of the export trade, on measures which modify the basis of a national economy as a whole.

Moreover, such measures, if successful, may be imitated by any other country, thus nullifying heir favourable effect. Accordingly, the abandonment of the gold standard has led to “a race for the worst currency”, in which the most powerful States are participating. It is obvious that in the long run such measures for the stimulation of the export trade are bound to prove worthless,

Moreover, devaluation does not only affect the export trade of a country. On the contrary it touches every branch of an economy. This holds most especially of imports since these must become dearer to the precise extent that exports become cheaper. So far as exporting, like in Germany for example, presupposes the importing of raw materials, a portion of the anticipated gain is thus necessarily lost. But in general, import restrictions, so popular today in many countries, lead eventually to a decline in exports, for the simple reason that in the last resort exports can only be paid with imports.

Furthermore, money debts abroad are augmented by a devaluation of the currency. However, what is of crucial importance is the fact that an abandonment of the gold standard involves the devaluation of the entire savings of a country, particularly as invested in savings banks, State loans, bonds, and mortgages. If the effect of a general adjustment of prices to the fall in the value of money is not visible at first or is deferred by artificial devices, eventually prices always rise when a currency has been devaluated,

Thus while the intended advantage for the export trade is emphatically dubious and at best only of passing importance, the loss in savings and capital is certain and lasting.

But even if the reason for abandoning the gold standard, as in the case of the United States, is the desire to devaluate capital savings, ‑ that is, if it is intended thereby to adjust the claims of creditors to fallen prices and to the shrunken turnover of the debtors ‑ the success is nevertheless more than doubtful. For what is decisive for the value of a claim is not so much its magnitude as the business turnover of the debtor. Everything depends therefore on increasing trade and this is nowise assured by attacking the rights of creditors, to say nothing of the moral and economic disorganization which this creates.

The heavy depletions on the capital market, the abrogation of the rights of creditors, the menace to State credit, and the decline in the standard of living, represent drawbacks which ultimately outweigh the transitory advantages.


The injuriousness of foreign exchange legislation is even more patent. Everybody agrees and the President of the Reichsbank, Dr. Schacht, has repeatedly expressed himself to this effect, that foreign exchange legislation constitutes a great evil, even though in most quarters such legislation is considered inevitable.

Foreign exchange legislation places obstacles in the way of settlement operations. These obstacles, in turn, hamper trade. But every decline in trading leads necessarily to a decline in well‑being, as the latter, in our age of the division of labour, depends on commerce. Everything therefore that obstructs trading tends to intensify want and unemployment.


All attempts to surmount the difficulties involved in foreign exchange legislation have hitherto proved abortive. This is especially true of the different forms of compensatory trading and is evident as regards its earlier developments. It could happen then, for instance, that an instrument manufacturer had to accept in exchange coffee and was thus compelled to become a produce dealer. In principle, later developments left matters unchanged. It is the very essence of a compensatory transaction that there is properly speaking no money payment involved and that the goods themselves have to fulfill this function. That is, every such transaction represents a barter operation.

Whilst it is agreed that movements of goods (in the broadest sense) underlay all settlements, nevertheless as much as fifteen centuries ago the Roman Emperor Justinian explained in his Corpus Juris why barter operations are necessarily inferior to monetary transactions. And yet the distinguishing mark of the present‑day international compensatory trade is the abandonment of the monetary system and the falling back on barter. It is evident that a form of economy which the Corpus Juris deemed obsolete, must be unequal to the task of conducting efficiently the exchange of goods in our highly developed age. A serious effort should therefore be made to reinstate the system of monetary settlements in international transactions.

Similarly with so‑called clearing agreements between countries. These have now become very common, but they cannot remove the difficulties arising from foreign exchange legislation, for apart from the fact that in most cases, as between Germany and Switzerland, they have led to a considerable dislocation of trade, they are necessarily confined to transactions between two countries while the realities of life demand freedom of movement in every direction and rebel against bilateral arrangements. They lack therefore the necessary fungibility and the attempts to supersede the monetary system in this way have hence failed. The League of Nations Committee for inquiring into international clearing agreements has accordingly drawn attention recently to their disadvantages and recommended their abolition.


Shrinkage of international and of intra‑national trade, want and unemployment, heavy investment losses; and a general uncertainty and loss of confidence, characterize the present situation in most countries. It is imperative to create once more reliable and stable standards of value and to secure the removal of the impediments to settlement operations between countries, The solution of these closely related problems has become a question of life and death for our social order.


Whatever the monetary system of a country, it is essential that the measure of value should be clearly and unequivocally determined. Thus where there is a gold currency, a silver currency, or an index currency, the value should be measured by gold, silver, and the index respectively. This basis of measuring economic values, and therefore of any monetary system, is destroyed when in the case of a gold or silver currency the notes of the bank of issue are made legal tender, for this compels everybody to accept these notes in payment regardless of their real value. Compulsory acceptance renders it even impossible to measure the notes by the unit of value and thus to ascertain their value within the country. Indeed, it establishes a legal fiction on the basis of which note and unit of value are identical. For this reason, the names of the units of value ‑ e.g., the terms dollar mark, pound, ‑ become ambiguous in that they mean now a fixed weight of gold and then the note of a bank of issue. Accordingly, the measure of value, on the unambiguity of which everything depends, comes to have two definitions. This renders impossible any real measurement and thus the whole monetary system is falsified.

This falsification is generally hidden from the public as long as the central bank is legally obliged to redeem its notes. This, however, only masks the reality, since convertibility introduces in the measurement of value an alien element. Indeed the fact that convertibility becomes a decisive factor, shows how the whole problem has assumed a different complexion. Where convertibility is suspended, we have only a pure paper currency, this despite strenuous legislative and administrative efforts to keep the value of be paper at a certain definite level, for what counts now is no longer the value of the gold but the question whether the note of the central bank, measured by gold, changes its value. In fact, the system that has been in general use since the beginning of the World War, including the so‑called gold standard or nominal gold currencies, may be described as paper currency.


Although the compulsory acceptance of banknotes appears today so natural that most people cannot imagine a means of payment not having that character, this system has in reality only come into general use in recent times. Here are two illustrations

Par. 2 of the German Bank Law of 1875, provided:
“Payments statutorily required to be made in money need not be accepted when tendered in banknotes and the constituent States cannot enact such an obligation for the State treasuries.”

This provision was only replaced by its converse in 1909. Article 3 of the Act of 1 June 1909 decreed :
“The notes of the Reichsbank are legal tender.”

The course of development was similar in Switzerland. Here article 39 of the Federal Constitution of 1874 prohibited once and for all the compulsory acceptance of banknotes.

However, already in 1891 the Constitution was amended and they became legal tender in 1914.


The statutory obligation to accept the notes of the central bank in settlement operations involves not only the falsification of the basis of the currency. It also makes the fate of the currency dependent on that of the central bank and frequently on that of the banking system generally. If for any reason the central bank can no longer redeem its notes or maintain their parity ‑ that is, if the market rate of the bonds it issues falls ‑ then, owing to the legal equivalence between the notes of this bank and the legal standard of value, the calculation of values generally will be prejudicially affected. Thus it was in the main the situation of the Bank of England, which led in 1931 to the abandonment of the gold standard and similarly, it was the National Bank of Belgium that suggested in 1935 the devaluation of the currency.

Almost a century ago Lord Overstone grasped this interdependence when he said that if he ruined his private bank, he would be ruined, but that if the Bank of England committed a gross blunder, the Bank could save itself, but the whole of the community might have to suffer grievously.


Moreover compulsory acceptance for banknotes forms the legal and conceptual basis of every inflation. In the absence of such an obligation, bank crashes, with all their dire consequences, may occur, but never an inflation, because the destruction of the standard of value and the falsifying  of all monetary relations, which are the mark of every inflation, can never result from the collapse of a single bank. This confusion is only possible when a legal equivalence has been established between the notes of this bank and the standard of value. It was compulsory acceptance that brought forth the ominous slogan of the German inflation period: “One Mark is as good as another” (“Mark gleich Mark”). History has not known an inflation not due to the legal obligation to accept.


If then, the preconditions’ of an inflation are to be eliminated and a reliable and stable currency is to be assured, if, more especially, the gold standard is to be restored, the falsification introduced in recent decades must be eradicated and the earlier separation between standard of value and means of payment must be reestablished. A compulsory exchange rate excludes a stable and safe currency. No wonder the distinguished German historian Niebuhr stigmatized compulsory acceptance as “a legislative provision both ridiculous and abominable.” (Nachgelassene Schriften nichtphilologischen Inhalts, 1842, p. 485 ff.).

The reintroduction of the gold standard in Germany in October 1923, after the inflation, offers an impressive and instructive illustration. The notes of the Reichsbank, which were legal tender, had completely collapsed and their value could only be stated in astronomical fractions. At long last it was decided to introduce calculating in gold value. First, taxes were thus calculated. Then a new institute of issue, the Rentenbank was founded, whose accountancy basis was to be gold units. There was, and this cannot be too strongly insisted on, no legal obligation for the public to accept the new notes in payment, and these notes have to this day never become legal tender. They have therefore never been identified with the unit of value, which was then the gold mark. This system, which lasted from the passing of the inflation in the autumn of 1923 until the introduction of the new Bank Act in the summer of 1924 (which Act formed part of the series of Dawes Acts), was therefore a pure system of calculating in gold units which was not falsified by any compulsory acceptance.

A similar example, although confined to one domain only, is offered by China which recently adopted a gold unit for its customs charges. By a decree of the Minister of Finance of 15 January 1930, counting in silver in the departments of maritime customs, the most important for the Chinese budget, was replaced by counting in gold. The basis for these calculations in gold is a weight of 60,186 centigrammes of fine gold and represents the customs gold unit. This customs unit is exclusively a calculating unit and the decree specifically provides that the payment of duties may, as before be made in local means of payment,  that is in silver dollars and in banknotes. Of course, these are only accepted in payment at the current exchange rate, this being measured by the customs gold unit.

In conformity with the foregoing illustrations it is therefore suggested here to introduce generally (whilst abrogating the statutory obligation to accept the notes of a central bank) calculating in gold value and thus to re­establish the conditions existing until 1909 in Germany and until 1914 in Switzerland.


It might be objected that before the War compulsory acceptance could be dispensed with because then, unlike now, the notes could at any time be converted into gold. The objection does not hold for convertibility is not of decisive importance, as will transpire from what follows.


It is generally believed that the value of banknotes resides in their being convertible into metallic currency, banknotes being considered in the main a substitute for gold or silver. Already Adam Smith, in a famous passage in his “Wealth of Nations” (bk, ch 2), declared that all paper money represented only gold or silver. Similarly the Bullion Report of the British Parliament of 1810, which was very strongly influenced by Ricardo, expressed itself to the same effect. Probably not a single theory in the whole domain of political economy has evoked such universal assent.

It is therefore natural that this view should have been incorporated in legislative enactments. Since paper money is regarded as a substitute for gold or silver, it must be at any time convertible into these. Banknotes and convertibility are therefore interdependent conceptions and hence all bank acts the world over contain definite provisions concerning convertibility, metallic cover, and the ratio of the notes issued to the current gold reserve. Indeed, every monetary claim is hence regarded as being in the last resort a claim for payment in gold, although there is no necessary connection between this and a gold standard, for a gold standard primarily presupposes, apart from calculating in gold, that a creditor cannot refuse acceptance of a payment in gold, i.e., that there is a general obligation to accept gold but by no means the right of a creditor to insist, in any and all circumstances, on being paid, directly or indirectly, in gold.

Firmly based, on the one hand, as seems the general conviction that convertibility is necessary (and practically everywhere the ratio of the gold cover is deemed to be of the utmost importance), there is, on the other, no doubt that today most banknotes have become legally, or at least actually, inconvertible, without thereby losing all their value. In fact, for some notes there exist today no provisions of redemption, and yet they possess value. Accordingly, there must be, apart from convertibility, another basis for the value of paper means of payment.


All eyes are turned towards the State to see whether, by its fiat, it is able to confer value on valueless paper. It becomes however quickly manifest that the power of the State is strictly limited in this sphere. All large‑scale monetary devaluations known to history have referred to means of payment the value of which rested on a State fiat. This holds, for example, of the notes which the Scotchman John Law issued in France at the beginning of the eighteenth century and even more so of the assignats of the French Revolution. In both instances acceptance was at first not compulsory. But presently the obligation to accept them was decreed and soon reenforced by penalties. On 11 April 1793 the French Government prohibited the use of all metallic money on pains of six years in chains, and in September of the same year the decrying ‑ that is, the verbal discrediting of the assignats ‑ became punishable with death and the confiscation of property.

These drastic measures proved, however, unavailing. The exchange value of the assignats declined steadily. At the close of 1793 it was only 22% and in 1795 it had fallen to under 1%.

Not so dramatic, but not less impressive and instructive, were the experiences during the two great American monetary devaluations on the occasion of the war of liberation and of the civil war. There, too, the fiat of the State was unable to prevent devaluation.

But by far the greatest financial catastrophe of modern times was the German postwar inflation. It is common knowledge that the legal obligation to accept the banknotes of the Reichsbank could not prevent their complete collapse and it is most significant that the notes of the Rentenbank, which succeeded in stopping the inflation, have never been legal tender.

Nor was it different in the case of the Austrian and Russian inflations. Nowhere, in fact has the power of the State been able to prevent devaluation.

But this was not only the fate of weak States crushed by defeat. France and Italy, both victors in the World War, had to suffer heavy devaluations which annihilated more than four fifths of the value of their currency

It cannot be therefore the State’s fiat which confers value on inconvertible paper money.


Not even confidence and national enthusiasm, revolutionary determination and religious belief, can accomplish this in the long run. One example may suffice. When during the French Revolution in April 1793 the above mentioned currency law was promulgated,  the whole population of Metz assembled on the Place de la Légalité, took a solemn oath, in the presence of the garrison, the National Guard, and the administrative and the judiciary staff, not to draw any distinction between the face value of the paper money and silver. Similarly, from Toulon where analogous ceremonies took place, the Government received a report stating that the population would carry out the law with the religious respect (respect religieux”) due to it. However, after a few days the workmen at the arsenal of Toulon petitioned that they might have their wages paid in silver, for, they declared, “try as we may, we cannot live otherwise”. (See Marion, “Histoire financiére de la France”, vol. 3; p. 47, Paris, 1921.) Even the powers of the soul cannot, therefore, permanently confer value on paper money.


The value of inconvertible paper means of payment has a different basis, clearly revealed in the history of German finance. Thus during the nineteenth century several German States issued paper money the value of which did not lie in its convertibility, but in that the State agreed to accept at its pay offices the notes it issued at their face value, regardless of their rate of exchange. German financial science called this the taxation foundation (“Steuerfundation”).

This acceptance by the State should not be confused with the current obligation to accept, for under the regime of compulsory acceptance the taxation offices, following the universal custom, accept notes at their actual and not at their nominal value. Thus the notes of the’ Bank of England are worth no more at the English fiscal offices than anywhere else. The State accepts them at their paper value and not at the value of the gold pound. Compulsory acceptance‑ and taxation foundation are therefore fundamentally different. The basis of value of this paper money lay in that it was accepted at the fiscal offices of the issuing authority at its nominal value and, accordingly, this obligation  to accept was the important element in the wording of the warrants. Thus the Saxon fiscal notes simply read:

“In conformity with the edict of 1 October 1818, this will be accepted at the Royal fiscal offices”, and the Prussian money orders of 1835 and 1856 contain, besides the value of the order only the statement: “Of full value in all payments.”

Similarly in most other countries. It is true that in many cases, as in Baden, Austria, and Wuerttemberg, there was in addition to the obligation of the fiscal offices to accept ‑ that is, to the fiscal foundation ‑ a more or less widely current obligation to redeem the warrants. However, here also the fiscal foundation was of prime importance and redemption constituted only a kind of supplementary guaranty. This is already expressed in the order of the two undertakings on the notes. Thus we read on the Baden paper money of 1849:

“Paper money of the Grand‑Duke of Baden, which all Baden fiscal offices accept in payments at its full nominal value i.e., as equivalent to the gross silver money struck at the country’s standard of coinage ‑ and is exchangeable at sight for gross silver coins at the redemption office at Karlsruhe.”

Likewise, in the Reich Act concerning the issue of Reich fiscal office notes of 30 April 1874 par. 5, we read:
“Federal fiscal notes are accepted at their nominal value in payments made to all fiscal offices of the Reich and all constituent States. They are redeemable at any time on demand for cash by the Reich’s Central Fiscal Office on the Reich’s account”.

In both cases therefore the fiscal foundation comes first. It covers all fiscal offices, whilst the notes can only be redeemed at one fiscal office.

The Wuerttemberg Act of 1 July 849 makes this relation still plainer. Article 2, par 1, provides:

“This paper money is accepted at its nominal value in payment at all fiscal offices of the State, as also at the tax collecting offices. These offices are instructed to redeem on demand this money, so far as the available funds permit, and in amounts not under twenty gulden at a time.”

Here the claim to redeem notes is conditional on means being available. According to the clear wording of the Act, the notes are in principle only covered by the fiscal foundation.

Later, this principle was further developed in the Rentenbank notes of 1923. Here no provision at all was made for direct redemption. Apart from utilizing them in connection with public pay offices, the holder had only a claim to convert them into annuity bonds, which fact has played no important part in practice. Lastly, in recent years the fiscal foundation has been most conspicuously exemplified in the fiscal warrants of 1932, although these are not intended to be means of payment proper. They cannot be, either directly or indirectly, converted into ready money. Nor can they be exchanged for securities. No redemption fund exists nor repayments or amortization. Their value is entirely due to their being accepted by the fiscal offices at the indicated value, regardless of their exchange rate. From the Saxon pay office notes to the fiscal warrants of the Reich, the same principle of a fiscal foundation is evident.


The principle of the commercial bill for the Scottish banknotes corresponds to that, of the fiscal foundation for State paper money. Whilst English banknotes have their origin in the receipts given by the London goldsmiths for gold deposited with them (which means that redemption is of their essence), the Scottish notes have a different history. In the latter case, the banks gave in exchange for commercial bills round sums in notes of small denominations, expressing themselves at the same time ready to accept the notes they had thus issued in payment at their face value for the commercial bills they had discounted. Thus the basis of the value of the English notes was the gold deposited, whereas the value of the Scottish notes was based on goods sold as expressed in commercial bills.

It is true that the Scottish banknotes were also redeemable in precious metal, very much as was frequently the case with the State paper moneys, but redemption played an indifferent part in practice. By means of a so‑called option clause the banks frequently reserved to themselves the right of postponing the redemption of the notes several months after they had been presented. Thus they could wait until the commercial bills had matured. When on the withdrawal of the bills, the notes flowed back, they disappeared from circulation and the question of their redemption did not arise. So far as the bill debtors redeemed their debts in coin or other means of payment and not in the notes of the bank, the bank, without drawing on its reserves, acquired thereby the necessary means of redeeming any floating notes.

Ultimately in any case, the value of these notes lay in their being accepted by the bank that had issued them.


The fiscal foundation for State paper money and the principle of the commercial bill for banknotes are therefore basically related. In both instances acceptance by the issuer at their nominal value, regardless of the exchange rate of the paper, is decisive. The significance of this acceptance (or reflux) is manifest. If, for example, the State pays an official with such a warrant, if the official passes this warrant on to his baker in payment for bread, and if, lastly, the baker liquidates his tax debt with it, then the baker clears his debt to the State with the warrant that the official has passed on to him. In the last resort we have here a clearing process, i.e., a balancing of mutual obligations. And these settlements, unlike barter or modern international compensatory transactions, are not made without resorting to money. On the contrary, the exchange is operated by means of a clearing process canceling the mutual claims through our monetary system.

An inconvertible paper means of payment assumes therefore a reflux. It represents, in fact, a clearing certificate and derives its value from the exchange of economic services. This indicates consequently the limit to issues of unredeemable notes. Since the value of freely quoted money, as for instance of the Reich pay notes of 1814 or of the Rentenbank notes is, like that of every other commodity, determined by supply and demand, no more notes may be issued than there is a demand for, that is, than must flow back each time to the issuing centre. Thus in the case of State paper money, the aggregate sum to be issued must depend on the aggregate tax claims due or all but due. Within this limit issues are always justifiable. If, however, this limit is exceeded by circulating paper money representing tax claims in the distant future, depreciation will inevitably follow, even if the State should promise to accept the notes at their face value in the future.

Similarly with banknotes. In principle, only short‑term obligations should be admitted as cover. The nearer the due date of the bill, the greater is the demand for means of payment in order to redeem it and the more assured is the value of the banknotes issued in connection with the bill. The more distant the due date and the less assured the payment the more in jeopardy are the notes having such a basis. Rightly, therefore, many bank acts, among them that of German, expressly provided that only sound commercial bills falling due within three months at most may be discounted by banks of issue. Indeed experience teaches that long‑dated financial bills have hurled whole monetary systems into the abyss.


There exist, accordingly, two entirely independent and wholly different foundations whereon the value of paper means of payment may be based: redeemability in precious metal and clearing. There is no third possibility.

Thus all paper means of payment in every country may be today actually resolved into these two elements of value. Insofar as notes can be really exchanged for gold, their value may be attributed to this. So far, however, as there is no redemption and there is an inadequate metallic cover, only clearing can confer value on the notes, and this either by clearing against commercial bills or against fiscal claims. Moreover, in order to extend the facilities for clearing and thus to increase their value, many countries, at a time when banknotes were not legal tender, introduced in addition a tax foundation for the banknotes, statutorily obliging public pay offices to accept them. (See, e .g , the Federal Act concerning the Swiss National Bank of 6 October 1905, article 23.)

The difference between the two elements of value, convertibility and clearing which latter means here the clearing of taxes due, is well exemplified in the Bank of England. Here, since the Peel’s Act, the note circulation is divided into a so‑called fiduciary and non‑fiduciary issue. The latter must have a 100 gold cover. It reflects the pure idea of convertibility. The fiduciary issue however, which at present amounts to 260 million pounds, is based exclusively on “an eternal debt of the English State’. It is not covered by any redemption fund. It rather embodies the idea of a fiscal foundation.

How, then, are these two elements of value related? Although in the public mind the idea of convertibility predominates, what is actually of decisive importance for prosperity of a country is clearing.

The idea of convertibility makes the quantity of means of payment basically dependent on the gold reserve available at any time. This is an entirely impracticable principle. All attacks on the gold standard, directed against this principle, which more especially combat the creditor’s right to claim from his debtor directly or indirectly gold, are to that extent justified. For there is never a possibility of meeting all liabilities by gold payments. This holds both of international obligations and of domestic payments. It was therefore of momentous importance and evinced profound insight, when Milhaud in his great work “A Gold Truce” (Annals of Public and Cooperative Economy, Genf, 1933. The Ed.), called for a universal gold truce.

Whether a country possesses a gold reserve, depends always more or less on chance. Accordingly, the amount of the means of payment in circulation should never, not even under the most orthodox gold standard system, be determined by the size of the gold reserve. The economic life of a country would have otherwise to shrink regularly with the shrinkage of its gold reserve. In reality, the exchange of goods remains a necessity and possible, even if there is no gold reserve at all. On the other hand, as the cases of France and the United States today show, not even the largest gold reserve of a central bank can save a people from widespread unemployment and poverty, whilst a monetary system intelligently based on mutual clearing, can provide a people with work and wealth, even in the absence of any store of gold.


The diverse possibilities of issuing inconvertible means of payment based on the idea of clearing can only be outlined here. In the first place, we may mention the inconvertible and freely quoted State paper money described above. In most countries today this is to be found in a more or less disguised form. Besides States, other corporations, especially local governments, may also issue clearing notes for the imposts they are entitled to raise. Thus in the nineteenth century Hanover city issued notes which. promoted most effectively the towns prosperity.

In the economic sphere, railways enter primarily into account as centres for the issue of special railway clearing notes. In Germany there is the noteworthy case of the Leipzig‑Dresden Railway, founded by Friedrich List. This Railway issued in the thirties of the last century railway money to the amount of 500.000 thalers and this money, to the general satisfaction, freely circulated until the establishment of the Reich. After the World War, the German Federal Railway also repeatedly issued its own means of payment, most of which exhibited the character of goods warrants, i.e. they were based on the principle of clearing. Naturally, other undertakings, for whose goods or services there is a general and constant demand, would also benefit by such facilities.

The clearing principle is most particularly useful in international settlements. Thus leading firms might issue purchasing certificates. For instance, certificates accepted at their face value by the I. G. Farben Company or by Siemens, could be disposed of in London, Cape Town, and generally. To make the certificates more widely acceptable, whole groups of undertakings concerned with agriculture, export, or the tourist traffic might agree to issue purchasing certificates jointly. Issues might also be undertaken by special foreign trade banks, whose clients would bind themselves to accept the certificates in payments up to a certain amount. Lastly work provision banks might similarly be established to combat unemployment within countries. For particulars on this subject the reader is referred to the valuable works of Milhaud and Beckerath published by the Annals of Public and Cooperative Economy in 1933 and 1934.

The central banks existing at present in most countries ought not to oppose the issue of such means of payment. In this connection we need not examine here whether these monopoly banks have fulfilled the hopes placed in them or have not rather aggravated all financial catastrophes, such as inflations and deflations. In any case so long as private enterprise exists at all, mutual clearing cannot be monopolized by a single central undertaking. The idea of a monopoly is most closely associated with the idea of convertibility. The issue of clearing warrants, which neither affect the store of gold nor are able, since they are freely quoted, to modify the average price level, cannot in principle be restricted to central banks. It ought rather, within the limits drawn by the State, to be allowed to develop freely.


Once it is recognized that the value of inconvertible paper money depends on the possibility of clearing it with the issuer, the way is open to abolish the pernicious compulsory acceptance and to reestablish the gold standard. This, in turn, would facilitate the abrogation of foreign exchange legislation, for this, too, is based in the last resort on the idea of convertibility and of compulsory acceptance.

In every country foreign exchange legislation is in the main identical. The endless number of laws, regulations, and principles maybe reduced to the following three aims:

  1. a) Retention of gold;
  2. b) Retention of foreign means of payment (foreign exchange proper); and
  3. c) Restriction of payments abroad.

Compared with these, all other provisions are of secondary importance. And these three aims may be reduced to one, namely the seizure of gold. The retention of foreign means of payment and the restriction of payments abroad ‑ are only means towards attaining that one object. The foreign means of payment are seized because they are regarded as substitutes for gold and because it is anticipated that by their redemption or, at least, by their being sold on the international market, gold might be obtained in exchange. Inversely, payments abroad are restricted as far as possible because it is feared that through the efflux of means of payment the gold reserve might be drained and the maintenance of the parity be thereby endangered. The idea that gold is not only the standard of value but ultimately the sole and supreme means of payment lies therefore at the root of this type of legislation.

Hence the efflux of gold and the associated threat to the redemption fund have been invariably the direct cause of the enacting of foreign exchange legislation. Here also the compulsion to accept the notes of the central bank plays a special part. Thus whilst the State may leave to their fate the freely quoted notes of a private bank, without the monetary unit being affected in case the bank should suffer exchange losses, the statutory obligation to accept the notes of the central bank implies that they are identified with the country’s standard of value. A loss in exchange in the case of the latter involves therefore a modification in the monetary basis itself. Accordingly, the State because of its self­imposed equalization is compelled to maintain the parity of the notes of the central bank as long as it is bent on saving its standard of value from fluctuations.

The identification of the monetary unit with the notes of the central bank has, lastly, created another source of danger which has repeatedly acted as a decisive factor in the introduction of foreign exchange legislation, namely its close association with certain large‑scale banks at home. Thus the collapse of the Kreditanstalt in Austria and the Darmstadt Bank in Germany instigated the introduction of foreign exchange legislation in those countries. Of itself there existed no direct connection between these banks and the monetary standard, and the example of Sweden, which did not rush to the aid of the collapsed Krueger undertakings, shows that in such emergencies other methods may be applied than those chosen by Austria and Germany. However where through the obligation to accept the notes of a given bank a statutory bridge has been built between the banking system and the national currency, the temptation will always exist to shift the liquidity difficulties of the banks onto the shoulders of the currency.

All these problems assume a very different complexion if we take inconvertible means of payment to be what they really are, namely means of clearing in relation to an issuer. A fundamentally inconvertible note, the value of which lies exclusively in the issuer accepting it and which therefore really involves no claim to payment in gold but a claim to the services of the issuer, can never lead to a reduction in the gold reserve. A State paper money based on such principles could be freely allowed to go abroad without this prejudicially affecting the currency. It could never entail liabilities in precious metal. On the contrary, it would necessarily lead to goods being exported and may be accordingly even followed by an influx of gold or foreign exchange.

What matters is that the means of payment shall not be legal tender; in other words, that it must be accepted only by the issuer at its face value. With this condition satisfied and the above mentioned limit to issues being respected, even the heaviest losses on foreign bourses would involve no danger; for the lower the market rate falls, the greater the temptation would be to acquire the warrants, inasmuch as the issuer has to redeem them at their nominal value, the exchange rate difference being thus converted into a gain. Everybody therefore who has to meet his obligation with these warrants ‑ in our example the taxpayer, will try to benefit by such falls and thereby produce a steady demand which has the peculiarity of rising as the market rate falls. This necessary demand provides the suction power that imparts value to the inconvertible paper money. Such a means of payment need not dread the throwing open of frontiers. In the words of a Swiss writer, it will, “like a carrier pigeon”, return always to its point of departure and necessarily lead to a demand for home products abroad, in this way furthering the export trade.

Since such a clearing warrant has by definition a free market rate and is therefore not linked to the currency unit, the latter cannot be affected by any fluctuations in the value of the former.

For clearing warrants of the type described, foreign exchange legislation ceases to have any meaning, for any stipulations relating to convertible money would lose their point. In all countries, therefore, sound clearing warrants should be freed from their fetters, since they were only imposed on them on account of the unsound ones. Thus one should at last bestow on our time the freedom of movement denied it today because of the past. That is, foreign exchange legislation should be abrogated at least for sound warrants. Given this first step and the basis for a new economic structure is laid. Further progress will be thereby facilitated, for in most countries the inconvertible notes could be without difficulty converted into State paper money, thus investing the present state of things in this matter with its proper form.

Should however, one or another central bank have transgressed the limits that would apply to a fiscal foundation, it will not be difficult to remedy this, without fettering a country’s general economic life by foreign exchange legislation.

Nor can it be objected that the national economy requires foreign exchange and can therefore not be satisfied with clearing transactions, for to begin with, we should remember that probably in all countries depending on foreign exchange the supply of the latter has shrunk despite the most stringent legislation and that therefore the object aimed at has not been fully attained. But a more important point is that both commercial and financial debts, according to firmly and universally established views, can only be liquidated by the goods or services of the debtor, and it is just these that are offered by clearing warrants. It must be therefore possible to start again payment operations on this basis. So far as the importing of raw materials is in question this cannot be challenged. A clearing warrant, unhampered by foreign exchange legislation and made out in gold units is a fully utilizable means of payment so long as there is a demand on the world market for the goods and services of the issuing country. If this condition is not satisfied, then foreign exchange legislation also would be of no avail. If an exchange of goods is actually impracticable, ‑ that is, if the country no longer plays a part in the world economy, ‑ no debt can possibly be liquidated, This holds equally of commercial and financial debts. No one can expect that a country possessing no gold should pay in gold. Every creditor should therefore recognize that a debtor can only offer a lien on (a right to -The Ed.) his goods and services.

Years ago the Economic Committee of the League of Nations expressed the same idea when it said that creditor countries must either agree that debtor countries may directly or indirectly redeem their obligations in goods or services or they must be resigned to not receiving any payments. The clearing warrants described here indicate the way in which even heavy liabilities may be in time honestly liquidated, to the common advantage of creditors and debtors.

Once the banknotes that had become inconvertible have been superseded by clearing warrants, the free gold market can be at last reestablished. As soon as gold has been divested of the property ascribed to it of being the exclusive means of payment and representing the core of all money claims, it becomes again a commodity like any other. Its free movement no longer causes alarm and may therefore the better fulfill the function of a standard of value. And this completes the circle of our proposals, for the free gold market which here appears as the result of clearing, at the same time presupposes it; since the various means of payment can only be reliably valued when gold may be freely moved.

It follows, lastly, that under the system above outlined there would be no objection to gold coins being minted for private firms and this not, as seems to be the intention at present in France, as a form of note cover, but for actual and immediate circulation with a view to measuring by them at any time the value of all other means of payment.


Accordingly, we propose

  1. The introduction of unambiguously determined gold units of account as a monetary basis
    (e.g. 1 mark – 1/2790 kg. fine gold, 1 Franc – y kg. fine gold. 1 L – z kg. fine gold)
  2. The transformation of the whole monetary system on the basis of a gold unit of account.
  3. The abolition of the statutory obligation of acceptance for the notes of the central banks.
  4. The removal of the monopoly of the central banks and its replacement by general regulations concerning the issue of means of payment.
  5. The abrogation of foreign exchange legislation and the reestablishment of a free gold market.
  6. The unrestricted right to mint gold coins.


  1. a) Through international agreements

Everybody is agreed that something should be done to remove the monetary chaos and the obstacles to settlement operations. Here also hope is commonly centered on international conferences which are expected to lead to an improvement. However, the value of such conferences for the solution of economic problems is trifling. Think in this connection of the World Economic Conference of 1933. This was eagerly looked forward to by all peoples, and yet, notwithstanding careful preparation, produced virtually no practical results,

However even if within the near future a monetary conference could be arranged to meet and if, moreover, full agreement could be reached among the participating States, the effect would be at best to reestablish the parity between the monetary units and to impose an unconditional obligation on the parties not to modify deliberately this parity nor to change it without the consent of the other contracting parties. But that would only dispose of a comparatively small part of the present difficulties, for only most rarely ‑ as in the United States recently ‑ has the abandonment of the gold  standard been an arbitrary act. In by far the majority of instances, the gold standard was most reluctantly dropped. Heavy calls on credit institutes, excessive claims on the central bank; a rising efflux of gold, and, finally the threatening impossibility, despite all efforts, of maintaining the parity of the banknotes, were for the most part the really decisive factors in the abandonment of the gold standard. For such cases an international agreement on monetary parities would offer no solution.

So long as the principle of central banks is retained and their notes, being made legal tender, are identified with the standard of value, no international agreement will be able to prevent the recurrence of present day conditions, for under this system the collapse of a single leading bank, to say nothing of the collapse of the State finances, may render impossible the maintenance of parity. However, even if, to meet such cases, the several countries concluded a convention providing that the gold reserves of the diverse central banks should be automatically mobilized to save the currency of a country in need, a scarcely credible supposition, even this would be insufficient in hard cases, for the magnitude of the monetary obligations maturing at any given moment would probably always exceed the quantity of available gold.

It also remains a moot point how the problem of foreign exchange legislation could be settled at such a conference, seeing that the granting of loans, useful as these would be to those immediately concerned, in no way solves this problem. And yet without this, even a fresh determination of parities would prove unsatisfactory. On the contrary, the object should be to rebuild on a sound basis the entire system of international payment arrangements.

  1. b) Through intrastate legislation

From what precedes follows the possibility that each separate State, having made up its mind, may, by abolishing the compulsion to accept and by recognizing the reflux principle, could reestablish for itself the gold standard and rescind its foreign exchange legislation; This may seem fantastic. But the reader should remember the German inflation period. Then, too, an escape without external assistance appeared impossible. “The hole in the West”, speculation on foreign bourses, and similar obstacles, which Germans could not control, were quite generally regarded as the causes of the depreciation of the mark. Eventually, when things were at their worst, when in some parts of the Reich serious disorders had broken out, and when external help was out of the question, the country succeeded, by its own efforts, without the aid of any foreign Government, without an international conference or convention, to stabilize the mark, to safeguard it against all foreign speculation, and even after a few months to abrogate the foreign exchange legislation then in force. Today, as at that time, the free resolve of any country may determine the fate of its currency.

The country that first finds the way out, that removes insecurity, and at the same time reestablishes the freedom of monetary operations, will march well ahead of the other countries. It may even be assumed that a mighty stream of capital for long‑term investment will pour into it from all sides.

  1. c) Through private initiative

Even if, however, no State were for the moment ready to proceed along this line, there remains the possibility of finding a way out of the monetary chaos through private initiative or at least to prepare the way for this. When in the eighteenth century the national monetary units, because of alleged State needs, continually fluctuated and when it was therefore impossible to rely on the value of currencies even for a moderate time, Hamburg merchants, more particularly, discovered a way out. By founding the famous Hamburger Girobank, they, following the centuries’ old Chinese Tael system, made themselves independent of the debased State coinage, by adopting as the basis of all their accounts an unminted definite weight of silver in the place of State money. This weight called Mark Banko, constituted an unchangeable unit of calculation, which came to be of the greatest service economically for the whole of Northern Europe.

There can be no doubt that the application of the same principle, with gold instead of silver as the basis naturally, might exercise a similarly far­reaching influence today. Just imagine that before the devaluation of the Belgian franc, a Belgian undertaking of good, repute had declared itself ready to take up a long‑term loan at its actual gold value, regardless of the exchange rate of the franc. The crowd of would‑be investors would probably have broken through all barriers.

In this connection the proposed private calculation in gold should on no account be linked with the national monetary system. Accordingly, I do not suggest here the appending of gold clauses (gold dollar gold pound, or gold franc) to a country’s currency unit, for experience teaches that such clauses ‑ which hold permanently before the statutory means of payment its ideal as in a mirror ‑ only rarely survive the fate of the national currency. Thus in Belgium, when the franc was devaluated, the gold clauses were revoked and only contracts in foreign currencies were left untouched. Private calculation in gold, like the erstwhile Mark Banko system, should be independent of the national currency. It should be therefore based on a separate unit, for which a simple calculation in grams of fine gold suggests itself here. This would also serve as a basis for an international calculating unit, which has been widely desiderated for many decades.

All that would be demanded of the State is not to prohibit calculating in gold. The State could never suffer through this, for even if a Government should deem it expedient to devaluate the national currency, as was frequently the case already in the seventeenth and eighteenth centuries, there can be no ground for imposing a devaluated currency unit on those who had freely agreed to use for their economic transactions among themselves another and constant value basis. Investors in all countries would joyfully welcome such a solution and there are not a few undertakings which, under these conditions would be decidedly ready to place long‑term loans at moderate interest rates. Once it is demonstrated however, that the monetary chaos, which seems to most men the work of an inscrutable fate, may be overcome, the profoundly beneficent effects of this for the peoples of the world will not fail to become apparent.

#     #     #