Chapter V
ENGLISH CURRENCY CONTROVERSIES,
1825-1865

The student who turns from the literature of the Heroic Age of British monetary controversy in order to attempt a study of the original sources relating to the antecedents of our modern banking situation will find himself confronted with a jungle of blue books and Parliamentary discussions, pamphlets and tracts and leading articles: a jungle at first sight so impenetrable that he may well despair. For it is characteristic of the period of middle-class ascendancy after 1832 that it produced much heat and little light; many massive volumes of evidence and statistics, but no classic reports; much legislation but, for a long time at least, no final solution of the various problems to be faced.

—T. E. Gregory, Select statutes, I, ix.

I. INTRODUCTION

The downward trend of the English price level, which persisted without any sustained reversal from 1815 to the 1850's, was for English industry and labor only partially compensated by the progress in manufacturing technique and the fall in the prices of imported raw materials. The occasional prosperous intervals were ordinarily terminated by sharp financial crises, and were followed by intervals of depression and unemployment. There was general agreement that these business fluctuations were inherent in the new structure of industry, but there was also a widespread conviction that they had been accentuated by chronic mismanagement—or misbehaviour—of the currency. It became apparent soon after resumption of cash payments that strict adherence to a fixed metallic standard was not sufficient to assure the smooth and beneficent working of the currency system. The Bank of England succeeded throughout the period in maintaining convertibility of its paper notes, but on several occasions only with great difficulty and after resort to emergency measures. In 1825, in 1836, and again in 1839, suspension of convertibility was avoided only by a narrow margin. In 1847, and 1857, and 1866, the Bank was again in serious difficulty. Each period of special strain gave rise to an extensive controversy, turning on the quality of the Bank's management of its affairs and on the principles which should be followed in the management of the currency.^*1 That the currency was operating badly no one disputed, although there were not a few who would have agreed with Cobden that "managing the currency [was]...just as possible as the management of the tides, or the regulation of stars, or the winds,"^*2 and that all that government could do, therefore, would be to place it on a wholly metallic basis, and then let "automatic" processes run their course.

During this period the English banking system underwent important structural changes. In the belief that it was mainly the small notes which were presented for redemption in gold during periods of financial stress, the renewal in 1822 of the right to issue notes under £5 was repealed by an act passed in 1826. To promote the establishment of financially stronger country banks, an act of the same year authorized the establishment in the provinces of note-issuing banks with an unlimited number of partners. An act passed in 1833 exempted the Bank of England, in so far as its discounts of short-term paper were concerned, from the legal maximum interest rate of 5 per cent under the usury laws, and thus gave it the power to use the discount rate as an instrument of credit control. This act also made the Bank's notes legal tender except by the Bank itself as long as the Bank maintained convertibility into gold, and definitely proclaimed—what had previously been questionable—the right to establish in the London area non-note-issuing banking establishments with more than six partners. The joint-stock banks grew rapidly in number and in importance, both in the provinces and in London. By the 1850's there were at least three joint-stock banks in London with deposits in excess of those of the Bank of England. The Bank of England had gradually given up most of its ordinary commercial discount business, and its "private deposits" came to consist largely of bankers' balances held with the Bank as the equivalent of cash reserves.

Finally, there was passed the Bank Charter Act of 1844, commonly referred to as Peel's Act, which was to remain the charter of English banking until the Great War. This act required the Bank formally to segregate the issue department from the banking department,^*3 and limited the issue department to a maximum issue of notes uncovered by bullion of £14,000,000, above which amount it could issue notes only in exchange for gold (or, within certain limits, silver). Country banks then issuing notes were permitted to continue such issue not to exceed the amount then in circulation, but the law made provision for the gradual absorption by the issue department of the Bank of England, as an addition to its uncovered issue, of the bulk of the note circulation of the country banks. Except as to note issue, the banking department of the Bank of England was left wholly free from statutory regulation, as were also the then existing private banks on all matters of consequence except the right to issue notes.

II. THE "CURRENCY SCHOOL"-"BANKING SCHOOL" CONTROVERSY

The currency controversies of this period were carried on mainly by the members of two groups, with divergent views, who came to be known as the "currency school" and the "banking school," respectively.^*4 The most prominent members of the currency school were Lord Overstone (Samuel Jones Loyd), G. W. Norman, R. Torrens, and William Ward. Thomas Tooke, John Fullarton, James Wilson, and J. W. Gilbart were the leaders of the banking school. There was not complete unanimity of doctrines within each group, and the currency school, under the impact of their opponents' criticisms, modified their doctrines fairly substantially in the course of the controversy. An attempt is nevertheless made in the next few pages to summarize the general position of the two schools, as a preliminary to a more detailed examination of such of the particular doctrines expounded in the course of the controversy as are of importance for the theory of international trade. The discussion between the two schools turned wholly, however, on short-run issues. On the question of what determined the quantity and the value of a metallic currency in the long run, both schools followed the "classical" or "Ricardian" doctrines.

The currency school maintained that under a "purely metallic currency" any loss of gold to foreign countries or any influx of gold from abroad would result immediately and automatically in a corresponding decrease or increase, respectively, in the amount of currency in circulation. The actual currency was a "mixed currency," that is, convertible paper notes were a constituent element of the currency. A mixed currency would operate properly only if it operated precisely as would a metallic currency, i.e., only if any efflux or influx of gold resulted in a corresponding (absolute, not proportional) decrease or increase in the quantity of the currency—the "currency principle." But a mixed currency would not operate in this manner automatically and immediately unless the issue of paper money were deliberately regulated so as to make the changes in its quantity conform to the changes in the quantity of gold held by the issuing agencies. In the absence of such regulation, paper money would at times be issued to excess, at other times unduly contracted; the maintenance of convertibility would not be definitely assured; the improper fluctuations in the currency would accentuate the tendency inherent in the economic structure toward recurrent booms and crises.

Since the ultimate objective of the currency school was that the value of the monetary unit, or the level of prices, should be the same under a "mixed currency" as it would be under a purely metallic currency, this could be accomplished by their rule of making the fluctuations in the amount of bank notes correspond to the fluctuations which would occur in the amounts of specie under a purely metallic currency only if the velocity of circulation of bank notes and of specie would under like circumstances be identical. This was apparently overlooked by the members of the currency school,^*5 although it may be that they took for granted that there would be such identity.^*6

The banking school denied almost all of these propositions. Generally waiving the question as to whether it was desirable that a mixed currency should operate precisely as would a purely metallic currency, they denied that a purely metallic currency would operate in the manner claimed by the currency school. They pointed out that under a purely metallic currency there existed in addition to specie, and under a mixed currency there existed in addition to specie and paper notes, a large quantity of bank deposits and bills of exchange which, they claimed, were also "currency" and in any case operated on prices in the same manner as did bank notes and specie. Under a purely metallic currency, moreover, some of the gold was not in circulation, but was in "hoards," in modern times held mainly in the bullion reserves of the Bank of England and other banks. Changes in the amounts of these hoards could not possibly have any effect on prices.^*7 Even under a purely metallic currency, therefore, a gain or loss in the nation's stock of gold need not result in corresponding fluctuations of the currency, but might merely change the amount of gold in hoards, or might be offset by an inverse fluctuation in the amount of deposits. Without control of hoards and of deposits, limitation of the note issues could not suffice, therefore, to attain the objective of the currency school of enforcing correspondence between the fluctuations in the total circulation and the fluctuations in the total stock of gold. The banking school did not present an alternative program of statutory control of the currency. They held that statutory control of the deposits was not demanded by anyone, was impossible, and even if possible was undesirable. The amount of paper notes in circulation was adequately controlled by the ordinary processes of competitive banking, and if the requirement of convertibility was maintained, could not exceed the needs of business for any appreciable length of time—the "banking principle." If unsound banking practices did occasionally lead to excess grant of credit, this brought its own corrective penalties. In any case it could not be prevented by legislative measures, and especially by mere limitation of note issue.

The bullionists, it will be remembered, had insisted that under an inconvertible paper money currency the issues should be so regulated as to conform to the aggregate circulation of specie and paper which could be maintained under a convertible currency, but usually maintained—or took it for granted, without argument—that if the requirement of convertibility were enforced there was no need of further regulation to insure against excess—or deficient—issue of paper money.^*8 The anti-bullionists, on the other hand, had ordinarily maintained that a paper currency could not be issued to excess whether convertible or not, if issued only by banks as loans on the security of good short-term commercial paper. The currency and the banking schools both rejected the anti-bullionist doctrine that an inconvertible paper money could not be issued to excess.^*9 The currency school went further; they claimed that even a convertible paper currency could be issued to excess, not permanently, but for sufficiently long periods to endanger the maintenance of convertibility and to generate financial crises. The "currency principle," i.e., the doctrine that a mixed currency should be made to operate as would a "purely metallic" currency, did resemble, however, the bullionist doctrine that an inconvertible paper currency should be made to operate as would a convertible currency, and was obviously derived from it.

The currency principle appears first to have been formulated during the 1820's. Joplin, in 1823, proposed a system of regulation of the issue of paper notes whose essence was the requirement of 100 per cent bullion reserves, so that "a paper circulation, by this system, would dilate and contract precisely in the same manner as a metallic currency."^*10 Henry Drummond, in 1826, similarly urged that the amount of paper money should be kept constant, so that all variations in the quantity of the currency should consist of corresponding variations in the quantity of specie.^*11

III. THE "PALMER RULE"

In 1827 the Bank of England adopted a rule—later commonly known as the "Palmer rule" or the "rule of 1832" because it was first publicly explained, in 1832, by the then governor of the Bank, J. Horsley Palmer—which aimed at making the fluctuations in the English currency conform with those which would occur under a purely metallic currency by keeping its security holdings, including discounted paper, constant. At the same time, although apparently wholly independently, Pennington, a confidential adviser of the Treasury on currency matters, had recommended the same rule in a memorandum to Huskisson, then Chancellor of the Exchequer. The problem as Pennington saw it was to make the paper currency operate as would a "purely metallic currency": "The great objection to a paper currency arises from the extreme difficulty of subjecting its expansion and contraction to the same law as that which regulates the expansion and contraction of a currency purely metallic."^*12 He offered as a solution that the Bank of England should be given an exclusive monopoly of note issue (or direct control over the issues of the other banks) and should hold a fixed amount of securities. There could then be no variations in its outstanding note circulation without corresponding variations in its holdings of bullion:

Nothing more would be necessary than that the bank should constantly hold a fixed amount of the same unvarying species of securities. If its outstanding liabilities amounted, at any particular time, to £26,000,000, and if, against these, it held £18,000,000 of government securities and £8,000,000 of bullion, then, by confining itself to the £18,000,000 of securities, the action of the foreign exchange would necessarily turn upon the gold: at one time the bank might have six, at another time ten, and at another eight millions of treasure; and in all cases, its paper would contract and expand according to the increase or diminution of its bullion.^*13

The Palmer rule was essentially the same. When the exchanges were at par and the currency "full," the Bank should try to have a bullion reserve of one-third of its combined note and deposit liabilities, so that its current assets should be one-third bullion and two-thirds securities. Thereafter all that would be required would be to maintain the securities at a constant amount. An influx of gold from abroad would thus act to increase the note circulation by a corresponding amount; an efflux of gold or a demand for coin for internal circulation would result in a corresponding decrease in note circulation. The internal circulation, specie plus paper, would thus remain constant unless acted upon by external gold movements.^*14

This rule had the fatal defect that it took no account of the necessity of also maintaining deposits constant if the maintenance of securities at a constant amount was to guarantee correspondence between the fluctuations in bullion and the fluctuations in note circulation. If the deposits were permitted to fluctuate, then as the bullion holdings fluctuated the note circulation might remain constant, or might fluctuate in the reverse direction. Under an inconvertible paper currency, a case could be made for the general policy of keeping the securities constant, if departure from this rule to offset variations in the velocity of circulation of notes and deposits were permitted, and if provision were made for adjustment of the amount of securities to the secular trend of production resulting from growth of population and capital. But under an international metallic standard, adherence to the rule of keeping the securities constant could lead to serious and lasting disequilibrium between the internal and the world price levels, and therefore to exhaustion or to indefinite accumulation of gold reserves.^*15

It is not easy to understand Pennington's original position in this connection. The objective of control stated by him seems to involve an unqualified acceptance of the currency principle. But the method of control which he advocated, like the Palmer rule, would have made the fluctuations in notes plus deposits correspond with the fluctuations in specie reserves, whereas the currency principle called for correspondence between the fluctuations in notes alone and the fluctuations in specie reserves. It especially seems to call for explanation that Pennington, who was an important factor in drawing attention to the importance of deposits as a means of payment, should have advocated a rule which would permit of withdrawal through the deposits of all the specie reserves without calling for any positive corrective action on the part of the Bank. When Pennington later, in 1840, published his memorandum, his views had apparently undergone some modification. He now made it appear that by a "purely metallic currency" he had meant one which consisted only of specie, and that by "paper circulation" he had meant notes and deposits.^*16 This would bring his rule of control into conformity with his objective. Against the currency principle proper he protested that it would make the fluctuations in the currency (= notes and deposits) exceed the fluctuations which would occur under a simple specie currency, with the result that "the public will be exposed to very great alternations of comparative ease and difficulty in the operations of the money market."^*17 What he now supported, apparently, was a provisional adherence to the Palmer rule, which would limit the fluctuations in notes and deposits to the fluctuations in specie reserves, with departure from it in the form of open-market sales of securities only when otherwise dangerous depletion of the specie reserves would occur.^*18

As was to be expected, the affairs of the Bank went badly while the Palmer rule was in operation. From 1836 to 1839 in particular, while the rule was presumably being followed, the Bank was in serious difficulties much of the time. The Bank found at times that gold was being withdrawn for export through the deposits without any compensating reduction in the note circulation. It also found itself unable—or unwilling—to keep its securities constant, and it even increased its securities while a drain of gold was under way. Its difficulties were due in part to misguided violations of its own rule,^*19 but in part they were due also to the utter impracticability of the rule under a metallic reserve currency whenever greater contraction of the currency (notes and deposits) should be requisite than the rule of keeping the securities constant would permit.

Torrens and Overstone were critical of the Palmer rule, although they held that the departures from it had been such as to accentuate rather than moderate its shortcomings. They pointed out that if the currency principle were to be carried out, gold movements should not be permitted to operate on the deposits alone. When the Bank found that its gold reserves were being drawn out through its deposits, it should have reduced its note circulation by "forcible operation on its securities," i.e., by deliberate contraction of its discounts or by sale of government securities in the open market. They held that maintenance of securities at a constant amount, instead of enforcing correspondence between the fluctuations in the amounts of bullion holdings and of note circulation, prevented the Bank from establishing such correspondence.^*20

According to Torrens and to Overstone, the error in the Palmer rule was that it aimed at keeping the whole of the Bank securities constant, including those upon which "the Bank lent its deposits," and that it permitted gold flows to act on the whole of the liabilities, including the deposits. The Bank, on the contrary, should keep constant only those securities upon which it put out its notes, i.e., should keep constant the amount of its uncovered note circulation. Only then would variations in the Bank's note circulation necessarily correspond with the variations in its bullion holdings.^*21 To enforce this procedure on the Bank, and to make certain that the securities held as backing for the notes should be segregated from those held as backing for the deposits, the banking and issue departments of the Bank should be formally separated, and the latter should be confined to the exchange of notes for bullion and of bullion for notes, pound for pound, except for a fixed maximum of notes to be covered by securities.^*22 The currency school undoubtedly wanted the note issue powers of the country banks to be withdrawn, or at least drastically restricted, but they did not enlarge upon this phase of the question,^*23 as a precaution, perhaps, against providing further stimulus to the already vigorous opposition of the country bankers to the currency school proposals.

IV. THE BANK ACT OF 1844

The Act of 1844 put into effect these proposals of the currency school. But any expectations which may have been held that the provisions of the act were sufficient to insure protection against currency disturbances were destined to meet with early disappointment. The Bank of England took too seriously the freedom from statutory regulation of its banking department under the act, and proceeded immediately to reduce its discount rate from 4 to 2½ per cent, the lowest rate in its history up to that time, and to expand its commercial discounts.^*24

Its reserves in the banking department soon began to fall. In 1847, the public, noting the decline in these reserves, and aware that under the Act of 1844 the Bank would be unable to meet the claims of its depositors with its own notes or with specie once the reserves of the banking department had been exhausted, took alarm, and proceeded to draw out their deposits. The Bank's attempts to check the drain by rationing, successive increases in the discount rate, sale of securities, and borrowing from the market, did not succeed. On October 22, 1847, the reserves in the banking department had fallen to £2,376,472, and a panic was in full sweep in the country. The Bank was still confident that it could continue to meet its payments, but the government, in order to allay the panic, stepped in, authorized the Bank to issue notes uncovered by gold in excess of the statutory maximum, and requested the Bank to discount freely, but at a high rate of interest. The panic ceased at once, gold began to flow back to the Bank, and no issue in excess of the statutory maximum was actually made. But it had been demonstrated that under the Act of 1844 gold could be withdrawn from the Bank by means of the deposits as well as by presentation of its notes for payment in specie, and that in a period of alarm knowledge that the power of the Bank to issue notes was legally restricted could operate to promote such withdrawal. In 1857 and 1866, suspension of the Bank Act was again invoked to prevent exhaustion of the reserves in the banking department. The Act of 1844 may have established an absolute guarantee of convertibility of the note issue, subject only to the condition that the amount of notes voluntarily remaining in the hands of the public did not fall below £14,000,000.^*25 It clearly failed to guarantee adequately good management of its credit operations on the part of the Bank of England.

The necessity of suspending the Bank Act three times within twenty-five years of its enactment was disappointing to its currency school advocates, but they denied that it justified the claims of the banking school that the currency school doctrines had been erroneous and that the act was injurious in its effects. Overstone even denied that the divergent fluctuations after 1844 in the note circulation of the Bank and in its bullion holdings disproved the contention of the currency school that the Act of 1844 would automatically enforce a correspondence between these fluctuations. He was able to show that when prior to the passage of the act he had supported the rule of forcing correspondence between the bullion holdings of the Bank and the note circulation in the hands of the public, instead of between the bullion holdings of the Bank and the notes outside the issue department, he had done so only because until 1844 information was not available as to the holdings of its own notes by the Bank as "till money." Had such information been available, he would have included notes held by the Bank in the banking department in the "circulation" whose fluctuations should be made to correspond with the fluctuations of the bullion holdings of the Bank.^*26 The Act of 1844 did guarantee absolute correspondence between the variations in the amount of notes outstanding at the issue department and the amount of bullion held by the issue department.

The great fault of the currency school was the exaggerated importance which they encouraged the public to attribute to the automatic regulation of the issue department as contributing to a proper functioning of the Bank of England as a whole. In his speech introducing the Bank Act in the House of Commons, Sir Robert Peel had stated: "With respect to the banking business of the Bank, I propose that it should be governed on precisely the same principles as would regulate any other body dealing with Bank of England notes."^*27 In his opinion regulation of the operations of the issue department would suffice—or perhaps more accurately, would be likely to suffice—to assure sound management of the currency. In this respect Peel went further than his currency school supporters, and he later admitted that he had been overoptimistic.^*28 Torrens and Overstone had never committed themselves to the doctrine that regulation of the note issues was a remedy for all banking ills, although this was often charged against them, both by contemporary and by later critics of the currency school. They had recognized that careful management by the Bank of its discounts would be necessary if its banking department reserves were not to be exhausted through drawing down of deposits. In their discussion of the Palmer rule, they had pointed out that suitable management of its discounts was an essential element in the proper functioning of the Bank. But they had believed that the Act of 1844, by requiring segregation of part of the bullion reserve as cover for the notes, beyond achieving its primary objective of assuring convertibility of the note issue, would force the Bank to give close attention to the fluctuations in the unsegregated or marginal reserve held in the banking department, and therefore to act more promptly to check a threatening drain of gold.^*29 They now held that the difficulties of 1847 were due to mismanagement of the Bank, not to the Act of 1844, and that had it not been for the Act of 1844 the Bank would have carried its imprudence even further:

It was a case of banking mismanagement on the part of the Bank of England acting upon the community, at that moment peculiarly susceptible of alarm under vague and ignorant apprehensions of the effect of the new law...Danger from undue exhaustion of the bullion is the evil against which the Act undertakes to protect the community; against an improper exhaustion of the banking reserve, and the consequent inconveniences, it is the duty of the Bank of England to take timely and effectual measures of precaution.^*30

But if the currency school were prepared to admit that proper functioning of the banking system required proper management by the Bank of England of its credit operations as a whole as well as of its note issues, why did they content themselves with proposals for the regulation of the note issue only? The answer lay partly in the fact that their primary objective was guarantee of convertibility of the note issue, and this the Act of 1844 substantially accomplished. As Overstone claimed: the Act of 1844 "has preserved the convertibility of the bank note; the purpose for which it was passed, and that which alone its authors promised that it should do."^*31 The currency school tended also to minimize or to deny the importance of bank credit in other forms than notes as a factor influencing prices, or, as in the case of Torrens, to claim that the fluctuations in the deposits were governed closely by the fluctuations in the note issues.^*32 They had a hankering also for a simple, automatic rule, and could find none suitable for governing the general credit operations of the Bank.^*33 They also had laissez-faire objections to extending legislative control of the banking system any further than seemed absolutely necessary.

The currency school held that their critics exaggerated the significance of suspension of the Bank Act. Overstone had prior to passage of the act conceded that, in case of an internal panic, suspension of the act would be desirable. In such a case, resort must be had to "that power, which all governments must necessarily possess, of exercising special interference in cases of unforeseen emergency and great state necessity." But an explicit provision in the act authorizing its suspension in an emergency would be objectionable, for it would tend to convert into a routine and anticipated procedure what should be regarded as only an emergency measure.^*34 Later he argued that the suspensions which had occurred were of small consequence. During a panic an interval would elapse before a contraction of the note circulation would be offset by an inflow of gold from abroad: "To meet this temporary difficulty, which was purely technical and not depending upon any principle, an important provision of the Act was for a short time suspended."^*35

The banking school objected to the Bank Act of 1844, both that it was no remedy against overexpansion of bank credit and that overexpansion of convertible bank notes was impossible. But they never supported any proposals for legislative control of the volume of bank credit, partly because they thought it impracticable, partly because, like the currency school, they objected to such control on general laissez-faire grounds. In spite of the past record of the English banking system, which they interpreted as unfavorably as did the currency school, they apparently saw no alternative but reliance on the hope that the English bankers would in time learn to do better:

If the country banks have erred at all, it has not been in their conduct as banks of issue, but in their conduct as banks for discounts and loans; a matter altogether different and distinct, with which the legislature has no more to do than with rash speculations in corn or cotton, or improvident shipments to China or Australia.^*36

Were it possible, by any legislative proceeding, to restrain effectually the errors and extravagances of credit, that would be the true course to a really beneficial reform of our banking system. But these errors and extravagances are unfortunately rather beyond the pale of legislation, and can only be touched by it incidentally, or by a sort of interference which would be more vexatious and intolerable than even the evil which it sought to correct.^*37

The banking school were not willing to concede any merit whatsoever to the Act of 1844. They either denied that it would force the Bank to contract its issues more promptly in case of an external drain,^*38 or, if they granted this, they denied that this was an advantage.^*39 John Stuart Mill took an intermediate position. While in general hostile to the Act of 1844, he conceded that when an external drain took place, the act forced upon the Bank a prompter contraction of credit than it might carry out in the absence of the act. But he held that where the drain was due to a temporary factor and would soon cease of its own accord, such contraction was undesirable. The act, moreover, hindered the Bank from taking the steps which would give relief when a crisis had already occurred.^*40

V. THE POSSIBILITY OF OVERISSUE OF CONVERTIBLE BANK NOTES

The Bank Act of 1844, in setting a maximum limit for the note issues of the country banks and in providing for the eventual absorption of their circulation by the Bank of England, was carrying out the recommendations of the currency school. The bullionists, it will be remembered, had denied the possibility of a relative overissue of country bank notes if they were convertible upon demand into Bank of England notes or specie. But the boom of 1824-25 and the resultant crisis of 1826 opened the eyes of many to the expansion possibilities even under convertibility, and the currency school on this point did not adhere to the bullionist doctrine. They insisted that the country banks could expand their issues relatively to the Bank of England note circulation for a long enough period to create difficulties, without being adequately checked by the resultant adverse balance of payments with London. A fortiori, they held that the Bank of England and the country banks, acting together, could issue to excess even under convertibility.

Torrens on this question held views closer to those of the banking school than to those of his currency school associates. He claimed that when a relative overissue of country bank notes occurred, country notes would be presented to be exchanged for bills on London, which would in turn be exchanged for gold for export; the balance of payments both with London and with foreign countries would turn against the provinces, and the country banks would quickly find themselves compelled to contract their issues. Similarly, when the Bank of England directors "decreed a contraction of the currency, the provincial banks of issue, instead of resisting, obeyed and suffered."^*41

Norman replied that Torrens did not make sufficient allowance for "friction" when he claimed that the Bank of England had complete control over the country bank issues.^*42 Overstone argued that Torrens's conclusion rested on two assumptions, neither of which was valid: that the districts in which the two types of notes circulated were distinct and completely separated from each other,^*43 and that there was no delay before a contraction of Bank of England issues exercised its full effect on the reserves of the country banks. To Torrens's statement that when the Bank of England decrees contraction, the country banks of issue, instead of resisting, obey and suffer, he replied that "the country banks first resist, then suffer, and in the end submit.^*44

Torrens similarly claimed that nothing the Bank of England could do could increase the circulation by one pound beyond the amount decreed by the "necessary and natural law which governs the amount at which a convertible currency can be maintained." If the Bank issued more notes it would displace an equal amount of bullion thereby driven abroad.^*45 Torrens and the remainder of the currency school thus meant different things by "excess" of note issue. Torrens by "excess" of note issue must have meant an amount of issue which was greater than was consistent with the retention of bullion in reserves or in circulation as coin at its existing and presumably appropriate volume and would therefore result in an immediate export of bullion. The currency school as a whole meant by excess of note issue an amount of issue such as to make the total circulation of notes and coin combined greater than could be permanently maintained consistently with maintenance of convertibility and of the gold standard. The latter explained the phenomena resulting from an excess note issue in terms of lags between the original excess issue and the consequent rise in prices, external drain of gold, and impairment of the Bank's bullion reserves. Torrens would here have no commerce with lags, and he gave no consideration to the possibility of a significant intervening period of excess aggregate circulation. It is not apparent, however, that Torrens ever realized the extent of the divergence of his views from those of the other prominent currency principle advocates, or the essential harmony between this phase of his analysis and that of the banking school writers whom he was vigorously attacking.

Against the possibility of overissue the banking school appealed to the alleged "law of reflux", which amounted to nothing more than that the notes issued by a banking system on loan at interest to their customers would return to the banks in liquidation of these loans when they matured, and therefore any excess "would come back to the banks."

New gold coin and new conventional notes are introduced into the market by being made the medium of payments. Bank-notes, on the contrary, are never issued but on loan, and an equal amount of notes must be returned into the bank whenever the loan becomes due. Bank-notes never, therefore, can clog the market by their redundance, nor afford a motive to anyone to pay them away at a reduced value in order to get rid of them. The banker has only to take care that they are lent at sufficient security, and the reflux and the issue will, in the long run, always balance each other.^*46

To Fullarton's "vaunted principle of reflux," Torrens made an inadequate reply. If the Bank issued notes by discount of 60-day paper, there would be an interval of sixty days before an increase of notes would return to the Bank.^*47 But Fullarton had pointed out that there was no necessity "that the particular securities on which those notes were advanced should also furnish the channel for their return."^*48 As earlier loans matured during the 60-day interval, the Bank could contract its circulation by failing to replace them with new loans. What Fullarton certainly failed and Torrens apparently failed to see was that the "reflux" gave the Bank the power, but did not compel it, to contract its issues, and that by granting new loans as rapidly as old ones matured, the Bank could keep any quantity of notes out for any length of time, provided only that its bullion reserves were not exhausted, and that the Bank lent on terms attractive enough to find willing borrowers.^*49

The essential fallacy of the banking school doctrine had already been exposed during the bullionist controversy by Ricardo and others. It lay in its assumption that the "needs of business" for currency were a definite quantity independent of the state of business psychology and the activities of the banks. The banking school were right in insisting that the volume of bank credit was dependent on the willingness of businessmen to borrow, as well as on the willingness of banks to lend. But the willingness of businessmen to borrow depended on their anticipations of the trend of business, on the interest rate, and on their anticipations as to the willingness of the banks, in case of need, to renew loans upon their maturity. The banks, by lowering their interest rates, or consciously or unconsciously lowering their credit standards, could place more loans, and the increase of loans, by increasing prices and physical volume of sales, would in turn increase the willingness of businessmen to borrow. As Joplin had pointed out in 1826, bankers ordinarily do not see this, because they do not see that they themselves as a group had created the conditions which make an expansion of credit possible and appear to make it "necessary":

Bankers, indeed, have the idea that their issues are always called forth by the natural wants of the country, and that it is high prices that cause a demand for their notes, and not their issues which create high prices and vice versa. The principle is absurd, but it is the natural inference to be deduced from their local experience. They find themselves contracted in their issues, by laws which they do not understand, and are consequently led to attribute the artificial movements of the currency to the hidden operations of nature, which they term the wants of the country.^*50

The banking school also argued, as against the possibility of overissue, that if any bank issued in excess of its usual amount it would find the balances running against it at the clearinghouse and would be forced to contract its issue. That the power to over-issue of a single bank, operating in competition with other banks, was closely limited, had long been known. It had been pointed out as far back as 1773 that if a single bank increases its note issue it at first causes a drain on the reserves of the other banks in its district, but that in time its balances to other banks become unfavorable and it is forced to contract its discounts in order to replenish its reserves.^*51 Lord King made the same point in 1804: "An excessive issue of notes by any particular banker is soon detected, if not by the public, at least by the interested vigilance of his rivals; an alarm is excited; and he is immediately called upon to exchange a very large portion of his notes in circulation for that currency in which they are payable."^*52

In the 1820's, in reply to the use of this argument to demonstrate the impossibility of overissue, a number of writers drew a distinction between what a single bank acting alone could do and what a large group of banks, or an entire banking system, could do, acting simultaneously.^*53 The Committee of 1826 on Joint-Stock Banks heard much evidence to the effect that the practice of the Scotch banks of making a periodic demand on each other for payment of their respective notes in cash, bills on London, or exchequer bills, was a complete safeguard against excess issue. The questions put to some of the witnesses indicate that the doctrine that banks acting together could issue to excess, though not accepted either by the questioners or the witnesses, was already current.^*54

In the same year, a number of writers denied the claims that were being made on behalf of the Scotch banks, that their regular procedure of presenting each other's notes for payment provided a guarantee against overissue, on the ground that if the banks all increased their issues simultaneously and in the same degree, they would not have adverse clearing balances against each other and therefore could overissue indefinitely.^*55 These writers overlooked or, in the case of Doubleday, denied, that, while simultaneous and equal expansion by the Scotch banks would not result in adverse clearing balances among themselves, it would result, at least after a time, in adverse balances with London. It is to their credit, however, that they perceived and expounded the important principle that there is less check to overexpansion by banks when they act in unison than when they act singly, and that it is an error to infer, from the limitations upon expansion to which a single bank acting alone is subject, that overexpansion for a time is impossible for an important group of banks, or a fortiori for a banking system as a whole, when acting in unison. After 1826, this principle was frequently stated,^*56 and it was adopted by the currency school as one of the elements in their reply to the banking school doctrine that overissue was impossible under convertibility. It became an important element in the then prevailing theory of business fluctuations that alternating waves of optimism and pessimism resulted in overtrading and speculation followed by collapse and contraction, and that the bankers, who as a group shared the optimistic or pessimistic views of their customers, fed the cycle by simultaneous expansion or contraction of their credits.^*57

Several writers, however, went further, and insisted that even a single bank could overissue for a time, and that credit expansion initiated by a single bank might spread to other banks. McCulloch, in 1831, started from the hypothetical case of ten banks in London, each with a note issue of £1,000,000. If one of them should increase its issue to £2,000,000, there would result a fall in the exchanges and a demand for gold. But the demand on the overissuing bank would be only in the same proportion to its issue as on the other banks. If to check the drain of gold general contraction takes place, then, when the reserves had been replenished the bank which had expanded its issue would find itself with a circulation of £1,818,000, and the other banks would have a circulation of only £909,000 each. The other banks "would certainly be tempted to endeavor to repair the injury done them by acting in the same way." Even a single bank can expand, therefore, and, more important, may arouse the other banks to a defensive expansion.^*58 McCulloch failed to point out that a single bank which expanded its note issue while other banks remained passive or contracted would suffer a drastic impairment of its reserves. He now also insisted, for reasons which are not clear but which arose probably more from considerations of Scotch patriotism than of Scotch logic, that, while expansion by a single London bank was possible, this did not hold for Scotch banks.

Scrope denied that McCulloch's reasoning was sound either for London or for Scotch banks. He did not explicitly raise the issue of the effect on the reserves of the expanding bank, but he claimed that a bank could expand its issue relatively to other banks only by discounting at a lower rate, or on inferior security, than its competitors, and that to maintain its increased circulation it must continue to discount on more favorable terms. "But if, as is presumable, the other banks are going as far in both these ways as a sound practice will permit,...the bank in question cannot go beyond them without risks, such as no stable or solvent establishment would hazard."^*59

Sir William Clay, in the hearings before the 1838 Committee on Joint-Stock Banks, received an affirmative reply to the following question put by him to a witness:

Is there not this circumstance with regard to a competition in the issue of money, that although it may be true that one bank, of many (issuing in competition in Dublin, we will say), if it issued more in a larger proportion than its rival banks, would have its notes returned upon it; and is it not true that would not operate as a check, if all, in the spirit of competition in a period of excitement, were also disposed to issue largely?

Longfield, citing this question and answer, objected that they took insufficient account of the part which even a single bank could play in bringing about an expansion of the circulation. If a single bank in a particular region expanded its discounts and permitted its cash reserve ratio to fall, there would result a gold drain from the banks of the region as a whole either to hand-to-hand circulation or for export, which all the banks in that region would feel in proportion to their circulation. If the other banks kept their discounts constant, they would find their reserves falling in greater proportion than their circulation (because since their circulation was several times larger than their reserves, the loss of a given amount of cash through presentation of notes for payment would represent a greater relative reduction in their reserves than in their circulation). To maintain their former reserve ratio, they must drastically contract their discounts. The expanding bank, if it had sufficient capital to withstand the drain on its own reserves, could by this procedure drive the other banks out of business. If the other banks in self-defense expanded their discounts, and allowed their reserve ratios to fall, there would result a general expansion of credit and circulation in the district. "Thus a bank may be driven in self-defense to take up the system of overtrading adopted by its competitors, and where there are several joint-stock banks of issue, the country will suffer under alternations of high and low prices, of confidence and panic, of great excitement and general depression of trade." Competitive issue of bank notes might therefore operate as a stimulus to, instead of as a protection against, the periodic recurrence of general overexpansion and overcontraction of the circulation.^*60

VI. THE ROLE OF DEPOSITS, BILLS OF EXCHANGE, AND "CREDIT" IN THE CURRENCY SYSTEM

It was, as we have seen, the position of the banking school that bank notes and bank deposits were both means of payment and parts of the circulating media, and that, since the proposals of the currency school dealt only with bank notes and left bank deposits free of control, they were bound to operate unsatisfactorily if put into practice. In a memorandum to Tooke in 1829, Pennington insisted that the deposits of London bankers performed exactly the same function as did the notes of country bankers: "the book credits of a London banker, and the notes of a country banker, are but two different forms of the same species of credit."^*61 This statement by Pennington is often credited with being the first statement of the identity between the economic functions of notes and deposits. It undoubtedly exerted a considerable influence not only on the members of the banking school, but also on the currency school, and especially on Torrens.^*62 But Pennington was merely repeating an old doctrine. At the very beginning of paper money in England, it was recognized that the transfer of bank notes and the transfer of book credits at the bank were alternative means of making payments.^*63 During the restriction on cash payments at the beginning of the nineteenth century the part played by the expansion of bank deposits in bringing about rising prices and the premium on bullion never became a subject of controversy, but a number of writers, both bullionist and anti-bullionist, in their analysis of the monetary process, assigned to bank deposits a role identical with that of bank notes. Boyd, in 1801, held that the "open accounts" of London bankers were, equally with country bank notes, an "addition to the powers of the circulating medium of the country." Bank notes were the "active circulation" of the banks; book credits were the "passive circulation" because they circulated only as their owners issued orders upon the bank.^*64 Thornton in 1802 treated bank deposits as a substitute for paper money.^*65 James Mill, in 1807, accepted "the common cheque upon a banker" as in the same class with the bank note, both being "currency," but neither being "real money."^*66 Lord Stanhope in 1811 presented a resolution in the House of Lords to authorize the Bank of England to establish branches throughout England, and to substitute for bank notes book credits with the Bank, to be legal tender and transferable without cost. Stanhope claimed for this proposal that it would avoid the disadvantage of paper money of its liability to forgery, and the disadvantage of metallic currency, of the influence on its quantity of the international balance of payments.^*67 He clearly regarded bank deposits as identical with bank notes in their monetary significance. Torrens, who was later radically to change his monetary views, in 1812 claimed that checks and bills of exchange were more important elements in the circulation than bank notes.^*68 Samuel Turner pointed out that "A country bank was a kind of clearing-house, where, without any actual interchange of notes or money, the greater part of all payments between man and man was effectuated by mere transfers in the books of their bankers."^*69 Senior stated that deposits subject to check were more important banking instruments for making payments than were bank notes.^*70 Other writers, while denying to bank deposits the dignity of constituting an independent element in the circulating medium, conceded that they were "economizing devices," which rendered a smaller amount of bank notes sufficient to mediate a given volume of monetary transactions.^*71

There might be debate among economists today as to whether bank deposits are "money" or are "currency." There would be general agreement, however, that they are, like bank notes, means of payment and therefore a part of the circulating medium. Many early writers, however, insisted that bills of exchange were also part of the circulating medium. Henry Thornton, in 1797, included as "means of payment," not only coin and bank notes, but also bills of exchange "when used as such," i.e., when they served as a means of final settlement of a transaction.^*72 An anonymous author wrote in 1802 that "Cash, or ready money, when considered as the medium of payment in a commercial country, comprehends every species of negotiable paper...."^*73 Ravenstone stressed the importance of bills of exchange as a means of payment, and declared that "I do not know how this species of paper has entirely escaped the attention of those who have treated this subject."^*74 Burgess,^*75 Parnell,^*76 and subsequently many other writers, included bills of exchange as parts of the circulating medium.

Some writers included "credit" as a part of the circulating medium, but meant by "credit" bank credit, and regarded it not as an item additional to bank notes and bank deposits, but as the source from which the two latter items arose.^*77 But other writers, most notably J. S. Mill, included credit in a broader sense as an element of "purchasing power":

The purchasing power of an individual at any moment is not measured by the money actually in his pocket, whether we mean by money the metals, or include bank notes. It consists, first, of the money in his possession; secondly, of the money at his banker's, and all other money due to him and payable on demand; thirdly, of whatever credit he happens to possess. To the full measure of this threefold amount he has the power of purchase. How much he will employ of this power, depends upon his necessities, or, in the present case, upon his expectations of profit. Whatever portion of it he does employ, constitutes his demand for commodities, and determines the extent to which he will act upon price.... Bank notes are to credit precisely what coin is to bullion; the same thing, merely rendered portable and minutely divisible. We cannot perceive that they add anything, either to the aggregate of purchasing power, or to the portion of that power in actual exercise.^*78

Modern writers on money as a rule include specie, government or bank notes, and bank deposits payable on demand by check, as constituting "money," or the "circulating medium," or the stock of "means of payment." They exclude bills of exchange and promissory notes, and treat checks as merely the instruments whereby bank deposits are transferred or "circulate." But during the early part of the nineteenth century bills of exchange for small amounts were still commonly used in some parts of England, and especially in Lancashire, as a means of payment between individuals, and sometimes passed through many hands in settlement of transactions before they matured and were canceled. To the extent that the receivers of these bills passed them on to others before their maturity in payment of debts or as payment for purchases, they functioned just as did bank notes and were properly to be included in the circulating medium. As one contemporary writer, Edwin Hill, pointed out, anticipating Francis Walker's dictum that "Money is that money does," the correct test of whether something is "currency" or not is not what it is, but what it does; bills of exchange, to the extent that they settled transactions without involving the use of any other medium, acted as currency.^*79 But even when bills of exchange do not pass from hand to hand, they are still entitled to be ranked with checks as instrumentalities whereby bank deposits are transferred, provided, as was generally the case in England, these bills were made payable at the acceptor's bank and when they matured were passed through clearings and credited to and debited against bank accounts in the same manner as checks.

In including personal command over credit and individual claims on other individuals as part of the stock of "purchasing power," J. S. Mill went too far. If valid individual claims to immediate payment are included as means of payment, then individual liabilities to immediate payment should be subtracted therefrom. Since these items are necessarily equal, they cancel each other, although they may in practice affect in different degree the willingness of the creditors and the unwillingness of the debtors to use their cash balances in other transactions. The case of command of credit in making purchases presents more difficulty. If all who can purchase on credit were simultaneously to do so, prices would rise even if demand deposits and notes in circulation remained unchanged in volume, for it is purchases which raise price levels, rather than payments for purchases. But the maintenance of the higher level of purchases requires, after some interval, an augmentation of the volume of payments, and this in turn requires either more means of payment of their greater "velocity" or rapidity of use. But the whole discussion as to what is and what is not "money" retains the appearance of significance only while velocity considerations are kept in the background. What mattered for the currency school-banking school controversy was the extent and the causes of the fluctuations in the volume of payments, i.e., of amount of money times its payment velocity, and therefore the wrongful inclusion as money of something which did not serve as a means of payment was of little consequence if its velocity coefficient was recognized to be zero. Moreover instruments which were not money at some particular moment could be so at some other moment. In this connection bills of exchange, time deposits, and overdraft privileges could be regarded as a sort of "potential money." The quality which one writer attributed solely to bills of exchange, as the result of which "they can be either kept in the circulation, as media of payment, or withdrawn from the circulation, and held for a time as interest-bearing investments,"^*80 was possessed also by time deposits, which could without much delay be transformed into demand deposits.

That differences in their velocity of circulation were the significant basis on which bank notes, deposits, and bills of exchange could be distinguished from each other with respect to their possession of the qualities of "currency" was by no means overlooked during this period. The existence of such differences was, indeed, the main ground on which the currency school refused to include deposits and bills of exchange, with their comparatively low velocity of circulation, on a parity with bank notes as parts of the circulating medium. This could be conceded to the currency school, however, without accepting their conclusion that deposits and bills of exchange should be treated as not constituting any part of the circulating medium, and should have led only to the assignment of greater weight to a given quantity of bank notes than to the same quantity of deposits. This was in effect done by several writers. Gilbart, for instance, although insisting that deposits were means of payment just like bank notes, argued that the extent to which they perform the functions of money must be measured, "not by the amount of the deposits, but by the amount of the transfers." Because only deposits payable on demand could be transferred, he considered only such deposits as a part of the currency.^*81 Longfield, similarly, held that the greater velocity of circulation of bank notes was a significant, but the only significant, difference between deposits and notes, in so far as their influence on prices was concerned.^*82 J. W. Lubbock expounded the same doctrine for cash, checks, and bills, with the aid of an algebraic formula which has close resemblance to Irving Fisher's celebrated "equation of exchange."^*83

Although the members of the currency school all supported a system of currency regulation which would place the issue of bank notes under rigorous control but would leave deposits wholly free from interference, they did not agree on the grounds which justified this discriminatory treatment of deposits and notes.^*84

Torrens freely conceded that bank deposits and bank notes were coordinate means of payment and acted similarly on prices. He claimed, however, that payments in specie and notes bore a constant proportion to payments by check and that an expansion of deposits could therefore not take place without an increase of gold or notes,^*85 regulation of the volume of note issues thus automatically involved regulation of the volume of bank deposits.^*86 Sir William Clay conceded the similarity between bank deposits payable upon demand and bank notes,^*87 and even admitted that the latter were a subsidiary circulating medium, and one whose importance was bound to decrease.^*88 He nevertheless insisted that the issue of bank notes needed to be and could be closely controlled. As for deposits, however, he knew of no practicable means of controlling their volume, and in any case there was no desire for such control in any quarter.^*89

Norman's main argument against the inclusion of deposits as a part of the currency was that the velocity of circulation of deposits was much less than that of bank notes or coin.^*90 He also claimed that the volume of deposits and bills of exchange was dependent on the volume of underlying credit, which in turn was regulated by the amount of bank notes and coin, and that in any case the influence on prices of these "economizing expedients" was only "trifling and transient."^*91 It was no more reasonable, moreover, to object to proposals for regulating note issues because the "economizing expedients" were left unregulated than it would be, in the absence of paper money, on similar grounds to object to the regulation of coin.^*92 Norman argued also that all that the currency school proposed was that the currency should be made to operate as if it were a purely metallic currency; but even under such a currency, i.e., even if bank notes did not exist, "the trade in money, like other trades, would be occasionally out of joint, although not probably so often or to so great an extent as now."^*93

Overstone's case for limiting regulation to note issues also consisted mainly of this argument that if a currency system which included bank notes could be made to operate as would a currency system in which bank notes did not exist, that was all that could be expected of it:

The utmost that can be expected from a paper-currency is that it shall be the medium of adjusting the various transactions of a country without greater inconvenience to the community than would arise under a metallic circulation.^*94

Deposits, debts owing, indeed credit in any form, may be made the means of purchasing and paying, of adjusting transactions; and they may therefore, in one sense, be considered as forming a part of what has been called "auxiliary currency." But the whole superstructure of "auxiliary currency" forms a subject, distinct from that of the management of the circulation. It may be raised equally upon a metallic or a paper circulation, and the fluctuations which it may undergo are subject to laws distinct from those which ought to regulate the substitution of paper for metallic money.^*95

The final outcome of the discussion was that the currency school agreed with the banking school that deposits and other forms of "auxiliary currency" or "economizing expedients," as well as bank notes, could be a source of difficulty, but that the two groups appraised differently the relative importance of variations in the two types of means of payment as causes of currency and credit disturbances. The currency school were not prepared to support government regulation of the credit operations of the banking system, but believed that statutory limitation of the note issues would bring a substantial measure of improvement. The banking school refused to support statutory restrictions on either bank deposits or bank notes, and maintained that the strict limitation of the amount of uncovered note issue would either have no effect or would operate to accentuate rather than to moderate the fluctuations in business conditions.^*96

It is not sufficient, to refute the currency school argument, to show that note circulation and bank deposits have divergent fluctuations, or even that there are divergent fluctuations in the volumes of payments by means of bank notes and checks, respectively, if, as the currency school assumed to be the case, the relative use of notes and checks is at any one moment, given the circumstances and habits then prevailing, fairly definitely fixed, and if the regulation of the quantity of notes does not of itself operate to induce a change in these relations. But given their price and business-stabilizing objectives, the currency school should have proposed such a method of regulation of note issue as would have resulted at all times, if velocity is left out of account, in the desired aggregate volume of means of payment, or taking velocity into account, in the desired aggregate volume of payments. Since the different stages of a business cycle are marked by variations in the proportions between bank notes and deposits, mere limitation of the amount of uncovered note issue would not suffice, and no method of regulation of note issue would suffice which did not make provision for cyclical changes in the ratio between bank notes and deposits and in their relative velocities, as well as for any changes in these ratios which the regulation of one type of means of payment might itself tend to bring about. Since these provisions could not be reduced to a simple formula, regulation of note issue alone, though it might still operate on the whole to make a "mixed currency" conform more closely to a "purely metallic currency" than if left wholly unregulated, would fail to bring about the desired results with respect to prices and the volume of business activity. An additional difficulty, with respect to the timing of regulatory measures, would arise, if, as appears to have been the case, the fluctuations in deposits preceded, instead of being simultaneous with or following, the fluctuations in note issues, so that if attention was confined to note issue alone the danger signals would come too late.^*97

During this period, however, the relative importance of bank notes in the English circulating medium, while steadily decreasing, was much greater than it is today. It is possible, moreover, to defend the currency school against these criticisms, even if deposits are acknowledged to be coordinate with notes as means of payment, if their objective of limiting the fluctuations in the volume of means of payment to such as would exist under a "purely metallic currency" is accepted as adequate, and if it is conceded that the variations in the proportions of deposits to specie and notes under a "mixed currency" would correspond, caeteris paribus, to the variations in the proportions of deposits to specie under a "purely metallic currency."

VII. THE TECHNIQUE OF CREDIT CONTROL

The Record of the Bank of England.

In appraising the record of the Bank of England during this period, allowance must be made for the lack of an adequate statistical account of the operations of the Bank and also for the absence of any serious attempt on the part of the Bank publicly to defend its record. It nevertheless appears to me that the evidence available warrants the verdict that during the period from about 1800 to about 1860 the Bank of England almost continuously displayed an inexcusable degree of incompetence or unwillingness to fulfill the requirements which could reasonably be demanded of a central bank. During the restriction of cash payments, it not only permitted the paper pound to depreciate, prices to rise, and the exchanges to fluctuate, but it repeatedly denied that there was any relationship between these phenomena and its own activities. William Ward relates that when he became a director of the Bank in 1817, he "could trace nothing directly that could fairly be said to constitute a plan or system" of credit management. It was not until 1827 that the Bank, upon Ward's motion, rescinded a resolution which it had solemnly adopted in 1819, which appeared to deny any connection between the volume of its note issues and the level of the foreign exchanges.^*98 The Bank even after 1827 apparently continued to be without any reasoned policy as to its discount rate, for Ward, in 1840, could still write:

I have often pressed on the Court the necessity of regarding the market rate of interest, but I generally found it an unwelcome subject. Low interest was said to encourage speculation; and on my enquiring the principle by which the rate should be governed, I was told in answer, to look to the practice of our forefathers.^*99

The growing authority of Horsley Palmer and G. W. Norman in the counsels of the Bank in the 1830's brought more enlightened pronouncements to the public, but does not appear to have improved the practice of the Bank. The adoption of the Palmer rule was a flagrant error, and the rule was repeatedly violated in such manner as to make things worse instead of better. The passage of the Act of 1844 by huge majorities was evidence of a general lack of confidence in the ability of the Bank properly to carry out its responsibilities to the public. When the Act of 1844 came into effect, the Bank at once proceeded to act as if the freedom from external control which the act left to the banking department had also rendered unnecessary any internal control. During this entire period, the management of the Bank showed an almost incomplete inability to profit not only from its own recent experience, and from the advice so freely offered to it by outsiders, much of it excellent, but even from "the practice of their forefathers" in the eighteenth century.^*100 The Bank then knew that there was a connection between their discount policy and their note issue, on the one hand, and the level of the foreign exchanges, on the other, and that a contraction of their discounts would operate to improve the exchanges and to check an external drain of gold. It recognized the difference between an internal drain due to impairment of confidence and an external drain due to a relative excess of note issue, and it was aware that different remedies were appropriate for the two cases: courageous extension of credit in the former and contraction of credit in the latter. Until the Bank was exempted in 1833 from some of the provisions of the Usury Acts, variations in the discount rate above 5 per cent were not available to it as an instrument of credit control, but it made use of informal rationing,^*101 of systematic borrowing from the market, and probably also of open-market operations, during the eighteenth century.

Variations in Discount Rate vs. Rationing.

From the beginning of the nineteenth century, writers had expressed regret that the Usury Laws prevented the Bank from substituting variations in the discount rate for rationing as an instrument of credit control.^*102 When in 1833 the Bank was exempted from the Usury Laws, in so far as its loans of three months' maturity or under were concerned, it was done with the approval of the Bank^*103 but apparently not at its request, and the Bank did not make systematic use of this new instrument until after 1844, and even then with a view too much to its own profit and not enough to its responsibilities as a central bank.^*104 The discovery by the Bank that the discount rate was an effective instrument of control seems to have surprised it,^*105 although dozens of writers had been scolding it for years for its failure to use it more extensively.

The arguments then used for the substitution of variations in the discount rate for rationing were that rationing was arbitrary and capricious in its mode of behavior, and that the fear to which it gave rise, that credit facilities in sufficient quantities would be unavailable on any terms in case of credit stringency, tended to promote panic. Formal rationing seems to have been practiced only at times of unusual credit strain. But even after the Bank had adopted variations in the discount rate as its chief instrument of control, it still upon occasion made use of rationing, in the form of shortening of the maturities of the paper which it would accept for discount, as a supplementary instrument of control.^*106

Open-Market Operations.

The present-day literature on banking commonly treats open-market operations, or the purchase and sale of securities by the Bank on its own initiative as a means of currency and credit control, as a recent development whether as idea or as practice.^*107 Given its legally and traditionally fixed rate of discount, there was for the Bank of England even during the eighteenth century no alternative to rationing of discounts except open-market selling operations and borrowing from the market when it wished for any reason to increase its bullion reserves. Since rationing meant refusal to its regular commercial customers of discount of what had hitherto been fully acceptable commercial paper, it was a drastic step which it could never readily have taken in the absence of emergency conditions. We know that during the Restriction period the Bank bought exchequer bills in the open market whenever it thought the circulation inadequate for the needs of the country but found no demand for additional discounts at the traditional rate.^*108 Ricardo regarded the volume of commercial discount business of the Bank as too small to serve as an adequate regulator of the volume of the currency, and he held that the conservative discount policy of the Bank made it necessary that it be in a position to use other means than increase in its commercial discounts to increase the amount of the currency, if this was to be maintained under a metallic standard at a sufficiently high level.^*109 He took it for granted that under a metallic standard open-market operations would be relied upon by the central bank whenever it desired to reinforce or to offset the effects of automatic gold movements.^*110 After the resumption of cash payments, open-market operations were without question the main instrument of credit control used by the Bank.^*111 It was reluctant to resort to the drastic step of formal rationing of discounts when it wished to contract its credit operations, and it had no means other that open-market operations of increasing the volume of outstanding bank credit when the demand for discounts at the traditional and legally the maximum permissible 5 per cent rate was insufficient either for credit control objectives or for its own income objectives. Under the Palmer rule, which called for the maintenance of the securities held by the Bank at a constant amount once the desired balance had been attained, the only scope for open-market operations would be to offset variations in the amount of commercial paper held by the Bank by counter-variations in the holdings of governments. The critics of the Palmer rule, when they insisted that "forcible operations upon the securities" would sometimes be necessary to check a drain of gold, or to make the note circulation expand to the same amount as did the gold reserves in case of an influx of gold, meant by such operations sales and purchases of government securities in the open market as well as contraction or expansion of commercial discounts. Norman, in fact testified in 1832 that, given the restrictions of the Usury Laws, open-market operations were the only practicable means of regulating the note issue,^*112 and Palmer, before the same committee, stated that if contraction was necessary the Bank would sell exchequer bills first, and would contract its discounts only as a last resort and only if the market rate of discount exceeded the legal maximum.^*113 It was common knowledge at the time that open-market operations were the main reliance of the Bank when it wished to act on the volume of its credits outstanding.^*114 Probably more important was the fact that prior to the passage of the Act of 1844 the Bank of England never lowered its commercial discount rate below 4 per cent, and after resumption of cash payments the market rate, except at crisis periods, was as a rule substantially lower than the Bank rate.^*115 The Bank, in consequence, lost most of its commercial discount business, and except at times of financial pressure what remained consisted largely of special accounts paying less than the nominal rate.^*116

There was little discussion, however, of the technique of open-market operations. One writer claimed that when the Bank wished to contract its note issue, the order of its operations was, first, to raise the discount rate, second, to sell government securities, and finally, if these did not suffice, resort to "putting on the screw" or rationing.^*117 It was pointed out that from the point of view of the Bank open-market operations suffered from the disadvantage that ordinarily it would be when securities were low in price that sales would be in order and when they were high that purchases would be made.^*118 During the crisis of 1847, the Bank, to escape the capital loss which would be involved in selling government stock, and to avoid arousing as much attention in the money market as would be involved in openly borrowing from the market, sold government securities for cash and at the same time bought an equal amount forward, thus in effect borrowing from the market.^*119 In 1875, and later, it appears to have resorted to analogous practices.

Internal and External Gold Drains.

The banking school regarded it as one of the defects of the Bank Act of 1844 that it failed to provide different treatment for an internal and an external drain on the Bank's gold, but in both cases aimed at forcing a corresponding contraction in the note circulation. They maintained that an internal drain due to mistrust called for an expansion instead of a contraction of credit.^*120 Palmer, in 1840, made a further distinction between external drains due to temporary causes which could be allowed to correct themselves and external drains due to a fundamental disequilibrium of price levels which could be corrected only by forcing down prices through contraction of credit:

I think the Bank are always called upon to look for the cause of the drain as far as they can form an opinion upon it when it commences, and to act upon the best opinion they can form of the occurrences then passing. There are two causes that will act upon the bullion of the country; one I take to be the deranged state of prices between this and other countries; the other, distinct payments which are to be made to foreign countries without any derangement of the general prices; if of the latter character,...that payment being made, and the commerce of the country not being deranged, I believe the bullion and currency would gradually resume their former state. If, as in the year 1825, a great derangement of prices existed, then it would only be by an adjustment of those prices, with reference to foreign countries, that the drain of bullion would be stopped.^*121

J. S. Mill drew similar distinctions between internal drains, external drains which were self-corrective in character, and external drains which could be checked only by a change in relative price levels, and criticized the Act of 1844 on the ground that it forced the Bank to apply identical treatment to all three types of drains.^*122 He claimed that a mechanical rule for the regulation of note issue was objectionable because it would prevent different treatment of the different types of drains,^*123 and he held that there would ordinarily be no difficulty for the Bank in determining the character of a drain, as the cause of a drain was generally notorious.^*124

The distinction between external drains according to their causes is valid and important, but Mill exaggerated the ease with which they could be so distinguished in practice, especially in a period of scanty statistical data.^*125 A drain, moreover, which is distinctly of one type in its origin, may imperceptibly become a drain of another type, or may, by causing alarm, give rise to another type of drain as well.^*126

Mill was in error also when, following Tooke, he held that while prior to 1844, and also under a purely metallic currency, a drain would generally be met from the "hoards" of bankers and merchants, under the Act of 1844 it must necessarily come out of the circulation.^*127 A contemporary writer pointed out the ambiguous way in which Mill here used the term "circulation." Mill's account of the manner in which the Act of 1844 must operate was correct only if by "circulation" he meant the "gross circulation" outside the issue department. But this gross circulation included the notes and bullion held by the banking department as well as whatever reserves of notes or bullion were held outside the Bank of England. These reserves outside the issue department, however, constituted "the identical hoards from which, as he so truly argues, when speaking of a [purely] metallic currency, nearly all drains must be taken." Mill's criticism of the Act of 1844 would be valid therefore only if under it drains must come from the "active" or net circulation, which was not the case.^*128 The Bank, in other words, was still able, under the Act of 1844, to discriminate in its treatment between different kinds of drains, and to meet drains out of its reserves without contracting its "active" circulation when it thought it desirable, if it kept adequate reserves in its banking department. Mill, however, later admitted in effect that at least as far as external drains was concerned the Bank of England could still deal with them as it had had the power to do before 1844, if it retained in its banking department as large reserves as before 1844 had sufficed for the Bank as a whole.^*129

Adequate Reserves.

Gold reserves yield no income, and banks operating for profit tend to reduce them to the lowest level that seems consistent with safety. In countries with central banks, all other banks tend to rely upon the central bank to provide the bulk of the gold reserves for the system as a whole. The Bank of England was never legally charged with this responsibility, and its obligations to its shareholders, who during this period still held its stock primarily because they wanted dividends and not as a social duty or because prestige attached to being a "Bank proprietor," necessarily loomed large in the minds of its directors. The other banks, on the other hand, behaved as if the Bank of England were a true central bank, with full responsibilities for looking after the gold reserves of the nation. When the Bank's charter was renewed in 1833, the government made a stiff financial bargain with the Bank, which reduced its earning power and made still more onerous for it the maintenance of any surplus reserves. The rapid growth of joint-stock banks in London further deprived the Bank of England of a large part of its commercial discount business, which had hitherto been the most remunerative form of employment of its funds. As a result of these circumstances, the English credit structure was marked, during the nineteenth century, by an extraordinarily low ratio of gold reserves to aggregate gross demand liabilities of the banking system. English banking statistics for this period are too meager to make possible an accurate determination of this ratio, but it seems that, disregarding the probably negligible amounts of coin and bullion held as reserves by the joint-stock and private banks, it fell at times to as low as 2 per cent and never between 1850 and 1890 exceeded 4 per cent.

From the late 1820's^*130 on to the end of the century a continuous succession of writers called attention to the inadequacy of the gold reserves, but without any visible results. One writer pointed out that the interest of the public in an adequate gold reserve was so great as to render the cost of its maintenance a matter of very minor importance from the national point of view. He took it for granted that the terms of the 1833 Bank Charter made the maintenance of an adequate reserve a greater burden for the Bank than it could bear. He therefore recommended that the Bank should be required to establish, at the expense of the government, an additional reserve, not to be encroached upon without a warrant from the Treasury.^*131 Richard Page saw that the ambiguous status of the Bank was a source of danger: "The double interests and duties of the Bank—as the proper institution for regulating the currency, and conducting a profitable banking business—are incompatible. The two things may often consist, but times will occur when they cannot be preserved together."^*132 He warned that the economy of the use of the precious metals had already been pushed too far, and that means should be found to restore the reserves to a satisfactory level:

A banker is now encouraged to keep but a small amount of specie by him; all his transactions resolve themselves into and are based upon ready money, and yet he is relieved of all labor and anxiety in procuring specie. The charge and responsibility of that obligation are taken from his shoulders, and put upon the Bank of England. The customers for gold in the market are therefore reduced to a single body; who, if the gold comes, take it in, but confess that they do not conceive it a part of their duty to go out of their way to obtain it. This is an evil. If every banker was obliged to market for himself, we should soon find our condition amended.^*133

Every recent improvement in banking has gone upon the principle that we should retain gold as a standard, but bring it forward as seldom as possible, and scarcely ever touch it. The perfection of the theory would be a refinement of the thing into nothing, a spiritualizing away of the reality, until gold and no gold became one and the same. Such improvers would make it "small by degrees, and beautifully less," until it had vanished altogether, and ceased to exist otherwise than argumentatively.^*134

The Bank Act of 1844 made the gold in the issue department of the Bank unavailable for external payments, except as the banking department had a disposable reserve of notes which it could exchange for issue department gold.