Studies in the Theory of International Trade

Chapter III

THE BULLIONIST CONTROVERSIES:

I. THE INFLATION PHASE

I. THE PARTICIPANTS IN THE CONTROVERSY

Of all the older controversies in the field of international trade theory, the inflation phase of the bullionist
*1 controversy has probably
been most fully and competently canvassed by modern writers.
*2 But there is still room for a resurvey of the controversy.

The contemporary literature of the bullionist controversy is of great importance for the history of the theory of international trade in its monetary aspects. The germs at least of most of the current monetary theories are to be found in it. It embodies the first detailed analysis of the relationships between currency phenomena and international balances, exchange rates, and price levels, under both metallic and inconvertible paper currencies. Foreign exchange theory is carried substantially forward, and the theory of the mechanism of adjustment of international balances is advanced substantially beyond the stage at which it was left by Hume. There are also discussions of a truly pioneer character of the functions of a central bank in a complex credit economy with respect to the maintenance of international monetary equilibrium and of internal business stability.

The contemporary participants in the controversy arrayed themselves fairly sharply in two opposing groups: the “bullionists” or “anti-Restrictionists” on the one hand, who criticized the course of monetary events, and the “anti-bullionists” on the other hand, who defended the government and the Bank of England against the attacks of the bullionists. But as will be seen, there were important divergences of opinion within each group. The essential doctrines of the bullionists were expressed by a small group of writers, of whom Boyd,
*3 King,
*4 Thornton,
*5 Wheatley,
*6and Horner,
*7 were most important, during the first period, 1801 to 1803, of marked premium on bullion and fall in the exchanges. Similar phenomena, even more marked in degree, in connection with the Bank of Ireland gave rise to a parliamentary inquiry
*8 and to the bullionist publications of John Leslie Foster,
*9 Henry Parnell,
*10 and Lord Lauderdale.
*11 The reappearance from 1809 on of a high premium on gold and a substantial fall in the exchanges gave rise to a flood of tracts and pamphlets, of which the most important on the bullionist side, in addition to the Report of the Bullion Committee of 1810, were the contributions of Ricardo, his first appearance in print as an economist,
*12 T. R. Malthus,
*13 Robert Mushet,
*14 and William Huskisson.
*15

The most effective statements of the anti-bullionist position were in speeches in Parliament by Nicholas Vansittart
*16 and George Rose,
*17 and in tracts by Henry Boase,
*18 Bosanquet,
*19 Coutts Trotter,
*20 and J. C. Herries.
*21

Ricardo made but few additions to the analysis of his predecessors,
*22 and, as will be shown later, on some important points he committed errors from which some of the earlier supporters of the bullionist position had been free. But the comprehensiveness and the force and skill of his exposition and the assurance and rigor of his reasoning made him at once the leading expositor of the bullionist position. It was largely through Ricardo’s writings, moreover, that the bullionist doctrines exercised their influence on the subsequent century of monetary controversy. Special attention is given, therefore, to Ricardo’s position in the following account of the bullionist controversy.

II. THE FACTUAL BACKGROUND

From the outbreak of the war with France in 1793, the Bank of England had been under a strain mainly because of the great demands for advances made upon it by the government, which it had resisted, but unsuccessfully. Early in 1797, a general panic, induced apparently by rumors of a French landing on English soil, and accentuated by failures and suspensions on the part of the country banks, led to a general clamor for gold. On February 25, 1797, there were only £1,272,000 of specie and bullion in the Bank, as compared to ordinary reserves of £5,000,000, or over. On February 26, 1797, the government, at the request of the Bank, issued an Order in Council prohibiting specie redemption of its notes by the Bank. By an Act of May 3, 1797, the restriction of cash payments was validated and continued in effect,
subject to minor qualifications, until June 24, 1797, and by a succession of later acts the suspension of specie payments was enforced until after the end of the war. With the factors responsible for the suspension of specie payments in 1797, we need not here concern ourselves.
*23 The suspension of specie payments was quickly followed by an inward flow of bullion, recovery of the Bank from its strained condition, and general restoration of confidence, and it was not until toward the end of 1799 that the exchange on Hamburg fell substantially below the pre-Restriction par and a premium was quoted on bullion over paper. From 1804 to 1808 the exchanges were again at or near parity, and paper was at no or a small discount in relation to bullion. But from 1809 to the end of the war there again prevailed low sterling exchanges and substantial premiums of bullion over paper.
*24

England, prior to the Restriction, although legally on a bimetallic basis, had for some time been in effect on a gold standard basis, since the mint ratio of silver to gold was such as generally to undervalue silver and thus keep it out of circulation. The metallic currency consisted of guinea pieces (= 21 shillings) and multiples and subdivisions thereof, and of silver coins from the crown (= 5 shillings) down. Of the silver coins, only the underweight coins remained in circulation. Except for coins surviving from ancient issues, the sovereign (= 20 shillings) was only a money of account. English coin could not legally be melted down unless underweight, and was not legally exportable, and gold bullion was exportable only subject to oath that it had not been obtained by melting down English coin. The metallic currency was supplemented by Bank of England notes in denominations of £5 or over, redeemable in specie upon demand, and by country bank notes, also in denominations of £5 or over, payable upon demand in specie or in Bank of England notes. London bankers had in 1793 voluntarily ceased to issue their own notes. Outside of the London area the Bank of England notes circulated freely only in Lancashire, where the local banks did not issue notes but where bills of exchange of small denominations were extensively employed as a medium of exchange. Bank deposits subject to check were also in existence, and constituted a part of what would
today be regarded as the circulating medium, although this was not yet widely recognized. Checks payable to order had only recently come into common use even in London and only for large payments. The private or non-governmental deposits at the Bank of England were small in amount throughout the Restriction period, and for the years after 1806, for which alone their precise amounts are known, they reached a yearly average of £2,000,000 in only one year.
*25 In the provinces also deposits seem to have been relatively unimportant, and to have been drawn upon mainly for cash, but the available evidence on this point is conflicting.
*26

III. PREMIUM ON BULLION AS EVIDENCE OF EXCESS ISSUE: THE BULLIONIST POSITION

The central issue of the controversy was made to turn on the question of whether the paper pound was depreciated, the bullionists insisting that it was depreciated, and most—though not all—of the anti-bullionists denying this. The answer to such a question obviously depends on how “depreciation” is defined, and the controversy suffered from a constant tendency to degenerate into merely terminological issues. As one bullionist writer caustically remarked: “Whether reduction of prices [of paper in gold] be depreciation or not, or equivalent to it, is a verbal question very fit to be argued in ‘Change Alley.'”
*27 But always present,
even when not clearly brought into the foreground of the discussion, were genuine and important issues of fact and policy.

For the bullionists the paper currency was depreciated if issued to excess, and many of the anti-bullionists also accepted this quantitative criterion of depreciation, or at least did not explicitly reject it. Defining depreciated currency as a currency issued to excess might seem merely to substitute one term of doubtful meaning for another. But the question, What is the proper amount of currency a country should have? Is an important one. To this question, as Hollander points out,
*28 Adam Smith had given no answer beyond saying vaguely that it was determined by “effectual demand,”
*29 and the participants in the bullionist controversy were the first seriously to tackle it. The bullionists argued, or more often simply asserted, that a circulation exceeding in amount what, under otherwise like conditions, could have been maintained under a metallic standard, was in excess. There was little express objection to this criterion of a properly-regulated currency during the inflation phase, and serious discussion of its adequacy came only with the deflation phase of the controversy.

During the inflation phase the main issue in controversy was as to the proper method of determining the existence of excess of issue. The chief test of excess issue used by the bullionists was the existence of a premium on bullion over paper currency, although since they held that the level of prices was determined by the amount of currency and that the amount of premium of bullion over paper and the amount of discount of sterling exchange from the metallic parities were closely related, they also held that a relative rise of prices in England as compared to abroad and a fall in the sterling exchanges below parity were evidence of depreciation. The bullionist position was well expressed by Boyd: “The premium on bullion, the low rate of exchange, and the high prices of commodities in general, [are]…symptoms and effects of the superabundance of paper.”
*30 Their conclusions rested on the following reasoning: the rate of exchange between two currencies depended solely or mainly on their relative purchasing power over identical transportable commodities in the two countries; on quantity theory of money grounds, prices in the two countries depended on the quantities of money circulating therein; the price of bullion in paper currency was governed by the exchange rates with metallic standard currencies; therefore, if the exchanges were below metallic parity, and if there was a premium on bullion over paper, this was evidence that prices were higher in England, and the quantity of currency in circulation greater, than would have been possible under the metallic standard prevailing prior to suspension of convertibility.
*31

While Wheatley and Ricardo held that the relative rise of the prices of particular commodities in England, as compared to the prices of the same commodities in foreign countries having metallic standard currencies, would be proportional to the degree of excess of the English currency, they did not suggest that the existence of excess issue could in practice be tested by such price comparisons.
*32 The notion of an index number was still in its infancy. Evelyn had published his crude index number of English prices for the preceding two centuries in 1798, and Wheatley had commented on it in laudatory terms.
*33 But no current index number yet existed for England, and there was but little information as to the prices prevailing in other countries. To Ricardo, moreover, it seemed an absurd notion that the trend of prices in
general, or of the general purchasing power of money, could be measured. Since prices fluctuated even under a metallic standard, he conceded that their fluctuations under an inconvertible currency could not be attributed solely to changes in the degree of excess of the currency. The only test from English prices alone of the existence of depreciation which he could consistently have accepted, therefore, would have been a comparison of the prices prevailing under inconvertibility with the prices which would have prevailed under convertibility, other conditions remaining the same, and in his treatment of arguments from price data Ricardo always adhered to this position.
*34 But Ricardo held that since the premium on bullion measured the degree of excess of the currency,
*35 it measured also the degree in which prices at anytime, say 1810, during the suspension of cash payments were higher, not than they had been in 1797, but than they would have been in 1810 if the currency were in 1810 at the amount which could then have been maintained in circulation under a metallic standard. Ricardo, however, put much stress on the question of the extent of the depreciation, as providing an answer to the question of how great a reduction in the currency would be needed to end the depreciation.

IV. QUALIFICATIONS CONCEDED BY THE BULLIONISTS

The bullionists were prepared to make several qualifications to this reasoning and therefore to concede that the existence of a premium on bullion over paper, or of a discount of sterling exchange from metallic parity, was not an absolute proof of excess issue, and was strong presumptive evidence of excess issue only
if it was substantial and prevailed for a considerable period of time.

Secondly, when the bullionists used the exchange rates as an alternative or supplementary test of the existence of depreciation, they conceded that since even under a metallic standard the exchanges could fall below the mint parity to the limit of the cost of shipping bullion, a fall in the exchanges which did not go beyond this limit was not proof that there was excess of currency under inconvertibility.
*36 Moreover, whereas England had been on a gold standard basis prior to the Restriction, Hamburg, Amsterdam, and Paris, the most important quotation points for the exchanges during the Restriction, were on a silver standard basis. Ricardo and other bullionists pointed out that since the relative values of gold and silver were not constant, the exchange parties between gold and silver currencies also were not constant, and that in computing the deviation of, say, the London-Hamburg exchange from parity it was necessary to make allowance for any alteration in the relative market values of the two metals. But the general trend of the price of silver as compared to gold was downward during the later stage of the controversy, and
Ricardo pointed out that comparisons of the trend of the Hamburg exchange during the Restriction period which used the prevailing rate before the Restriction as the base therefore underestimated the extent of the real fall in the exchange value of English paper currency in terms of gold bullion abroad.
*37

Thirdly, even before 1797, English gold coin, or bullion derived therefrom, was not legally exportable, and at a time when the exchanges were against England exportable bullion would command a premium over its mint price in coin or in paper. Ricardo and other bullionists insisted, however, that the prohibition of export of English coin or bullion could not be successfully enforced and that a small premium would in practice suffice to compensate for the risks involved in melting and false swearing, or in smuggling English bullion out of the country. They conceded that a premium on gold not exceeding this risk-premium was not necessarily indicative of excess. Bullion also could command a premium over coin and paper even under convertibility if the coinage was generally underweight as compared to its nominal standard, and for this also the bullionists were willing to make allowances. But the gold coinage was in good condition in 1797, and only a minor allowance was called for on this account.
*38 The bullionists tended to agree that 5 per cent premium on gold was ample allowance for both these factors, and it seems that in the years prior to 1797 the premium on gold at no time exceeded this except in isolated and special transactions.
*39

Fourthly, the bullionists recognized that the substitution in England of paper for gold and the export of the displaced gold would tend to result in a rise of prices in other countries in terms of gold, and that England could share in this rise of prices, and could therefore circulate a greater quantity of currency than before, other things remaining the same, without suffering a
premium on gold or fall in the exchanges.
*40 They did not attach any importance to this factor, however, presumably on the ground that any such release of gold would be negligible in comparison to the world supply.
*41 Since to the extent that this consideration had weight it would tend to make the bullionists’ tests of excess as they defined it too generous rather than too exacting, the anti-bullionists also made no use of it, although it became an important element in the controversy of the deflation period.

V. POSSIBLE OBJECTIONS TO THE BULLIONIST POSITION

Similarly, a currency might depreciate because of an increase in its velocity of circulation, its amount meanwhile remaining constant or even falling. This was generally recognized at the time, but it was tacitly assumed, then and later, not that velocity remained constant—for it was known that it was subject to variation with the state of business confidence, with improvements in the means of communication, and with the development of clearinghouse
and other arrangements for “economizing currency,”
*42 but that velocity would not be altered merely by the suspension of convertibility. If changes in velocity due to changing degrees of confidence in the future of the currency be disregarded, this assumption could not be expected to be a source of serious error. Under convertibility the actually circulating medium, if deposits and bills of exchange be disregarded, was partly coin, partly paper; under inconvertibility it was wholly paper. It is conceivable that individuals would tend to hold smaller cash balances in proportion to the volume of their transactions if the currency was paper than if it was coin. Holding of paper involved risk of loss through fire, or through failure of the issues. Paper money could be shipped from one point to another more promptly, more safely, and if in small quantities more economically, than could specie, for paper money could be sent by post, whereas specie remittances required private couriers, who had to be convoyed because of the danger of robbery on the highways. This would tend to lead to the holding on the average of larger cash balances relative to volume of transactions if the currency were specie than if it were paper.
*43 But it seems doubtful that this could have been an important factor.

On both
a priori and empirical grounds, however, velocity should be expected to rise as the volume of means of payments and the price level was rising, and thus measurement of the percentage of excess of currency from the percentage of discount of paper in terms of gold would tend to exaggerate the degree of excess during rising prices and to underestimate it during falling prices.

A more serious qualification to the validity of the bullionist position lies in the fact that under inconvertibility speculative anticipations of depreciation or appreciation of the currency would affect the willingness of individuals to hold the currency and would thus influence its velocity of circulation and its value in
relation to gold, to foreign currencies, and to commodities, independently of the effects of variations in its quantity. In modern times, as we now know only too well, such speculative factors can dominate for an appreciable length of time the metallic or exchange value of an inconvertible paper currency. There is every reason to believe that such speculative factors were also operative in some degree during the period of the bullionist controversy.

Silberling and Angell misread into the bullionist writings in general the positive charge that the depreciation of the paper pound in relation to bullion was in part at least a “qualitative” depreciation, and they find something absurd in such a charge. Silberling claims to find in Ricardo’s writings the doctrine, which he clearly regards as a strange one, that the “fall” in paper money was due to “a mere inherent debasement in quality” of the paper currency rather than to its issue to excess. He concedes that “debasement” could readily be translated into “excess,” if by excess is meant the amount exceeding the quantity at which the price of gold in paper would be at its mint par.
*44 But Ricardo
repeatedly and uniformly insisted that he meant just this by excess.

Angell follows Silberling in finding among the bullionists adherence to the notion of a
qualitative depreciation of the currency, and in treating it as an absurd notion, but his interpretation of the bullionist position in this connection is different from Silberling’s. Angell claims that Boyd, Ricardo, and other bullionists held that an excess of currency led first to “a positive degradation of the standard” and that this degradation in turn led to a rise in prices, “the degradation thus being a distinct and ‘intermediate’ step between the increase in currency and the rise in prices.”
*45 Angell gives no specific references to Ricardo, but he refers to the following passage in Boyd:

He would say, that not only the currency of the country had been changed from a
certain to an
uncertain standard, but that the quantity of it, in all probability, had been greatly augmented by the issuing of paper, without the obligation of paying it on demand, and that thus the prices of all objects of exchangeable value necessarily feel the influence of a
positive degradation of the standard, and of a
probable augmentation of the quantity of money in the country, any one of them amply sufficient to discount for a considerable rise, but both united, adequate to still greater effects than any that had already been produced.
*46

Boyd here clearly assigns to “degradation” a distinct influence on prices over and above that resulting from any increase in the quantity of the currency. But there is no trace here of the time-sequence imputed to him by Angell. The context shows that the word “positive” which qualifies “degradation” is to be understood to mean “certain,” as contrasted to the “probable” increase of the amount of the currency. At the time Boyd wrote no report had been made as to the issues of the Bank since the Restriction, and increase in such issue could be only a matter of inference from circumstantial evidence. The question remains, what did Boyd mean by “degradation”? No light is afforded by the context, but a reasonable explanation which makes his position intelligible
is made possible by reference to a doctrine of other contemporary writers. Henry Thornton in 1797 had argued that the quantity of notes which it was proper at any time to issue depended much “on the state of the public mind, that is, on the disposition of persons to detain them.” Thus an impairment of the general credit “while Bank notes sustain their credit” would make possible, and desirable, an increase of the issue of notes without any impairment of their value.
*47 In 1802 he repeated this argument and supported it by reference to the effect of confidence in the paper money on the velocity of its circulation and on the size of the cash balances generally held by individuals.
*48 He pointed out, moreover, that while paper was falling in value, foreigners generally would expect “that the paper, which is falling in value, will, in better times, only cease to fall, or, if it rises, will experience only an immaterial rise, and this expectation serves of course to accelerate its fall.”
*49 Thus the suspension of cash payments could conceivably result in a premium of bullion over paper even if no increase in the issue of paper had occurred. But Thornton denied that the loss of confidence in the English currency which could bring this about had occurred.

Lord King and George Woods expressed similar views:

But when the obligation to pay in coin ceases, the currency no longer retains this determinate value, but is in danger of being depreciated from two different causes; viz., by want of confidence on the part of the public, and an undue increase of the quantity of notes…. Though the persons who have the regulation of a currency not payable on demand should confine their issues within the most just and reasonable limits; yet if their credit or solvency is doubted, it is impossible that their notes can circulate at the full nominal value.
*50

Whether the depreciation of bank notes be owing to excess of issue or to the ticklish foundation upon which their present validity is built, the ever-varying standard of public opinion, the fact itself… [i.e., of depreciation of paper in terms of bullion] is undeniable…. If it be alleged that the issues of the Bank, compared
with the wants of the public, are not greater now than formerly, I answer, that this reasoning may imply a decreased confidence in the Bank of England, but that it does not throw the smallest light upon the question of depreciation.
*51

Ricardo likewise disclaimed any belief that in 1810 lack of confidence in the paper pound was a factor in its depreciation: “I am not aware of any causes but excess, or a want of confidence in the issues of the paper (which I am sure does not now exist), which could produce such effects as we have for a considerable time witnessed.”
*52

The bullionists on this point were in error. Their error, however, lay not, as Silberling and Angell claim, in attributing some of the depreciation of the paper pound to loss of confidence in it, but in their refusal to do so, although this refusal may have been due to prudential considerations. For as Horner and Ricardo later acknowledged,
*53 some of the sharp fluctuations in the premium on gold could not be adequately explained as due to corresponding fluctuations in the quantity of paper money, and could be adequately explained only with reference to changes in anticipations as to the future of the paper pound, resulting in changes in willingness of Englishmen to hold cash balances in paper and of foreigners to hold securities payable in sterling.

The bullionist position is open to one further correction, but one of probably minor practical importance. Under a metallic standard, if due to foreign remittances or abnormally heavy grain imports there occurs a temporary rise in the relative demand for
foreign bills, an export of specie will tend to occur, which will operate both to lower the amount of the domestic circulation and directly to increase the supply of foreign bills by the amount for which the exported specie itself can be exchanged. Under an inconvertible currency which has been on a depreciated basis for some time, so that all the bullion has already either been exported or passed into more or less permanent hoards, there will be no specie export to constitute a direct equilibrating element in the international balance of indebtedness. With the same volume of foreign remittances to be made, a greater contraction of the currency, therefore, will be necessary under inconvertibility than under a metallic standard if the exchanges are to be kept from falling by more than the cost of shipping gold, and conversely, a fall of the exchanges by more than the cost of shipping gold will not be absolute proof that the currency has been contracted in less degree than would have been necessary if the standard were metallic.

VI. THE ANTI-BULLIONIST POSITION

The anti-bullionists often attempted to show from statistics as to Bank of England note issues either that the issues had not increased or that there was no relation in time or degree between the fluctuations in issue and the fluctuations in the premium on bullion or the exchanges. But Ricardo was able to show that even if the data were as alleged—as they often were not—they did not refute his argument. He was claiming not that the currency had been increased during the Restriction, but that it existed in an
amount greater than could have been maintained at that time, other things remaining the same, if convertibility had been maintained. Whether the amount of actual issue in say 1810 was greater or less than in 1797 was beside the point if it was greater than could have been maintained under convertibility in 1810:

I do most unequivocally admit, that whilst the high price of bullion and the low exchanges continue,…it would to me be no proof of our currency not being depreciated if there were only 5 millions of bank notes in circulation [as compared to about 10 millions in 1797 and 23 millions in 1810]. When we speak, therefore, of an excess of bank notes, we mean that portion of the amount of the issues of the Bank, which can now circulate, but could not, if the currency were of its bullion value.
*54

Some of the anti-bullionists contended that to prove depreciation it was necessary to prove that gold
coin commanded a premium over paper, since bullion was only a commodity and its price therefore of no special significance.
*55 Since it was unlawful to melt or export English coin, and since persons buying such coin at a premium would come under suspicion of intent to violate the law, it is not surprising that there were no open dealings in gold coin at a premium over paper.
*56 What happened was that the full-weight coin quietly but rapidly passed out of circulation and was either exported on government account or went into hoards or into the melting pot for industrial use or for illegal export abroad. As Ricardo pointed out, if the law against melting and export had been repealed, gold coin and gold bullion would have commanded the same premium over paper money;
*57 on the other hand, if the law against melting and export could have been fully enforced, exportable bullion would have commanded the same premium over coin and paper money.
*58

VII. THE BALANCE OF PAYMENTS ARGUMENT

The anti-bullionists, however, had a more serious objection to raise against the acceptance of the premium on bullion over paper as a proof of excess currency. It was agreed on both sides that under an inconvertible as under a convertible standard the price of bullion was governed by the foreign exchanges. Both sides were also agreed that under convertibility the exchanges could not ordinarily fall below the gold export point, since below that point, representing by its distance from mint parity of the two currencies the cost of transmitting bullion, it would be more profitable to ship bullion than to buy foreign bills. The anti-bullionists argued, however, that under inconvertibility this limit to the fall in the exchanges did not exist; that the exchanges and the premium on bullion would be governed solely by the state of the balance of international payments;
*59 and that in a period when heavy military remittances and extraordinary importations of grain because of deficient English harvests had to be made, there was no definable limit beyond which the exchanges could not fall or the premium on bullion rise without demonstrating that the currency was in excess.

In their treatment of this crucial issue, the bullionists were divided into two groups, offering different answers. One of these groups consisted of only two men, Wheatley and Ricardo.
*60 To
the argument that foreign remittances would under inconvertibility operate to depress the exchanges, Wheatley and Ricardo both replied that foreign remittances would have no effect on the exchanges whether under convertibility or inconvertibility; in both cases, they maintained, the demands of England and the rest of the world for each other’s products would so adjust themselves automatically to the remittances that they would be transferred in goods without changes either in relative prices or in exchange rates. If under inconvertibility, therefore, there appeared a depreciation of sterling exchange, this was evidence of excess issue of currency. Ricardo later made some minor concessions to his critics,
*61 but Wheatley adhered rigidly to this doctrine to the end.
*62 The other bullionists took an intermediate position. They conceded that foreign remittances would affect the exchange rates, and conceded also, though without adequately explaining why, that while such remittances were under way a premium on bullion and exchanges below parity were not proof that the amount of currency in circulation was in excess of what could be maintained under a metallic standard currency. They confined themselves to the argument that a continued and substantial premium of gold over paper, and fall of the exchanges below parity, established strong presumptions that the currency was in excess of what could be maintained under a metallic standard. The Bullion Report, for instance, cited the persistence of a high premium on gold and low foreign exchanges “for a considerable time” as the evidence pointing to the existence of excess currency.
*63

It is arguable that the above account exaggerates the difference between Ricardo and the other bullionists, although the publications of the latter and Ricardo’s correspondence show that they were conscious of their existence and were unable to reach a mutually satisfactory reconciliation. Ricardo could very rarely interest himself in the immediate and transitory phases of an economic process sufficiently to trace it in detail through its successive stages, and he frequently confined his analysis to the end results, either passing over without mention or even denying the existence
of the intermediate stages. Ricardo, moreover, tended to omit at times explicit mention of qualifications whose validity he was prepared to acknowledge, if he regarded these qualifications as of minor importance or if he had already in some other connection conceded them. The result of these two habits was a rigor and a precision in his formulation which perhaps gave added force to his exposition when he was dealing with the general public, but which enabled more sophisticated critics to expose him to rebuttal often more damaging in appearance than in fact. These characteristics of Ricardo’s methods of thought are now familiar to economists, and Ricardo was to some extent conscious of them himself.
*64 They are well illustrated by the following passages from Ricardo, of which the first appears to involve an absolute denial of the existence of intermediate stages in the process of international adjustment to a currency disturbance, while the second recognizes their existence but reveals that his interest lay wholly in what occurs after they have fully worked themselves out:

To me…it appears perfectly clear, that a reduction of bank notes would lower the price of bullion and improve the exchange, without in the least disturbing the regularity of our present exports and imports…. Our transactions with foreigners would be precisely the same….
*65

I am not disposed to contend that the issues of one day, or of one month, can produce any effect on the foreign exchanges; it may possibly require a period of more permanent duration; an interval is absolutely necessary before such effects would follow. This is never considered by those who oppose the principles of the Committee. They conclude that those principles are defective, because their operation is not immediately perceived.
*66

After a time Ricardo gave way somewhat to the pressure of dissent from his views not only by the anti-bullionists but by the bulk of the bullionists. In response to criticism by Malthus, he conceded that when remittances were under way the currency in the remitting country would be in excess unless it were reduced
in the proportion which the commodity export surplus constituted of the total stock of goods in that country, which still implied that the remittances could be effected under a convertible standard, or without depreciation of the currency under an inconvertible standard, without involving a relative fall in the level of prices in the remitting country.
*67 He later introduced into his exposition qualifying words and phrases of a kind not to be found in his first writings and which brought him closer to the position of the other bullionists.
*68

The bullionists other than Wheatley and Ricardo conceded that extraordinary remittances would affect the exchanges adversely, but insisted, as against either the express denial of or the failure to give consideration to this factor by the anti-bullionists, that the quantity of note issues, through its effect on commodity prices, and thus on the trade balance, was an additional factor determining the exchange value of the English currency, and ordinarily would be the dominant factor.
*69 Perhaps the best brief statement of the moderate bullionist position was the following by Malthus:

The real state of the case seems to be, that though the effects of a redundancy of currency upon the exchange are sure, they are slow, compared with the effects of those mercantile transactions not
connected with the question of currency; and, while the former of these causes is proceeding in its operations with a steady and generally uniform pace, the more rapid movements of the latter are opposing, aggravating or modifying these operations in various ways, and producing all those complex, and seemingly inconsistent, appearances, which are to be found in the computed exchange.
*70

Wheatley and Ricardo, it appears to me, were clearly wrong in their denial that extraordinary remittances would operate to depress the value of the English currency on the exchanges and in their insistence that in the absence of currency changes the demands of England and the rest of the world for each other’s products would necessarily so immediately and completely adjust themselves to extraordinary remittances as to result under both a metallic and an inconvertible paper standard in the maintenance of equilibrium in the balance of payments without the aid of specie movements, changes in the relative level of prices in the two areas, or movements of the exchange rates. The theoretical grounds for holding these views to be erroneous are presented at length in a later chapter.
*71 Silberling and Angell, moreover, have shown, in the case of Silberling by a comparison of the English foreign remittances with the price of silver in English paper currency, and in the case of Angell by a comparison of the premium on silver and the Hamburg exchange, that there was a close correlation between these remittances and the status of the English currency. These comparisons are reproduced in table I and chart I. Adequate data do not exist to permit a tabulation of the international balance of payments of Great Britain for the period, or even of its trade balance. In the absence of such data, it is reasonable to assume that the extraordinary remittances are a fair presumptive index of the degree of pressure operating to force upwards the foreign exchanges and the price of silver in terms of English paper currency. The correlation shown is in fact closer than could reasonably have been expected, given the partial character of the data made use of for the purpose of the comparison, and I know of no equally striking results from similar comparisons for other countries or periods. Whether by design or by accident the English paper currency remained at or near parity with silver and with foreign metallic currencies in the years in
which no, or small, foreign remittances had to be made, and departed from parity in roughly corresponding degree in the years in which heavy foreign remittances were necessary.
*72

Ricardo could, however, have conceded to his opponents the point that extraordinary remittances tend to depress the exchanges without surrendering his main contention that actual depreciation of the exchanges was evidence of greater issue of currency than could be maintained in circulation under a metallic standard. Extraordinary remittances tend to depress the exchanges alike under a metallic and under a paper standard, but under a metallic standard this depreciation is prevented from going beyond the gold export point by contraction of the currency. If speculative factors be abstracted from, or if it be assumed that their mode of degree of operation is not affected by extraordinary remittances, then such remittances,
if accompanied by equal contractions of the currency of the remitting country in the two cases, should not result in appreciably greater
*73 depreciation of the exchange under a paper than under a metallic standard. It was primarily because under the paper standard the English currency was not contracted as it would necessarily have been contracted under a metallic standard that the foreign remittances resulted in such marked depreciation of the paper pound on the exchanges.

If Wheatley and Ricardo erred in their exposition of the relation under inconvertible currency between the exchange rates and the state of the balance of payments, the anti-bullionists erred more grievously. The anti-bullionists insisted rightly that under inconvertibility the exchanges were immediately determined solely by the demand for and supply of foreign bills, but failed to see that this was equally true of a metallic standard and that a very important factor determining the relative demand for and supply
of foreign bills was the relative level of prices in the two countries, which in turn was determined largely by the relative amounts of currency. Many of the anti-bullionists, moreover, must have thought that in some way a fall in the foreign exchanges made possible the payment of foreign remittances without the need of a commodity export surplus. No other explanation is available of their repeated insistence that throughout the period of low exchanges England either had an unfavorable balance of payments on trade account, or else had a balance insufficiently favorable to offset the military expenditure abroad. As Ricardo pointed out, in reply to reasoning of this sort by Bosanquet, this left it a mystery how the military expenditures abroad were actually met, since specie was not available.
*74

Not all the anti-bullionists, however, were confused on this point. One of them stated very compactly and clearly the possibilities under such circumstances:

[Under a depreciated paper currency] it would be literally impossible that the balance of payments should be any longer against us, because we could have no means of paying an unfavorable balance. Our receipts from, and payments to, foreign nations must therefore be reduced to an equality (or the balance must be turned in our favor) either by an increase of our exports of merchandise, a diminution of our imports and of the foreign expenditure of government, or by some…international transfers of capital….
*75

Another anti-bullionist, Herries, explained that foreign remittances could exceed the export surplus for a time if the balance was met by borrowing abroad, and, writing no doubt from first-hand knowledge, since he had been engaged in the task of making the remittances for the government, said: “This is, probably, the case, with respect to our drafts from abroad at this time:—we are borrowing money to carry on our foreign expenditure, at a high rate of interest.”
*76

There are passages in Silberling’s critique of the bullionists which seem to indicate that Silberling also subscribes to the notion that the fall in the exchange value of a remitting country’s currency
can operate to supply it with foreign funds with which to meet its foreign liabilities in some other way than by stimulating its exports and restricting its imports.
*77 Silberling cannot consistently fall back on the argument that a decline in the exchanges under inconvertible currency would lead to a debt-liquidating shipment of bullion, for he and Angell have characterized this as one of the erroneous doctrines of the bullionists, and especially of Ricardo. As I have elsewhere shown,
*78 Ricardo distinguished carefully between an inconvertible paper currency depreciated in terms of bullion and one not so depreciated—a distinction which Silberling and Angell fail to make—and denied the possibility of bullion shipments as a part of the regular mechanism of adjustment of international balances in the case of the former. Curiously enough, both Silberling and Angell place some emphasis on bullion shipments as part of the regular mechanism of adjustment of international balances in the case of the former. Curiously enough, both Silberling and Angell place some emphasis on bullion shipments as part of the explanation of the phenomena of the Restriction period, and tacitly, and probably wrongly, assuming that the Bank of England’s gold losses were mainly to the government, that when the Bank sold gold it ordinarily did not charge the market price, and that most or all of the gold exported while the Restriction was in effect came from the Bank’s holdings instead of from private stocks, cite these bullion shipments as an item in the meritorious record of the Bank of England during the period of suspension of cash payments.
*79

The notion that even under
depreciated inconvertible paper exchange fluctuations will give rise to bullion movements as an ordinary everyday occurrence is not as absurd on
a priori grounds as Silberling and Angell regard it for
any inconvertible currency.
While there is no internal demand for bullion for monetary purposes at a market price in excess of the mint price, there still remains an internal demand for industrial use, for hoarding, and for speculative purposes. A rise in the paper premium on bullion resulting from a fall in the exchanges will operate to induce some of the holders of bullion to offer it for sale for export. There is considerable evidence, both for the Restriction period in England and for other past and present cases of countries with depreciated paper currencies, that, where legal restrictions do not prevent, bullion moves fairly freely into and out of such countries in response to changes in its paper price.
*80 It is quite conceivable that the net export of bullion from England during the suspension period was even greater than it would have been if the metallic standard had been retained, and that the absence of the direct debt-liquidating effect of bullion shipments cannot therefore be invoked as even a partial explanation of the depreciation of the paper currency.

VIII. THE POSSIBILITY OF EXCESS ISSUE BY BANKS

There were among the anti-bullionists some crude inflationists for whom no amount of currency could be too great. Most of the anti-bullionists, however, recognized that there were limits beyond which it was not desirable to go in the issue of currency. What these limits were, they failed to specify, except in terms of the “needs of business.” They claimed that as long as currency was issued only by banks, and was issued by them only in the discount of genuine and sound short-term commercial paper, it could not be issued in excess of the needs of business, since no one would borrow at interest funds which he did not need. If currency should perchance be issued to excess, it would rapidly return to the banks either in liquidation of bank loans or, under convertibility, for redemption in specie.
*81 To this doctrine the directors of
the Bank of England and prominent members of the Cabinet also subscribed, and the authority of Adam Smith was appealed to in support thereof.
*82

The bullionists explicitly denied the validity of this doctrine, at least for an inconvertible currency. Thornton in 1797 had objected against the usury laws that they limited the Bank of England to a rate at which “there might be a much greater disposition to borrow of the Bank…than it might become the Bank to comply with,”
*83 and in 1802 he pointed out that the extent to which the charge of interest acted as a check on the demand for discounts depended on the rate of interest which was charged; the Bank of England was prevented by the usury laws from charging more than 5 per cent, and if the prevailing rate of commercial profit were higher than that, the demand for loans would be greater than the Bank should meet.
*84 Lord King put it more strongly: when the market rate of interest exceeds the bank rate, the demand for discounts “may be carried to any assignable extent,”
*85 and in this somewhat extreme form it was repeated by other bullionists.
*86 In a speech in the House of Commons on May 7, 1811, Henry Thornton expounded with great ability, and with interesting references to the experience of other central banks, the mode of operation of the rate of interest as a regulator of the volume of note issue. He pointed out that even John Law’s
bank had issued only on loans at interest, and that it was Law’s error that “he considered security as every thing and quantity as nothing” and failed to see the significance of the rate at which he offered to lend.
*87 Thornton argued, moreover, that during a period of rising prices the real rate of interest was less than the apparent rate; while businessmen did not generally perceive this, they did realize that borrowing at such times was usually profitable, and therefore increased their demands for loans if the bank rate did not rise.
*88

Ricardo agreed with the other bullionists that the “needs of commerce” for currency could not be quantitatively defined, and that through a resultant change in prices commerce could absorb whatever amount was issued.
*89 But he ordinarily denied any relationship between the rate of interest and the quantity of money, and presumably also between the rate of interest and the demand for loans: “Whilst the Bank is willing to lend, borrowers will always exist, so that there can be no limit to their overissues, but that which I have just mentioned,” i.e., convertibility.
*90 In a speech in Parliament he expressly denied that the rate of interest was a check to the amount of issues: “For…what the directors thought a check, namely, the rate of interest on money, was no check at all to the amount of issues, as Adam Smith, Mr. Hume, and others had satisfactorily proved.”
*91 Here once more Ricardo was applying long-run considerations to a short-run problem. But in his
Principles we find Ricardo at one point expounding the same views as the other leading bullionists:

The applications to the Bank for money, then, depend on the comparison between the rate of profits that may be made by the employment of it, and the rate at which they are willing to lend it. If they charge less than the market rate of interest, there is no amount of money which they might not lend,—if they charge more
than that rate, none but spendthrifts and prodigals would be found to borrow of them. We accordingly find, that when the market rate of interest exceeds the rate of 5 per cent at which the Bank uniformly lends, the discount office is besieged with applications for money; and, on the contrary, when the market rate is even temporarily under 5 per cent, the clerks of that office have no employment.
*92

To the denial by the bullionists that the charge of interest on loans was a sufficient guarantee, irrespective of the rate charged, against overissue, the anti-bullionists apparently never attempted to reply.
*93 In evidence before the Bullion Committee, Bank of England officials had emphatically denied that the security against overissue by the Bank would be reduced if the discount rate were to be lowered from 5 to 4 or even to 3 per cent. No person, they insisted, would pay interest for a loan which he did not need, whatever the rate, unless it were for the purpose of employing it in speculation, “and provided the conduct of the Bank is regulated as it now is, no accommodation would be given to a person of that description.”
*94

That the quantity of bank loans demanded is dependent on the rate of discount is now universally accepted by economists and need not be further argued. On the question whether or not the rate of 5 per cent uniformly charged by the Bank of England during the Restriction was lower than the market rate, there is,
however, a conflict of opinion. The usury laws would operate to prevent any overt charge of more than 5 per cent, and the uniform 5 per cent rate which is often said to have prevailed during the Restriction period
*95 may have been only nominally that. There is contemporary evidence that bankers found means of evading the restrictions of the usury laws. In 1818, the Committee on the usury laws stated in its Report that there had been “of late years…[a] constant excess of the market rate of interest above the rate limited by law.”
*96 Thornton notes that borrowers from private banks had to maintain running cash with them, and borrowers in the money market had to pay a commission in addition to formal interest, and that by these means the effective market rate was often raised above the 5 per cent level.
*97 Another writer relates that long credits were customary in London and a greater discount was granted for prompt payment than the legal interest for the time would amount to.
*98

More convincing evidence that the 5 per cent rate was not of
itself always an effective barrier to indefinite expansion of loans by the banks is to be found in the fact that the directors of the Bank of England, although they professed that they discounted freely at the rate of 5 per cent all bills falling within the admissible categories for discount,
*99 in reply to questioning admitted that they had customary maxima of accommodation for each individual customer and occasionally applied other limitations to the amount discounted.
*100

Even if it were conceded that the Bank rate was never lower than the market rate of discount for the same classes of loans, it might still be low enough to permit or even to foster a wild inflation, if the Bank rate was low absolutely, and if it was the Bank rate which determined the market rate. On important classes of loans the Bank of England was a direct competitor with other lending agencies, and it was certainly important enough as a lending agency to exercise at least an important influence on the market rate. Also, by lowering its credit standards, or offering its credit to a wider range of applicants for commercial loans, it could actively promote currency expansion without lowering its interest rate below the hitherto prevailing level. It may be accepted, therefore, that the 5 per cent rate was not necessarily an adequate check to the volume of bank credit extended to commercial borrowers.

The powers of the Bank of England to expand its note issues, moreover, were not confined to its commercial discount activities. The Bank could also, and did, get its notes into circulation by advances to the government, by purchases of exchequer bills and public stocks in the open market, and by advances to investors in new issues of government stocks. Since even many of the anti-bullionists conceded that there was no automatic check to excess issue where the issues were made in connection with loans to the government, there should have been no occasion for extended controversy as to the existence of a possibility of excess issue.
*101

IX. RESPONSIBILITY FOR THE EXCESS ISSUE: BANK OF ENGLAND VS. COUNTRY BANKS

Since bank notes were issued both by the Bank of England and the country banks, responsibility for any excess issue of paper currency could lay with either or with both. With the exception of Wheatley, who held the country banks largely responsible,
*102 the bullionists were united in assigning responsibility for the excess, as between the Bank of England and the country banks, wholly or predominantly to the former. Boyd, in 1801, laid down the formula which was to be the text of the bullionists: “The Bank of England is the great source of all the circulation of the country; and, by the increase or diminution of its paper, the increase or diminution of that of every country bank is infallibly regulated.”
*103 His argument rested on the postulate that the country banks must keep a fixed percentage of reserves against their own note circulation in Bank of England notes, whereas the Bank of England was not subject to such a limitation.
*104 In a note added to the second edition, he conceded that the country banks, by allowing their reserve ratio to fall, may have contributed independently to the then existing excess of currency, but he blamed the Restriction, which left to holders of country bank notes the possibility only of converting them into Bank of England notes with which they were not familiar, for making this fall in reserve ratio possible. He apparently believed that once this fall had taken place, the Bank of England would again have control, through
regulation of its own issues, over the volume of country bank issues.
*105

Thornton reached the same conclusion, that the volume of Bank of England issues regulated the volume of country bank note issues, but by a more elaborate chain of reasoning. He applied to different regions within a country the Hume type of analysis of adjustment of international balances of payments.
*106 If country banks took the initiative in increasing their issues, country prices would rise; the provinces would buy in London commodities which formerly they had bought locally; there would result an adverse balance of payments on London, which would be met through shipment of Bank of England notes to London or by drafts on the balances of country banks with London bankers. The impairment of their reserves would force the country bankers to contract their note issues.
*107

Thornton pointed out, however, that this did not mean that the proportion of country bank notes to Bank of England notes must always remain the same. This would hold true only if the areas of circulation of the two types of notes and also the relative volumes of payments to be made in the respective areas remained unaltered:

By saying that the country paper is limited in an equal degree, I always mean not that one uniform proportion is maintained between the quantity of the London paper and that of the country paper, but only that the quantity of the one, in comparison with the demand for that one, is the same, or nearly the same, as the quantity of the other in proportion to the call for the other.
*108

Similar views were expressed by Horner,
*109 King,
*110 Ricardo,
*111 the Bullion Committee,
*112 Malthus,
*113 and other bullionists.

Since the anti-bullionists denied any excess in the currency as a whole, they ordinarily showed little interest in the attempts of the bullionists to apportion responsibility for such excess. Some of the anti-bullionists agreed that convertibility of country bank notes into Bank of England notes was as effective a restriction on country bank note issue as convertibility into gold.
*114 Others of them, however, apparently determined that if any blame was to be assigned it should not be to the Bank of England, denied that the amount of country bank notes was in any way dependent on the amount of Bank of England notes, and cited in confirmation the evidence of country bankers before the Bullion Committee that their reserves consisted only slightly of Bank of England notes and the apparent absence of correspondence between the fluctuations in the issues of the two types of paper money.
*115

Silberling and Angell reject the bullionist claim that the country bank note circulation was dependent on that of the Bank of England notes. Silberling ridicules the notion that if prices rise
in the provinces, it will result in a shift of purchases from the provinces to London: “London and the rest of England were not then, and are not now economic areas producing identical wares. If the price of iron or hops or wool rose in the provinces by reason of liberal credit accommodation to farmers and speculators,…it could not result in purchases from London of what London did not produce.”
*116 This is a valid criticism of the manner in which the bullionists expressed their argument, but leaves the essence of the argument untouched. A relative increase in country bank note issues will not lead the provinces to increase their purchases in London of
country products, but it will, nevertheless, lead to a debit balance of payments of the country with London. The increase in spendable funds in the country will lead to increased purchases of
London products by the provinces, and the rise in prices in the provinces will lead to decreased purchases of
country products by London. When two regions have currencies convertible into each other at fixed rates and have commercial relations with each other, one of these regions cannot issue currency to any extent, irrespective of what the other does, without encountering serious exchange and balance of payments difficulties, even if the two regions do not have a single identical product.
*117

Silberling and Angell object further that the explanation given by Thornton and Ricardo is unilateral, instead of bilateral; it fails to take account of the upward effect on London prices of the release by the country to London of Bank of England notes and
balances with London bankers resulting from an expansion of country bank note issues. They contend, in rebuttal, that a rise in prices in the country resulting from an increased issue of country bank notes would spread to London.
*118

This is a valid criticism of Thornton.
*119 It is not applicable, however, to Boyd
*120 or Ricardo,
*121 for both of these writers took it for granted that it was necessary for the country banks to maintain constant cash reserve ratios whether in Bank of England notes or not. If the Bank of England did not increase its issues, then the country banks could not at the same time increase their circulation and maintain a constant reserve ratio. It is this assumption of constancy in the country bank reserve ratios, to which neither Silberling nor Angell refers, which is the vulnerable point in the bullionist argument. If, as Boyd conceded, the country banks allowed their reserve ratios to fall, they could, as long as their reserves were not wholly exhausted, force their issues even while the Bank of England remained passive. If they tolerated a lowering of their reserve ratio, they could bring about a new price equilibrium and a new equilibrium in the balance of payments between London and the provinces, with the circulation greater, and prices higher than before, in each area. Even if the country banks expanded rapidly and extravagantly, and the Bank of England did not follow suit, it might be some time, as Joplin later pointed out,
*122 before their reserves were exhausted, and in the interval before the collapse prices would be higher and the premium on bullion greater for England as a whole. The question still remains, however, as to what were the obligations of the Bank of England in such a situation.

Silberling further supports his argument that the issues of the country banks were not dependent upon those of the Bank of England by the claim that the country bank reserves consisted mainly of balances with London private bankers, while the reserves of the London bankers “were wholly uncontrolled by law and had never been more than very moderate sums; and their
ability to create credits was now but very little controlled by the Bank of England.”
*123 The London bankers, unless they were of a banking species hitherto unrecognized, must, in practice, have found it necessary to have on hand in case of need cash or its equivalent. But the only “cash” at the time was Bank of England notes, and its only equivalent at the time was a demand deposit with the Bank of England. The private bankers in London in fact began during the Restriction period the practice of opening accounts at the Bank of England and of rediscounting bills in their portfolios with the Bank, instead of, as before, selling exchequer bills or government stock on the open market, when they needed to replenish their cash reserves.
*124 The then deputy governor of the Bank admitted to the Bullion Committee, in reply to a searching question on this point, that a considerable amount of the bills discounted with the Bank of England by the London private banks was country bank paper.
*125 Willingness of the London bankers to allow their cash balances to run down would enable them to expand their credits to country banks in some degree, even if the Bank of England did not make available to them increased rediscount facilities. But since such expansion would involve a persistent drain of their cash to the Bank of England and to hand-to-hand circulation, it could not have been carried far without active Bank of England support. The Bank of England, moreover, could, by positive action, have prevented even such expansion of the volume of discounts of the London private banks as had been independent of increased discounts with the Bank of England.

Silberling and Angell fail completely to give any consideration to the proposition that while England had an inconvertible paper currency special responsibility attached to some agency, and presumably to the Bank of England as in effect a central bank, to keep the currency in good order, even if to do so it should prove necessary to countervail the activities of the country banks and the London private bankers. Silberling even goes to the length of characterizing as a “truly remarkable opinion,” unfortunately, however, without indicating why, Ricardo’s argument (as summarized
by Silberling) that “one of the causes of the ‘excess’ of Bank notes was the expansion of the country issues, which had thereby narrowed the field within which the Bank’s issues could circulate; the latter overflowed, in other words, a contracted channel.”
*126 The Bank of England, it is true, was organized as a profit-making establishment. But it enjoyed valuable special privileges, and whatever some of its shareholders may have thought,
*127 it was the general opinion of the time that it also had special obligations, what we should today term the obligations of a central bank. Silberling himself refers to the Bank of England of that period as a “central bank,” and states that the Bank claimed to be a “regulator” of the currency. The Bank could not plead financial inability to carry out these obligations, for the “supposedly enormous profits,” to which Silberling refers in a manner clearly intended to suggest that they existed only in the imagination of the bullionists, were genuine.
*128 There is nothing obviously remarkable in the proposition that a central bank should contract when the rest of the banking system is dangerously expanding, in order to check and to offset that expansion. It should, on the contrary, be obvious that there is a fatal conflict between the regulatory functions of a central bank and determination on its part to maintain, willy-nilly, its accustomed proportion of the country’s banking business.
*129

The Bank of England could have followed any one of three alternative lines of policy with respect to the relationship between its own issues and those of the country banks, which can be distinguished as (1) regulatory, (2) passive or indifferent, and (3) sympathetic. If it followed a regulatory policy, this should show itself in a negative correlation between the fluctuations in the two types of issues, with the changes in the Bank’s issues lagging behind those of the country banks. If the Bank took a passive or indifferent attitude toward the operations of the country banks, there should be no marked correlation, positive or negative, unless:
(a) the country banks, either from policy or from necessity, followed the Bank of England, when there should be a positive correlation between the fluctuations in the two types of issues, with the changes in the country issues lagging after the changes in the Bank issues; or (b), both the Bank and the country banks responded to the same factors in the general situation pulling for credit expansion or contraction, when there should be positive correlation between the fluctuations in the two types of issues, with the existence and the character of a lag depending on the time-order in which London and the provinces, respectively, felt the stimuli to expansion or contraction and the rapidity with which they responded to such stimuli. Finally, if the Bank followed a sympathetic policy, there should be positive correlation between the changes in the two types of issues, but with the changes in the Bank of England issues lagging after the changes in the country bank issues. This does not, of course, exhaust the range of possible relationships, since the types of relationship distinguished above need not in practice have been mutually exclusive, but could have been present in varied and varying combinations.

From his examination of the statistical data Silberling concludes that “the quarterly cyclical fluctuations in the country notes preceded…the discounts of the Bank of England (a much more accurate measure of accommodation than their notes).”
*130 If this were the case, it would indicate that the Bank of England either had followed the “sympathetic” policy toward country bank issues, surely the least defensible of all if it was its function to keep the currency in order, or had had no policy at all but had reacted in the same way as the country banks, but more slowly, to the forces operating in the country at large to bring about a currency inflation. Silberling nevertheless presents this conclusion as an important element in his exoneration of the Bank from blame.

Angell, using Silberling’s data, finds that the Bank of England note circulation “was a comparatively stable element” and that
“the great element of fluctuation in the volume of currency was, rather, the issues of the country banks. These issues usually expanded greatly before and during a rise in prices, while they contracted even more abruptly before and during a fall,”
*131 i.e., the Bank of England followed a passive policy.

The statistical conclusions of both writers rest, unfortunately, on faulty data with respect to the country bank note circulation. There was no record of the actual amounts of notes issued, but the notes had to carry tax stamps, and all contemporary estimates were based on the official statistics of the tax stamps, sold by the government, and on the estimated average life of the notes. Country bank notes were subject to tax only upon their original issue. Subject to some complex qualifications, prior to 1810 these notes could not be reissued after three years from the date of their original issue. This limitation was removed in 1810, on the assumption that on the average the notes would, because of wear and tear, have a life of about three years. If the notes could be presumed to last, prior to 1810, on the average for three years, if after 1810 all the notes could be presumed to last for the full three years, and if the country banks always succeeded in maintaining in circulation the full amount of notes for which they had purchased stamps, then the circulation at the beginning of any quarter would be equal to the amount of notes for which stamps had been sold during the preceding twelve quarters. There was no available mode of estimating the circulation which did not necessitate making doubtful assumptions of this kind.
*132 Silberling’s estimate of country bank note circulation, which Angell also uses, has, moreover, a special and catastrophic defect of its own. It consists merely of the amount, for each quarter, of £1 and £5 notes for which stamps had been sold in such quarter,
arbitrarily multiplied by ten, i.e., with the decimal point moved one place to the
right, presumably as the result of an error in copying.
*133 It bears no resemblance in its fluctuations to the other available estimates of country bank note circulation, as the following table shows:

Silberling claims for his series that “since this stamp duty involved expense to the issuing bankers, it is wholly probable that
the volume of notes stamped each quarter affords a safe index, at any rate, of the variability of the actual issues.”
*134 But Silberling overlooks that the amount of stamps issued each quarter indicates at best only the amount of
new notes which were issued during that quarter. Since it gives no indication of the amounts of
old notes which went out of circulation during that quarter, it is a wholly unreliable index of the
net change during the quarter in country bank note circulation. Silberling’s series, as its method of compilation would lead one to expect, shows much greater quarter-to-quarter and year-to-year variability than do the other available estimates of country bank note circulation. These last do not indicate any appreciably greater instability in country bank note than in Bank of England note circulation. But even these other estimates are probably too defective to warrant any confidence on conclusions based on their use.
*135

X. RESPONSIBILITY FOR EXCESS ISSUE: THE CREDIT POLICY OF THE BANK OF ENGLAND

Silberling finds other statistical evidence of the high quality of the Bank of England’s management of its affairs during the Restriction: “the Bank’s loans to the State tended to expand when discounts were moderate, and vice versa. In other words, the Bank granted accommodation to the government during the war rather sparingly and according to the state of their mercantile accounts. They put the business interests foremost and assumed a primary responsibility for the maintenance of British trade and industry, which, in an essentially commercial war, was of vast consequence.”
*136 But as can be seen from chart II and table III, all that the data show is that there was somewhat of an inverse correlation between the short-run changes in commercial discounts and advances to the government. As they stand, the data will equally well support the conclusion that commerce got only what was left after the government’s demands had been satisfied. An even more plausible explanation of the inverse correlation between commercial discounts and advances to government, because it does not involve the attribution to the Bank of England
of any consistent regulatory policy, is that advances to the government supplied commerce with funds as effectively as, though less directly than, commercial discounts. When the government borrowed freely from the Bank, the borrowed funds flowed into commerce and consequently lessened the demand of businessmen for discounts.
*137 But while Silberling here praises the Bank for giving commerce a preference over the government, and for treating the latter only as a residual claimant for credit, he later attacks the bullionists for their alleged failure to recognize the extent to which the Bank’s expansion of its credit was due to the demands made upon it by the government.
*138

Angell, from the same data, derives a substantially different defense of the Bank’s operations. Instead of finding that the Bank treated the government as a residual borrower, he claims,
probably justly,
*139 that the Bank was not a free agent in deciding the amount of credit it should grant to the government, and concludes, from his analysis of the data, that the Bank in its commercial discounts, “that part of the credit extensions over which it had independent control,” exercised “a moderating policy, of restraint in boom times and of assistance in stringency…. The Bank of England, in so far as its independent and uncontrolled loan-extensions were concerned, was largely blameless.”
*140

There is substantial ground for accepting Angell’s plea in defense of the Bank’s advances to the government that the Bank with respect to these was not a free agent. It nevertheless had much greater scope for regulating the currency through control of its commercial discounts than it made use of. Ample facilities for direct credit to private business had developed during this period outside the Bank of England both in London and in the provinces, and it is by no means clear that there was any longer any urgent need, as far as the nation’s commerce and industry were concerned, for the Bank to grant any genuinely “commercial” discounts at all. Its “commercial discounts” increased, however, from 6.3 millions in 1800 to 15.3 millions in 1809 and
19.5 millions in 1810,
*141 amounts which were never again reached until after 1914! The proportion of the Bank’s commercial discounts to its total advances increased, from an average of approximately 25 per cent during the years 1794 to 1796 inclusive, to 33 per cent for 1797 to 1800, 42 per cent from 1801 to 1805, 50 per cent from 1806 to 1810, and 36 per cent from 1811 to 1815;
*142 in 1820, after resumption, it fell to 19 per cent, a level which it appears barely to have maintained during the 1820’s.
*143

Even these percentages apparently minimize the extent to which the Bank’s expansion of its credit facilities provided funds for use by private industry rather than by government. The government during the Restriction kept a large proportion of the advances to it by the Bank, for the years 1807 to 1816 exceeding 50 per cent on the average,
*144 on deposit with the Bank, presumably as an emergency reserve. The commercial discounts, on the other hand, were in the main drawn out immediately in cash, and the private deposits at the Bank of England during the same period averaged under 12 per cent of the commercial discounts.
*145 It is likely, therefore, that the funds resulting from the commercial discounts had a greater velocity of circulation and consequently,
pro rata, a greater influence on the level of prices, than the advances to the government.

Angell makes some attempt to determine the responsibility of price inflation in England for the fall in the exchanges and the premium on bullion by a comparison of their fluctuations with the fluctuations in the English price level, as shown by Silberling’s index number of English prices for the period, but with admittedly inconclusive results.
*146 Such a comparison is by its
nature without significance. Even a comparison of the fluctuations in the exchange rates with the
relative fluctuations of English to foreign price levels would not yield conclusive results unless there was very marked agreement or disagreement in the fluctuations.
*147

That resort to price inflation is necessary if a great war is to be financed seems to Silberling so axiomatic that without argument he makes their failure to acknowledge this one of the most heavily stressed counts in his indictment of the bullionists.
*148 There is no need to debate this issue here, but several considerations of which the bullionists were aware call for notice. Contemporary writers pointed out that England had financed successfully the Seven Years’ War and the American War, both of which involved fairly comparable financial strain, without resort to price inflation involving serious depreciation of the currency in terms of bullion. Napoleon financed his side of the war on a strictly metallic currency basis. There was, moreover, a substantial rise in English prices even in terms of gold and silver, and England could, therefore, have had a substantial inflation even if she had remained on the gold standard.