Chapter III: The Theory of Money and of the Foreign Exchanges (2)

The argument, however, is often carried further than this, and it is supposed that, if a country’s budget, currency, foreign trade, and its internal and external price levels are properly adjusted, then, automatically, its foreign exchange will be steady.[36] So long, therefore, as the exchanges fluctuate--thus the argument runs--this in itself is a symptom that an attempt to stabilise would be premature. When, on the other hand, the basic conditions necessary for stabilisation are present, the exchange will steady itself. In short, any deliberate or artificial scheme of stabilisation is attacking the problem at the wrong end. It is the regulation of the currency, by means of sound budgetary and bank-rate policies, that needs attention. The proclamation of convertibility will be the last and crowning stage of the proceedings, and will amount to little more than the announcement of a _fait accompli_.

[36] Dr. R. Estcourt, criticising one of my articles in _The
Annalist_ for June 12, 1922, writes: “The arrangement
would not last for any appreciable period unless, as a
preliminary, the Governments took the necessary steps to
balance their budgets. If that were done, the so-called
stabilisation speedily would become unnecessary; exchange
would stabilise itself at pre-war rates.” This passage puts
boldly an opinion which is widely held.

There is a certain force in this mode of reasoning. But in one important respect it is fallacious.

Even though foreign trade is properly adjusted, and the country’s claims and liabilities on foreign account are in equilibrium over the year as a whole, it does not follow that they are in equilibrium every day. Indeed, it is well known that countries which import large quantities of agricultural produce do not find it convenient, if they are to secure just the quality and the amount which they require, to buy at an equal rate throughout the year, but prefer to concentrate their purchases on the autumn period.[37] Thus, quite consistently with equilibrium over the year as a whole, industrial countries tend to owe money to agricultural countries in the second half of the year, and to repay in the first half. The satisfaction of these seasonal requirements for credit with the least possible disturbance to trade was recognised before the war as an important function of international banking, and the seasonal transference of short-term credits from one centre to another was carried out for a moderate commission.

[37] Whilst the fact of seasonal pressure is well ascertained,
the exact analysis of it is a little complicated. Food
arrivals into Great Britain, for example, are nearly 10
per cent heavier in the third and fourth quarters of the
year than in the first and second, and reach their maximum
in the fourth quarter. (These and the following figures
are based on averages for the pre-war period 1901–1913
worked out by the Cambridge and London Economic Service).
Raw material imports are more than 20 per cent heavier
in the fourth and first quarters than in the second and
third, and reach their maximum in the three months November
to January. Thus the fourth quarter of the year is the
period at which there are heavy imports of both food and
raw materials. Manufactured exports, on the other hand,
are distributed through the year much more evenly, and are
about normal during the last quarter. Allowing for the fact
that imports are paid for, generally speaking, before they
arrive, these dates correspond pretty closely with the date
at which seasonal pressure is actually experienced by the
dollar-sterling exchange. In France, since the war, imports
in the last quarter of the year seem to have been quite 50
per cent heavier than, for example, in the first quarter.
In Italy the third quarter seems to be the slackest, and
the last quarter, again, a relatively heavy period. When we
turn to the statistics for the United States we find the
other side of the picture. August and September are the
months of heavy wheat export; October to January those of
heavy cotton export. The strength of the dollar exchanges
in the early autumn is further increased by the financial
pressure in the United States during the crop-moving
period, which leads to a withdrawal of funds from foreign
centres to New York.

It was possible for this service to be rendered cheaply because, with the certainty provided by convertibility, the price paid for it did not need to include any appreciable provision against risk. A somewhat higher rate of discount in the temporarily debtor country, together with a small exchange profit provided by the slight shift of the exchanges within the gold points, was quite sufficient.

But what is the position now? As always, the balance of payments must balance every day. As before, the balance of trade is spread unevenly through the year. Formerly the daily balance was adjusted by the movement of bankers’ funds, as described above. But now it is no longer a purely bankers’ business, suitably and sufficiently rewarded by an arbitrage profit. If a banker moves credits temporarily from one country to another, he cannot be certain at what rate of exchange he will be able to bring them back again later on. Even though he may have a strong opinion as to the probable course of exchange, his profit is no longer definitely calculable beforehand, as it used to be; he has learnt by experience that unforeseen movements of the exchange may involve him in heavy loss; and his prospective profit must be commensurate with the risk he runs. Even if he thinks that the risk is covered actuarially by the prospective profit, a banker cannot afford to run such risks on a large scale. In fact, the seasonal adjustment of credit requirements has ceased to be arbitrage banking business, and demands the services of speculative finance.

Under present conditions, therefore, a large fluctuation of the exchange may be necessary before the daily account can be balanced, even though the annual account is level. Where in the old days a banker would have readily remitted millions to and from New York, hundreds of thousands are now as much as the biggest institutions will risk. The exchange must fall (or rise, as the case may be) until either the speculative financier feels sufficiently confident of a large profit to step in, or the merchant, appalled by the rate of exchange quoted to him for the transaction, decides to forgo the convenience of purchasing at that particular season of the year, and postpones a part of his purchases.

The services of the professional exchange speculator, being discouraged by official and banking influences, are generally in short supply, so that a heavy price has to be paid for them, and trade is handicapped by a corresponding expense, in so far as it continues to purchase its materials at the most convenient season of the year.

The extent to which the exchange fluctuations which have troubled trade during the past three years have been seasonal, and therefore due, not to a continuing or increasing disequilibrium, but merely to the absence of a fixed exchange, is not, I think, fully appreciated.

During 1919 there was a heavy fall of the chief European exchanges due to the termination of the inter-Allied arrangements which had existed during the war. During 1922 there was a rise of the sterling exchange, which was independent of seasonal influences. During 1923 there has been a further non-seasonal collapse of the franc exchange due to certain persisting features of France’s internal finances and external policy. But the following table shows how largely _recurrent_ the fluctuations have been during the four years since the autumn of 1919:--

PERCENTAGE OF DOLLAR PARITY

+------------+------------------+------------------+------------------+ |August-July.| Sterling. | Francs. | Lire. | | | Lowest. Highest. | Lowest. Highest. | Lowest. Highest. | +------------|------------------+------------------+------------------+ | 1919–1920 | 69 88 | 31 66 | 22 56 | | 1920–1921 | 69 82 | 30 45 | 18 29 | | 1921–1922 | 73 92 | 37 48 | 20 28 | | 1922–1923 | 90 97 | 29 41 | 20 27 | +------------+------------------+------------------+------------------+

On the experience of the past three years, francs and lire are at their best in April and May and at their worst between October and December. Sterling is not quite so punctual in its movements, the best point of the year falling somewhere between March and June and the worst between August and November.

The comparative stability of the highest and lowest quotations respectively in each year, especially in the case of Italy, is very striking, and indicates that a policy of stabilisation at some mean figure might have been practicable; whilst, on the other hand, the wide divergences between the highest and lowest are a measure of the expense and interference that trade has suffered.

These results correspond so closely to the facts of seasonal trade (see above, p. 108) that we may safely attribute most of the major fluctuations of the exchanges from month to month to the actual pressure of trade remittances, and not to speculation. Speculators, indeed, by anticipating the movements tend to make them occur a little earlier than they would occur otherwise, but by thus spreading the pressure more evenly through the year their influence is to diminish the absolute amount of the fluctuation. General opinion greatly overestimates the influence of exchange-speculators acting under the stimulus of merely political and sentimental considerations. Except for brief periods the influence of the speculator is washed out; and political events can only exert a lasting influence on the exchanges, in so far as they modify the internal price level, the volume of trade, or the ability of a country to borrow on foreign markets. A political event, which does not materially affect any of these facts, cannot exert a lasting effect on the exchanges merely by its influence on sentiment. The only important exception to this statement is where there exists on a large scale a long-period speculative investment in a country’s currency on the part of foreigners, as in the case of German marks. But such investments are comparable to borrowing abroad and exercise a different kind of influence altogether from a speculative transaction proper, which is opened with the intention of its being closed again within a short period. And even speculative investment in a currency, since it is bound to diminish sooner or later, cannot permanently prevent the exchanges from reaching the equilibrium justified by conditions of trading and relative price levels.

It follows that, whilst purely seasonal fluctuations do not interfere with the forces which determine the ultimate equilibrium of the exchanges, nevertheless stability of the exchange from day to day cannot be maintained merely by the _fact_ of stability in these underlying conditions. It is necessary also that bankers should have a sufficiently certain _expectation_ of such stability to induce them to look after the daily and seasonal fluctuations of the market in return for a moderate commission.

After recent experience it is unlikely that they will actually entertain any such expectation, even if the underlying facts were of a kind to justify it, with sufficient conviction to act, unless it is backed up by a guarantee on the part of the Central Authority (Bank or Government) to employ all their resources for the maintenance of the level of exchange at a stated figure. At present the declared official policy is to bring the franc and the lira (for example) back to par, so that operations favouring a fall of these currencies are not free from danger. On the other hand no steps are taken to make this policy effective, and the conditions of internal finance in France and Italy indicate that their exchanges may go much worse. Thus, since no one can have complete confidence whether they are to be a great deal better or very much worse, there must be a wide fluctuation before financiers will come in, purely from motives of self-interest, to balance the day-to-day fluctuations and the month-to-month fluctuations round about the unpredictable point of equilibrium.

If, therefore, the exchanges are not stabilised by policy, they will never come to an equilibrium of themselves. As time goes on and experience accumulates, the oscillations may be smaller than at present. Speculators may come in a little sooner, and importers may make greater efforts to spread their requirements more evenly over the year. But even so, there must be a substantial difference of rates between the busy season and the slack season, until the business world knows for certain at what level the exchanges in question are going to settle down. Thus a seasonal fluctuation of the exchanges (including the sterling-dollar exchange) is inevitable, even in the absence of any decided long-period tendency of an exchange to rise or to fall, unless the Central Authority, by a guarantee of convertibility or otherwise, takes special steps to provide against it.

IV. _The Forward Market in Exchanges._

When a merchant buys or sells goods in a foreign currency the transaction is not always for immediate settlement by cash or negotiable bill. During the interval before he can cover himself by buying or selling (as the case may be) the foreign currency involved, he runs an exchange risk, losses or gains on which may often, in these days, swamp his trading profit. He is thus involuntarily engaged in a heavy risk of a kind which it is hardly in his province to undertake. The subject of what follows is a piece of financial machinery--namely, the market in “forward” exchanges as distinguished from “spot” exchanges--for enabling the merchant to avoid this risk, not, indeed, during the interval when he is negotiating the contract, but as soon as the negotiation is completed.

Transactions in “spot” exchange are for cash--that is to say, cash in one currency is exchanged for cash in another currency. But merchants who have bought goods in terms of foreign currency for future delivery may not have the cash available pending delivery of the goods; whilst merchants who have sold goods in terms of foreign currency, but are not yet in a position to sell a draft on the buyer, cannot, even if they have plenty of cash in their own currency, protect themselves by a “spot” sale of the exchange involved, save in the exceptional case when they have cash available in the foreign currency also.

A “forward” contract is for the conclusion of a “spot” transaction in exchanges at a later date, fixed on the basis of the spot rate prevailing at the original date. Pending the date of the maturity of the forward contract no cash need pass (although, of course, the contracting party may be required to give some security or other evidence for his ability to complete the contract in due course), so that the merchant entering into a forward contract is not required to find cash any sooner than if he ran the risk on the exchange until the goods were delivered; yet he is protected from the consequences of any fluctuation in the exchanges in the meantime.

The tables given below show that in London, in the case of the exchanges which have a big market (the dollar, the franc, and the lira), competition between dealers has brought down the charges for these facilities to a fairly moderate rate. During 1920 and 1921 the cost to an English buyer of foreign currency for forward delivery was a little more expensive than for spot delivery in the case of francs, lire, and marks, and a little cheaper in the case of dollars. Correspondingly, French, Italian, and German merchants were generally in a position to buy both sterling and dollars for forward delivery at a slightly cheaper rate than for spot delivery--that is to say, if they dealt in London. As regards the rates charged in foreign centres my information is not extensive, but it indicates that in Milan, for example, very much less favourable terms for these transactions are frequently charged to the seller of forward sterling than those ruling in London. During 1922, however, the effect of the progressive cheapening of money in London was, for reasons to be explained in a moment, to cheapen the cost to English buyers of foreign currency for forward delivery, forward francs falling to an appreciable discount on spot francs, and forward dollars becoming at the end of the year decidedly cheaper than spot dollars. Later on, the raising of the bank-rate in June 1923 acted again, as could have been predicted, in the opposite direction.

Proceeding to details, we see below (pp. 118, 119) the quotations for forward exchange ruling in the London market since the beginning of 1920. During 1920–21 forward dollars were generally cheaper than spot dollars to a London buyer to the extent of from 1 to 1½ per cent per annum. Occasionally, however, when big movements of the exchange were taking place, the discount on forward dollars was temporarily much higher, having risen, for example, in November 1920, when sterling was at its lowest point, to nearly 6 per cent--for reasons which I will endeavour to elucidate later. During the first half of 1922 the discount on forward dollars dwindled, but rose again during the latter half of the year, reacting again in the middle of 1923 after money rates in London had been slightly raised. Thus a London merchant, who has had dollar commitments for the purchase of goods, has not only been able to cover his exchange risk by means of a forward transaction, but on the average he has got his exchange a little cheaper by providing for it in advance.

TABLE OF EXCHANGE QUOTATIONS IN LONDON ONE MONTH FORWARD[38]

+--------------------------------------------------+ | NEW YORK. | +----------+------------+-------------+------------+ | | | One Month | Difference | | Date. | Spot. | Forward. | per cent | | | | | per annum. | +----------+------------+-------------+------------+ | 1920 | | | | |January | 3·79 | + ⅜ cent | +1·2 | |February | 3·48⅞ | + ¼ „ | + ·9 | |March | 3·41⅜ | + ¼ „ | + ·9 | |April | 3·90¾ | + ⅜ „ | +1·2 | |May | 3·82⅞ | + ½ „ | +1·6 | |June | 3·89-15/16 | + ⅜ „ | +1·2 | |July | 3·96⅛ | + ⅝ „ | +1·9 | |August | 3·67 | + ½ „ | +1·6 | |September | 3·56⅞ | + ½ „ | +1·7 | |October | 3·48-5/16 | + ½ „ | +1·7 | |November | 3·44⅜ | +1⅝ „ | +5·7 | |December | 3·49 | + ½ „ | +1·7 | | 1921 | | | | |January | 3·58⅜ | + ⅜ „ | +1·3 | |February | 3·84¾ | +1 „ | +3·1 | |March | 3·88⅜ | + ⅞ „ | +2·7 | |April | 3·92 | + ⅜ „ | +1·1 | |May | 3·98 | + ½ „ | +1·5 | |June | 3·90⅝ | + ¾ „ | +2·3 | |July | 3·71-15/16 | + ⅝ „ | +2·0 | |August | 3·56⅜ | + ½ „ | +1·7 | |September | 3·71⅝ | + ⅜ „ | +1·2 | |October | 3·76⅛ | + ½ „ | +1·6 | |November | 3·92-1/16 | + ⅞ „ | +2·7 | |December | 4·08-5/16 | + ⅜ „ | +1·1 | | 1922 | | | | |January | 4·20½ | + ⅛ „ | + ·4 | |February | 4·30½ | par | ... | |March | 4·42 | „ | ... | |April | 4·39 | „ | ... | |May | 4·44½ | „ | ... | |June | 4·46¾ | + 3/16 cent | + ·5 | |July | 4·44¾ | + 1/16 „ | + .17 | |August | 4·45¼ | + 3/16 „ | + .5 | |September | 4·46 | + ⅜ „ | +1 | |October | 4·42 | + ¼ „ | + .68 | |November | 4·46½ | + ⅝ „ | +1·68 | |December | 4·51¾ | +1 „ | +2·65 | | 1923 | | | | |January | 4·64¾ | +1¼ „ | +3·23 | |February | 4·67 | + ⅞ „ | +2·25 | |March | 4·70⅝ | +1 „ | +2·55 | |April | 4·66⅞ | + ¾ „ | +1·93 | |May | 4·62½ | + 15/16 „ | +2·43 | |June | 4·62¾ | + ⅞ „ | +2·27 | |July | 4·56½ | + ½ „ | +1·31 | |August | 4·57 | + ¼ „ | +0·66 | +----------+------------+-------------+------------+

+--------------------------------------------------+ | PARIS. | +----------+------------+-------------+------------+ | | | One Month | Difference | | Date. | Spot. | Forward. | per cent | | | | | per annum. | +----------+------------+-------------+------------+ | 1920 | | | | |January | 40·90 | + 6 centime | +1·7 | |February | 46·90 | + 4 „ | +1·0 | |March | 48·55 | + 3 „ | + ·7 | |April | 57·80 | + 3 „ | + ·6 | |May | 64·04 | + 1 „ | + ·18 | |June | 50·45 | - 5 „ | -1·2 | |July | 47·05 | -10 „ | -2·8 | |August | 49·00 | -10 „ | -2·4 | |September | 51·22½ | - 5 „ | -1·2 | |October | 52·10 | -10 „ | -2·3 | |November | 54·45 | -15 „ | -3·3 | |December | 57·45 | -15 „ | -3·2 | | 1921 | | | | |January | 61·07½ | -30 „ | -5·9 | |February | 54·50 | -20 „ | -4·4 | |March | 54·40 | -27 „ | -5·9 | |April | 55-37½ | -15 „ | -3·3 | |May | 50·22½ | -12 „ | -2·9 | |June | 46·35 | -10 „ | -2·6 | |July | 46·72½ | -10 „ | -2·6 | |August | 46·77½ | + 2 „ | + ·5 | |September | 48·68½ | + 3 „ | + ·7 | |October | 52·27½ | + 1 „ | + ·2 | |November | 53·44 | + 4 „ | + ·9 | |December | 54·24 | + 2 „ | + ·4 | | 1922 | | | | |January | 52·32½ | par | ... | |February | 51·62½ | „ | ... | |March | 48·45 | „ | ... | |April | 48·15 | - 1 centime | - .25 | |May | 48·47 | + 1 „ | + .25 | |June | 49·00 | + 2 „ | + ·49 | |July | 56·20 | + 8 „ | +1·8 | |August | 54·10 | +10 „ | +2·21 | |September | 57·40 | + 3 „ | + ·63 | |October | 58·25 | + 3 „ | + ·62 | |November | 64·65 | +14 „ | +2·59 | |December | 64·30 | + 8 „ | +1·49 | | 1923 | | | | |January | 66·40 | + 5 „ | + ·9 | |February | 75·50 | +16 „ | +2·54 | |March | 77·50 | +11 „ | +1·70 | |April | 70·40 | + 5 „ | + .85 | |May | 69·35 | + 5 „ | + ·86 | |June | 71·60 | + 5 „ | + ·84 | |July | 78·35 | + 4 „ | + ·61 | |August | 79·20 | + 9 „ | + ·60 | +----------+------------+-------------+------------+

First day of month in 1920, first Wednesday in 1921, and first Friday thereafter.

TABLE OF EXCHANGE QUOTATIONS IN LONDON ONE MONTH FORWARD

+--------------------------------------------------+ | ITALY. | +----------+------------+-------------+------------+ | | | One Month | Difference | | Date. | Spot. | Forward. | per cent | | | | | per annum. | +----------+------------+-------------+------------+ |1920[38] | | | | |January | 50 | - ⅛ lire | - 3·0 | |February | 55 | - ⅛ „ | - 2·7 | |March | 62¾ | - ¼ „ | - 4·7 | |April | 80½ | - ¼ „ | - 3·7 | |May | 83 | - ½ „ | - 7·1 | |June | 66⅜ | - ½ „ | - 9·1 | |July | 65⅜ | - ½ „ | - 9·2 | |August | 70 | - ½ „ | - 8·5 | |September | 76¼ | - ½ „ | - 7·9 | |October | 83-9/16 | - ½ „ | - 7·2 | |November | 93-11/16 | - ½ „ | - 6·4 | |December | 94-11/16 | - ½ „ | - 6·3 | | 1921 | | | | |January | 104⅜ | par | ... | |February | 105½ | - ¾ lire | - 8·5 | |March | 106½ | - ⅝ „ | - 7·0 | |April | 92¼ | - ½ „ | - 6·5 | |May | 81⅜ | - ⅝ „ | - 9·1 | |June | 73-11/16 | - ½ „ | - 8·1 | |July | 77 | - ½ „ | - 7·8 | |August | 85-1/16 | - ¼ „ | - 3·5 | |September | 85-9/16 | - ⅜ „ | - 5·2 | |October | 94⅛ | - ⅜ „ | - 4·8 | |November | 96⅝ | - ¼ „ | - 3·1 | |December | 93-15/16 | - ½ „ | - 6·4 | | 1922 | | | | |January | 97⅛ | - ¼ „ | - 3·0 | |February | 92½ | - 7/16 „ | - 5·7 | |March | 83-3/16 | - ¼ „ | - 3·6 | |April | 83-5/16 | -15 pts. | - 2·16 | |May | 83 | -10 „ | - 1·45 | |June | 85⅞ | - 3 „ | - ·41 | |July | 100 | par | ... | |August | 96 | par | ... | |September | 101 | -11 „ | - 1·31 | |October | 103 | -10 „ | - 1·16 | |November | 106 | - 8 „ | - ·91 | |December | 93¾ | -20 „ | - 2·56 | | 1923 | | | | |January | 92 | -11 „ | - 1·43 | |February | 97½ | -23 „ | - 2·83 | |March | 97⅜ | -23 „ | - 2·82 | |April | 93¾ | -18 „ | - 2·30 | |June | 99 | -15 „ | - 1·82 | |July | 106⅞ | -22 „ | - 2·47 | |August | 105½ | -28 „ | - 3·18 | +----------+------------+-------------+------------+

+-------------------------------------------------------+ | GERMANY. | +----------+------------+------------------+------------+ | | | One Month | Difference | | Date. | Spot. | Forward. | per cent | | | | | per annum. | +----------+------------+------------------+------------+ |1920[38] | | | | |January | 187 | marks | | |February | 305 | | | |March | 337 | | | |April | 275 | | | |May | 218½ | - 1 „ | - 5·5 | |June | 150½ | - 1 „ | - 8·0 | |July | 150 | - ½ „ | - 4·0 | |August | 160½ | - 1 „ | - 7·5 | |September | 176 | - ½ „ | - 3·4 | |October | 215 | - 1 „ | - 5·6 | |November | 266½ | - ½ „ | - 2·2 | |December | 241½ | - 1 „ | - 4·9 | | 1921 | | | | |January | 269½ | - 2 „ | - 8·9 | |February | 243½ | - 1 „ | - 4·9 | |March | 24½ | - 1 „ | - 4·9 | |April | 239½ | - 2 „ | -10·0 | |May | 262½ | - 1¾ „ | - 8·0 | |June | 245¼ | - 1½ „ | - 7·3 | |July | 279½ | - 1½ „ | - 6·45 | |August | 286 | - 1 „ | - 4·2 | |September | 347½ | - 1½ „ | - 5·1 | |October | 471 | - 5 „ | -12·7 | |November | 764½ | - 2¼ „ | - 3·5 | |December | 855 | - 1½ „ | - 2·1 | | 1922 | | | | |January | 777½ | - 3½ „ | - 5·4 | |February | 872 | - 2½ „ | - 3·4 | |March | 1117 | - 1½ „ | - 1·6 | |April | 1440 | - 8 „ | - 6·6 | |May | 1270 | - ½ „ | - ·47 | |June | 1222 | par | ... | |July | 2320 | + 5 marks | + 2·59 | |August | 3175 | +20 „ | + 7·56 | |September | 5700 | nominal | ... | |October | 9900 | + 450 mks | + 54·54 | |November | 26,250 | + 6,000 „ | +274·3 | |December | 35,000 | + 5,500 „ | +188·58 | | 1923 | | | | |January | 39,500 | + 1,750 „ | + 53·16 | |February | 190,000 | + 27,000 „ | +170·53 | |March | 105,000 | + 10,000 „ | +114·28 | |April | 97,500 | + 20,000 „ | +141·18 | |June | 350,000 | + 40,000 „ | +137·14 | |July | 900,000 | + 30,000[38] „ | + 40·00 | |August | 5,500,000 | +1,500,000[38] „ | +327·27 | +----------+------------+------------------+------------+

[38] Nominal.

Forward purchases of francs, after being dearer than spot transactions by 2½ per cent per annum or more from the middle of 1920 to the middle of 1921, were nearly level in price from the middle of 1921 to the middle of 1922, whilst since that time they have been ½ to 2½ per cent per annum cheaper. In the case of lire there have been much wider gaps, forward purchases being frequently 3 per cent or more dearer than spot. In the case of German marks, the forward rate, after ranging round about 5 per cent per annum dearer than spot, has reached, since the autumn of 1922 and the complete collapse of the mark, a figure fantastically cheaper, thus reflecting the sensational rate of interest for short loans current inside Germany.

But in all these cases (except in Germany since the complete collapse of the mark), whether forward exchange is at a discount or at a premium on spot, the expense, if any, of dealing forward has been small in relation to the risks that are avoided.

Nevertheless, in practice merchants do not avail themselves of these facilities to the extent that might have been expected. The nature of forward dealings in exchange is not generally understood. The rates are seldom quoted in the newspapers. There are few financial topics of equal importance which have received so little discussion or publicity. The present situation did not exist before the war (although even at that time forward rates for the dollar were regularly quoted), and did not begin until after the “unpegging” of the leading exchanges in 1919, so that the business world has only begun to adapt itself. Moreover, for the ordinary man, dealing in forward exchange has, it seems, a smack of speculation about it. Unlike Manchester cotton spinners, who have learnt by long experience that it is not the hedging of open cotton commitments on the Liverpool futures market, but the failure to do so, which is speculative, merchants, who buy or sell goods of which the price is expressed in a foreign currency, do not yet regard it as part of the normal routine of prudent business to hedge these indirect exchange commitments by a transaction in forward exchange.

It is important, on the other hand, not to exaggerate the extent to which, at the present time, merchants can by this means protect themselves from risk. In the first place, for reasons, some of which will be considered below, it is only in certain of the leading exchanges that these transactions can be carried out at a reasonable charge. It is not clear that even the banks themselves have yet learnt to look on the provision for their clients of such facilities at fair and reasonable rates as one of the most useful services they can offer. They have been too much influenced, perhaps, by the fear that these facilities might tend at the same time to increase speculation.

But there is a further qualification, not to be overlooked, to the value of forward transactions as a protection against risk. The price of a particular commodity, in terms of a particular currency, does not exactly respond to changes in the value of that currency on the exchange markets of the world, with the result that a movement in a country’s exchanges may, in the case of a commodity of which that country is a large seller or a large purchaser, change the commodity’s world-value expressed in terms of gold. In that case a merchant, even though he is hedged in respect of the exchange itself, may lose, in respect of his unsold trading stocks, through a movement in the world-value of the commodity he is dealing in, directly occasioned by the exchange fluctuation.

* * * * *

If we turn to the theoretical analysis of the forward market, what is it that determines the amount and the sign (whether plus or minus) of the divergence between the spot and forward rates as recorded above?

If dollars one month forward are quoted cheaper than spot dollars to a London buyer in terms of sterling, this indicates a preference by the market, on balance, in favour of holding funds in New York during the month in question rather than in London,--a preference the degree of which is measured by the discount on forward dollars. For if spot dollars are worth $4.40 to the pound and dollars one month forward $4.40½ to the pound, then the owner of $4.40 can, by selling the dollars spot and buying them back one month forward, find himself at the end of the month with $4.40½, merely by being during the month the owner of £1 in London instead of $4.40 in New York. That he should require and can obtain half a cent, which, earned in one month, is equal to about 1½ per cent per annum, to induce him to do the transaction, shows, and is, under conditions of competition, a measure of, the market’s preference for holding funds during the month in question in New York rather than in London.

Conversely, if francs, lire, and marks one month forward are quoted dearer than the spot rates to a London buyer, this indicates a preference for holding funds in London rather than in Paris, Rome, or Berlin.

The difference between the spot and forward rates is, therefore, precisely and exactly the measure of the preference of the money and exchange market for holding funds in one international centre rather than in another, _the exchange risk apart_, that is to say under conditions in which the exchange risk is covered. What is it that determines these preferences?

1. The most fundamental cause is to be found in the interest rates obtainable on “short” money--that is to say, on money lent or deposited for short periods of time in the money markets of the two centres under comparison. If by lending dollars in New York for one month the lender could earn interest at the rate of 5½ per cent per annum, whereas by lending sterling in London for one month he could only earn interest at the rate of 4 per cent, then the preference observed above for holding funds in New York rather than in London is wholly explained. That is to say, forward quotations for the purchase of the currency of the dearer money market tend to be cheaper than spot quotations by a percentage per month equal to the excess of the interest which can be earned in a month in the dearer market over what can be earned in the cheaper. It must be noticed that the governing factor is the rate of interest obtainable for short periods, so that a country where, owing to the absence or ill-development of an organised money market, it is difficult to lend money satisfactorily at call or for very short periods, may, for the purposes of this calculation, reckon as a low interest-earning market, even though the prevailing rate of interest for longer periods is not low at all. This consideration generally tends to make London and New York more attractive markets for short money than any Continental centres.

The effect of the cheap money rates ruling in London from the middle of 1922 to the middle of 1923 in diminishing the attractiveness of London as a depository of funds is strikingly shown, in the above tables, by the cheapening of the forward quotations of foreign currencies relatively to the spot quotations. In the case of the dollar the forward quotation had risen by the beginning of 1923 to a rate 3 per cent per annum above the spot quotation (_i.e._ forward dollars were 3 per cent per annum _cheaper_ than spot dollars in terms of sterling), which meant (subject to modification by the other influences to be mentioned below) that the effective rate for short loans approached 3 per cent higher in New York than in London.

In the case of francs forward quotations which had been below spot, so long as money was dear in London, rose above the spot quotations, thus indicating that the relative dearness of money in London as compared with Paris had disappeared; whilst in the case of lire forward quotations, although still below spot quotations, rose, under the same influence, nearer to the spot level. Nevertheless, in the case of both these currencies, a preponderance of bearish anticipations about their future prospects probably also played a part, for the reasons given in detail below, in producing the observed result.

The most interesting figures, however, are those relating to marks, which illustrate vividly what I have mentioned on page 23 above concerning the enormous money rates of interest current in Germany subsequent to the collapse of October 1922, as a result of the effort of the real rate of interest to remain positive in face of a general anticipation of a catastrophic collapse of the monetary unit. It will be noticed that the effective short money rate of interest in terms of marks ranged from 50 per cent per annum upwards, until finally the quotations were merely nominal.

2. If questions of credit did not enter in, the factor of the rate of interest on short loans would be the dominating one. Indeed, as between London and New York, it probably is so under existing conditions. Between London and Paris it is still important. But elsewhere the various uncertainties of financial and political risk, which the war has left behind, introduce a further element which sometimes quite transcends the factor of relative interest. The possibility of financial trouble or political disturbance, and the quite appreciable probability of a moratorium in the event of any difficulties arising, or of the sudden introduction of exchange regulations which would interfere with the movement of balances out of the country, and even sometimes the contingency of a drastic demonetisation,--all these factors deter bankers, even when the exchange risk proper is eliminated, from maintaining large floating balances at certain foreign centres. Such risks prevent the business from being based, as it should be, on a mathematical calculation of interest rates; they obliterate by their possible magnitude the small “turns” which can be earned out of differences between interest rates plus a normal banker’s commission; and, being incalculable, they may even deter conservative bankers from doing the business on a substantial scale at any reasonable rate at all. In the case of Roumania or Poland, for example, this factor is, at times, the dominating one.

3. There is a third factor of some significance. We have assumed so far that the forward rate is fixed at such a level that the dealer or banker can cover himself by a simultaneous spot transaction and be left with a reasonable profit for his trouble. But it is not necessary to cover every forward transaction by a corresponding spot transaction; it may be possible to “marry” a forward sale with a forward purchase of the same currency. For example, whilst some of the market’s clients may wish to sell forward dollars, other clients will wish to buy forward dollars. In this case the market can set off these, one against the other, in its books, and there will be no need of any movement of cash funds in either direction. The third factor depends, therefore, on whether it is the sellers or the buyers of forward dollars who predominate. To fix our minds, let us suppose that money-market conditions exist in which a sale of forward dollars against the purchase of spot dollars, at a discount of 1½ per cent per annum for the former, yields neither profit nor loss. Now if in these conditions the purchasers of forward dollars, other than arbitragers, exceed sellers of forward dollars, then this excess of demand for forward dollars can be met by arbitragers, who have cash resources in London, at a discount which falls short of 1½ per cent per annum by such amount (say ½ per cent) as will yield the arbitragers sufficient profit for their trouble. If, however, sellers of forward dollars exceed the purchasers, then a sufficient discount has to be accepted by the former to induce arbitrage the other way round--that is to say, by arbitragers who have cash resources in New York--namely, a discount which exceeds 1½ per cent per annum by, say, ½ per cent. Thus the discount on forward dollars will fluctuate between 1 and 2 per cent per annum according as buyers or sellers predominate.

4. Lastly, we have to provide for the case, quite frequent in practice, where our assumption of a large and free market breaks down. A business in forward exchange can only be transacted by banks or similar institutions. If the bulk of the business in a particular exchange is in a few hands, or if there is a tacit agreement between the principal institutions concerned to maintain differences which will allow more than a competitive profit, then the surcharge representing the profit of a bank for arbitraging between spot and forward transactions may much exceed the moderate figure indicated above. The quotations of the rates charged in Milan for forward dealings in lire, when compared with the rates current in London at the same date, indicate that a bank which is free to operate in both markets can frequently make an abnormal profit.

But there is a further contingency of considerable importance which occurs when speculation is exceptionally active and is all one way. It must be remembered that the floating capital normally available, and ready to move from centre to centre for the purpose of taking advantage of moderate arbitrage profits between spot and forward exchange, is by no means unlimited in amount, and is not always adequate to the market’s requirements. When, for example, the market is feeling unusually bullish of the European exchanges as against sterling, or of sterling as against dollars, the pressure to sell forward sterling or dollars, as the case may be, may drive the forward price of these currencies to a discount on their spot price which represents an altogether abnormal profit to any one who is in a position to buy these currencies forward and sell them spot. This abnormal discount can only disappear when the high profit of arbitrage between spot and forward has drawn fresh capital into the arbitrage business. So few persons understand even the elements of the theory of the forward exchanges that there was an occasion in 1920, even between London and New York, when a seller of spot dollars could earn at the rate of 6 per cent per annum above the London rate for short money by converting his dollars into sterling and providing at the same time by a forward sale of the sterling for reconversion into dollars in a month’s time; whilst, according to figures supplied me, it was possible, at the end of February 1921, by selling spot sterling in Milan and buying it back a month forward, to earn at the rate of more than 25 per cent per annum over and above any interest obtainable on a month’s deposit of cash lire in Milan.

It is interesting to notice that when the differences between forward and spot rates have become temporarily abnormal, thus indicating an exceptional pressure of speculative activity, the speculators have often turned out to be right. For example, the abnormal discount on forward dollars, which persisted more or less from November 1920 to February 1921, thus indicating that the market was a bull of sterling, coincided with the sensational rise of sterling from 3.45 to 3.90. This discount was at its maximum when sterling touched its lowest point and at its minimum (in the middle of May 1921) when sterling reached its highest point on that swing, which showed a remarkably accurate anticipation of events by the balance of professional opinion. The comparatively high discount on forward dollars current at the end of 1922 may, in the same way, have been partly due to an excess of bull speculation in favour of sterling based on an expectation of its recovery towards par, and not merely to the cheapness of money in London as compared with New York.

The same thing seems to have been true for the franc. In January and February 1921, the abnormal premium on the forward franc indicated that, in the view of the market, the franc had fallen too low, which turned out to be the case. They turned round at the precise moment when the franc reached its highest value (end of July 1921), and were right again. During the first five months of 1922, when the franc was almost stable, spot and forward quotations were practically at par with one another, whilst the progressive fall of the franc since June 1922 has been accompanied by a steady and sometimes substantial discount on forward francs; indicating, on this test, that the professional market was bearish of francs and therefore right once more. The lira tells somewhat the same tale. Thus, whilst the reader can see for himself by a study of the tables that no precise generalisation would be accurate, nevertheless the market has been broadly right when it has taken a very decided view, as measured by forward rates.

This result may seem surprising in view of the huge amounts which exchange speculators in European currencies, more particularly on the bull side, are reputed to have lost. But the mass of amateur speculators throughout the world operate by cash purchases of the currency of which they are bulls, forward transactions being neither known nor available to them. Such speculation may afford temporary support to the spot exchange, but it has no influence on the difference between spot and forward, the subject now under discussion. The above conclusion is limited to the fact that when the type of professional speculation which makes use of the forward market is exceptionally active and united in its opinion, it has proved roughly correct, and has, therefore, been a useful factor in moderating the extreme fluctuations which would have occurred otherwise.

* * * * *

Out of the various practical conclusions which might be drawn from this discussion and the figures which accompany it, I will pick out three.

1. Those exchanges in which the fluctuations are wildest and the merchant is most in need of facilities for hedging his risk are precisely those in which facilities for forward dealing at moderate rates are least developed. But this is to be explained, not necessarily by the instability of the exchange in itself, but by certain accompanying circumstances, such as distrust of the country’s internal arrangements or its banking credit, a fear of the sudden imposition of exchange regulations or of a moratorium, and the other analogous influences mentioned above (pp. 126–7). There is no theoretical reason why there should not be an excellent forward market in a highly unstable exchange. In those countries, therefore, where regulation is still premature, it may nevertheless be possible to mitigate the evil consequences of fluctuation by organising facilities for forward dealings.

This is a function which the State banks of such countries could usefully perform. For this they must either themselves command a certain amount of foreign currency or they must provide facilities for accepting short-period deposits in their own currency from foreign bankers, on conditions which inspire these bankers with complete confidence in the freedom and liquidity of such deposits. Various technical devices could be suggested. But the simplest method might be for the State banks themselves to enter the forward market and offer to buy or sell forward exchange at a reasonable discount or premium on the spot quotation. I suggest that they should deal not direct with the public but only with approved banks and financial houses, from whom they should require adequate security; that they should quote every day their rates for buying and selling exchange either one or three months forward; but that such quotation should take the form, not of a price for the exchange itself, but of a percentage difference between spot and forward, and should be a quotation for the double transaction of a spot deal one way and a simultaneous forward deal the other--_e.g._ the Bank of Italy might offer to sell spot sterling and buy forward sterling at a premium of ⅛ per cent per month for the former over the latter, and to buy spot sterling and sell forward sterling at par. For the transaction of such business the State banks would require to command a certain amount of resources abroad, either in cash or in borrowing facilities. But this fund would be a revolving one, automatically replenished at the maturity of the forward contracts, so that it need not be on anything like the scale necessary for a fund for the purpose of supporting the exchange. Nor is it a business which involves any more risk than is inherent in all banking business as such; from exchange risk proper is free.

With free forward markets thus established no merchant need run an exchange risk unless he wishes to, and business might find a stable foothold even in a fluctuating world. A recommendation in favour of action along these lines was included amongst the Financial Resolutions of the Genoa Conference of 1922.

I shall develop below (Chap. V.) a proposal that the Bank of England should strengthen its control by fixing spot and forward prices for gold every Thursday just as it now fixes its discount rate. But other Central Banks also would increase their control over fluctuations in exchange if they were to adopt the above plan of quoting rates for forward exchange in terms of spot exchange. By varying these rates they would be able, in effect, to vary the interest offered for _foreign_ balances, as a policy distinct from whatever might be their bank-rate policy for the purpose of governing the interest obtainable on _home_ balances.

2. It is not unusual at present for banks to endeavour to distinguish between speculative dealings in forward exchange and dealings which are intended to hedge a commercial transaction, with a view to discouraging the former; whilst official exchange regulations in many countries have been aimed at such discrimination. I think that this is a mistake. Banks should take stringent precautions to make sure that their clients are in a position to meet any losses which may accrue without serious embarrassment. But, having fully assured themselves on this point, it is not useful that they should inquire further--for the following reasons.

In the first place, it is almost impossible to prevent the evasion of such regulations; whilst, if the business is driven to methods of evasion, it tends to be pressed underground, to yield excessive profits to middlemen, and to fall into undesirable hands.

But, what is more important and is less appreciated, the speculator with resources can provide a useful, indeed almost an essential, service. Since the volume of actual trade is spread unevenly through the year, the seasonal fluctuation, as explained above, is bound to occur with undue force unless some financial, non-commercial factor steps in to balance matters. A free forward market, from which speculative transactions are not excluded, will give by far the best facilities for the trader, who does not wish to speculate, to avoid doing so. The same sort of advantages will be secured for merchants generally as are afforded, for example, to the cotton trade by the dealings in “futures” in the New York and Liverpool markets. Where risk is unavoidably present, it is much better that it should be carried by those who are qualified or are desirous to bear it, than by traders, who have neither the qualification nor the desire to do so, and whose minds it distracts from their own business. The wide fluctuations in the leading exchanges over the past three years, as distinct from their persisting depreciation, have been due, not to the presence of speculation, but to the absence of a sufficient volume of it relatively to the volume of trade.