(3.8)
7 The regulation of the quantity of convertible notes by the standard
The only use of a standard is to regulate the quantity, and by the quantity the value of a currency.
(Proposals; IV: 59)
According to Ricardo in his letter to Francis Horner of 5 February 1810, the third cause of depreciation of money was “a superabundance of paper circulation” (VI: 2). By “superabundance” Ricardo meant “that quantity of paper money which could not by any means be kept in circulation if it were immediately exchangeable for specie on demand”, as was expected in a discussion of the consequences of inconvertibility. This implied that under convertibility paper money could not be permanently in excess. The reason for that would be a matter of debates, not only during the Bullion Controversy but also long after the resumption of convertibility, including during the controversy in the 1840s between the Currency School and the Banking School. Some authors contended that an excess issue of notes was absorbed in circulation, but this absorption triggered a corrective process that eliminated the excess. Others contended that, if notes were issued through the discounting of “real bills”, an excess could not even occur.
Ricardo maintained that, however cautious the Bank of England could be in discounting bills for bona fide transactions, this did not prevent the aggregate quantity of money from being in excess, under inconvertibility and convertibility as well. Contrary to what happened under inconvertibility, however, depreciation was only temporary when notes were convertible into the standard (gold bullion), even if convertibility was indirect – notes being converted into coin with a view to melting the coin into bullion. During the transition, the economy suffered inconveniences and Ricardo proposed the Ingot Plan as a way to get rid of them.
The present chapter analyses how, according to Ricardo, the value of money adjusts to a change in the quantity issued of convertible notes, first by departing from the value of the standard (depreciation or appreciation), then by being brought back to it. This requires introducing the quantity of money into the Money–Standard Equation formulated in Chapter 4 above (Section 7.1). Two cases of self-adjusting process are then studied, in response to an increase in the “wants of commerce” (Section 7.2) or in the discretionary note issue (Section 7.3). An important consequence ensues: contrary to what is generally found in the literature, Ricardo’s monetary theory was not a quantity theory of money (Section 7.4). I will then ask whether this theory applies to money without a standard (Section 7.5), before concluding on the specificity of the market for gold bullion (Section 7.6).
With the exception of Section 7.5, this chapter considers a mixed monetary system of undebased coins and Bank of England notes convertible into coin – like that ruling in England before 1797 and from 1821 onwards – leaving to Chapter 9 below the analysis of Ricardo’s proposals for a circulation exclusively composed of notes convertible into bullion (the Ingot Plan) and for a mixed monetary system of undebased coins and notes convertible into coin but issued by a public bank (the Plan for a national bank).
7.1 The Money–Standard Equation and the quantity of money
As emphasised in Chapter 3 above, the condition of conformity of money to the standard is expressed by the equality between the value of money and the value of the standard (both in terms of all commodities except the standard), given the legal price of the standard:
(3.8) |
with 
the value of money conforming to the standard, 
the value of gold bullion (the standard of money), and 
the legal price of gold in coin.
Let me call 
the conformable quantity of money (measured in pounds) – that is, the quantity consistent with the condition of conformity (3.8) – and M the actual quantity of money, each unit of 
circulating at a value 
to which the condition (3.5) of coherence of the price system applies:
(3.5) |
with 
the market price of gold bullion. Since 
circulates the same aggregate value of commodities as 
the following equality applies:
(7.1) |
Replacing 
and 
in (7.1) by their respective values in (3.8) and (3.5) gives:
(7.2) |
When the actual quantity of money is in excess (deficient) relative to the quantity that fulfils the condition of conformity of money to the standard, money is depreciated (appreciated) in proportion to the excess (deficiency), as testified by the market price of gold bullion being above (below) the legal price of gold in coin. It should be observed that equation (7.2) is a direct consequence of: (a) the normative condition (3.8) of conformity of money to the standard; (b) the normative condition (3.5) of coherence of the price system; and (c) Ricardo’s contention (7.1) that whatever the quantity of money issued it was absorbed in circulation (“a circulation can never be so abundant as to overflow”; Principles; I: 352). Equation (7.2) does not result from a description of the working of the market for the standard (gold bullion) in the case of convertible notes (more on this point in the Afterword of the present book).
With 
remaining unchanged, one derives from (7.2) the following relation:
(7.3) |
with 
the rate of change of the market price of gold bullion, 
the rates of change of respectively the actual and the conformable quantities of money.
Let me recall the Money–Standard Equation in simplified form analysed in Chapter 4:
(4.8) |
From (7.3) and (4.8) one obtains:
(7.4) |
Equation (7.4) reformulates the Money–Standard Equation to take into account how a change in the aggregate quantity of money affects the market price of gold bullion. The two causes of change in the value of money (a change in the value of the standard and a change in its market price), distinguished in Chapter 4 above, now become a change in the value of the standard (with a positive sign) and the difference between the rates of change in the actual and the conformable quantities of money (with a negative sign). In Chapter 5, I analysed how an exogenous change in the value of gold bullion 
– such as the discovery of a new mine in a foreign gold-producing country or an unusual demand from the bank of issue in the domestic gold-importing country – caused a change in the domestic value of money. As for the second cause of change in the value of money, namely the difference between the rates of change in the actual and the conformable quantities of money, Ricardo exposed it in Proposals:
The value of money then does not wholly depend upon its absolute quantity, but on its quantity relatively to the payments which it has to accomplish; and the same effects would follow from either of two causes – from increasing the uses for money one tenth – or from diminishing its quantity one tenth; for, in either case, its value would rise one tenth.
(Proposals; IV: 56)
In this example, 
because either 
I will now examine how, according to Ricardo, the value of money adjusts to a change either in the conformable quantity of money 
following a change in the “wants of commerce”, or in its actual quantity M, consequent upon a discretionary change in the note issue. In what follows, I will consequently assume that no exogenous variation in the value of gold bullion occurs.
7.2 The adjustment of the value of money to an increase in the “wants of commerce”
The notion of wants of commerce
The expression “wants of commerce” was used in the Appendix to the fourth edition of High Price where Ricardo outlined his Ingot Plan:
On just principles [of Ricardo’s plan] we should possess the most economical and the most invariable currency in the world. […] The amount of the circulation would be adjusted to the wants of commerce with the greatest precision.
(High Price; III: 127)
Developing this plan in Proposals, Ricardo again argued that the quantity of notes could be endogenously adjusted to the needs of the circulation in a more secure and cheaper way than with coins:
Amongst the advantages of a paper over a metallic circulation, may be reckoned, as not the least, the facility with which it may be altered in quantity, as the wants of commerce and temporary circumstances may require: enabling the desirable object of keeping money at an uniform value to be, as far as it is otherwise practicable, securely and cheaply attained.
(Proposals; IV: 55)
The same expression had been introduced by Ricardo in Reply to Bosanquet when he criticised the view (later called the Real Bills doctrine in the literature) according to which notes could not be issued in excess as long as they were put into circulation by discounting bills generated in actual transactions:
The refusal to discount any bills but those for bona fide transactions would be as little effectual in limiting the circulation; because, though the directors should have the means of distinguishing such bills, which can by no means be allowed, a greater portion of paper currency might be called into circulation, not than the wants of commerce could employ, but greater than what could remain in the channel of currency without depreciation.
(Reply to Bosanquet; III: 219)
Ricardo quoted Bosanquet who supposed that the Bullion Report had used “the term excess of currency […] in the sense in which it is used by Dr. Smith, as denoting a quantity greater than the circulation of the country can easily absorb or employ” (ibid: 228), and he objected:
This is not the sense in which I consider the Committee to use the word excess. In that sense there can be no excess whilst the Bank does not pay in specie, because the commerce of the country can easily employ and absorb any sum which the Bank may send into circulation. It is from so understanding the word excess that Mr. Bosanquet thinks the circulation cannot be excessive, because the commerce of the country could not easily employ it. In proportion as the pound sterling becomes depreciated will the want of the nominal amount of pounds increase, and no part of the larger sum will be excessive, more than the smaller sum was before. By excess, then, the Committee must mean the difference in amount of circulation between the sum actually employed, and that sum which would be employed if the pound sterling were to regain its bullion value. This is a distinction of more consequence than at first sight appears, and Mr. Bosanquet was well aware that it was in this sense that it was used by me.
(ibid: 228‒9; Ricardo’s emphasis)
The wants of commerce consequently not only depended on the aggregate value of the commodities to be circulated but also on the value of the currency which circulated it. As Ricardo declared when on 4 March 1819 he was examined by the Commons’ Committee on the Resumption of Cash Payments:
I consider, in all cases, that the quantity of circulation must depend upon its value, and the quantity of business which it has to perform.
(V: 372‒3)
Such contention was consistent with what he had written in Principles:
A circulation can never be so abundant as to overflow; for by diminishing its value, in the same proportion you will increase its quantity, and by increasing its value, diminish its quantity.
(I: 352)
If the currency was depreciated, “the want of the nominal amount of pounds increase[d]”, and all the notes issued were absorbed by circulation:
If money be but depreciated sufficiently, there is no amount which may not be absorbed.
(Reply to Bosanquet; III: 217)
In other words, notes could be in excess while being wanted by commerce and absorbed by circulation, because “commerce is insatiable in its demands” for currency (ibid: 215) when the latter is depreciated:
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.
(ibid: 230)
The quantity of money which would circulate “if the currency were of its bullion value” is what I have called the conformable quantity of money 
– that is, the quantity consistent with the condition (3.8) of conformity of money to the standard. In what follows, an increase in the “wants of commerce” will be understood as requiring a positive 
consistent with the maintenance of the equality between the value of money and the value of bullion (in other words the absence of depreciation or appreciation).
It should be observed that for Ricardo, even supposing its value unchanged, the quantity of money required to circulate a given aggregate value of commodities did not remain constant, since it depended on the “state of credit”. Being examined on 24 March 1819 by a Lords’ Committee, he observed:
One of the causes which operate on the value and the quantity of currency, I have omitted to mention, namely, the varying state of credit, which considerably affects the quantity necessary to perform the same business.
(V: 419)
To the question “Do you conceive that the amount of trade, capital, and revenue, and the amount of currency required, must necessarily bear any fixed ratio or proportion to each other?”, Ricardo answered:
Certainly not; I think the proportion must depend on the economy in the use of money, which again must depend on the state of credit at the time.
(ibid: 420)
Already in High Price, Ricardo had observed that this “proportion” varied across countries, and in the same country according to “the degree of confidence”:
The value of the circulating medium of every country bears some proportion to the value of the commodities which it circulates. In some countries this proportion is much greater than in others, and varies, on some occasions, in the same country. It depends upon the rapidity of circulation, upon the degree of confidence and credit existing between traders, and above all, on the judicious operations of banking. In England so many means of economizing the use of circulating medium have been adopted, that its value, compared with the value of the commodities which it circulates, is probably (during a period of confidence*) reduced to as small a proportion as is practicable. (High Price; III: 90) [The footnote* reads: “In the following observations, I wish it to be understood, as supposing always the same degree of confidence and credit to exist.”]
Ricardo thus abstracted from these variations in order to determine the quantity of money required by the wants of commerce. Let me call k this supposedly constant “proportion” which “the value of the circulating medium […] bears […] to the value of the commodities which it circulates”, and Y the latter aggregate value, as given by the “real” system of exchangeable values and natural quantities. By definition, the conformable state is such that:
(7.5) |
In Chapter 3 above, the condition of conformity of money to the standard read:
(3.8) |
In other words, the value of money in the conformable state is independent of its quantity: given the definition of the monetary unit in terms of the standard – hence fixing 
– it only depends on the value of the standard. Combining (7.5) with (3.8) gives the level of the conformable quantity of money 
that is, the quantity of money consistent with the condition of conformity:
(7.6) |
The conformable quantity of money 
is thus endogenously determined: it is an increasing function of the aggregate value Y of the commodities to be circulated and a decreasing function of the value 
of the standard. When the conformable state is maintained through time, one derives from (3.8) that the value of money only varies positively with the value of the standard:
(7.7) |
Since by definition the maintenance of the conformable state implies 
one checks that this result is also obtained by making 
in (7.4). As for the conformable quantity of money, one derives from (7.6) that it varies positively with the aggregate value of commodities and inversely with the value of the standard:
(7.8) |
Let me now consider the following case: starting from a situation in which money conformed to the standard, the wants of commerce increased, while the actual quantity of money remained the same 
so that according to (7.3) 
The symmetrical case of a reduction in the wants of commerce implied an excess quantity of money, if the note issue was not contracted in proportion 
The adjustment was thus the same as in the case of a discretionary increase in the note issue, while the wants of commerce remained unaltered 
and it will be analysed in Section 7.3 below.
I recall that the adjustment is analysed under the assumption that no exogenous variation in the value 
of gold bullion occurs. As a consequence, (7.7) gives 
and (7.8) 
the conformable quantity of money rises at the same rate as the aggregate value of commodities. In the absence of any change in the actual quantity of money 
equations (7.4) and (7.8) give:
(7.9) |
The unchanged actual quantity of money M circulates a greater aggregate value of commodities Y because the actual value 
of each unit of money increases in the same proportion. This case was described by Ricardo in Proposals as follows:
When the number of transactions increase in any country from its increasing opulence and industry – bullion remaining at the same value, and the economy in the use of money also continuing unaltered – the value of money will rise on account of the increased use which will be made of it, and will continue permanently above the value of bullion, unless the quantity be increased, either by the addition of paper, or by procuring bullion to be coined into money. There will be more commodities bought and sold, but at lower prices; so that the same money will still be adequate to the increased number of transactions, by passing in each transaction at a higher value.
(Proposals; IV: 56)
Since 
the actual value of money 
rose above its conformable level 
in the same proportion as the wants of commerce increased. It has been shown in Chapter 3 that for Ricardo the condition of conformity of money to the standard could be written in value (equation (3.8)) or in price (equation (3.9) so that 
is identical with 
: the market price of gold bullion is below the legal price of gold in coin. The sequence of events was thus as follows. The increase in the aggregate value of commodities Y required an increase in the conformable quantity of money 
which was not immediately met by an increase in the actual quantity of money M. According to (7.3) the market price of gold bullion 
fell below the legal price of gold in coin 
and, according indifferently to (4.8) or (7.4), this raised the actual value of money 
above its conformable level 
. In other words, all prices 
fell in the same proportion as .
The next step was described by Ricardo supposing that “the increase in the circulation were supplied by means of coins” (ibid: 57). With 
below 
, a profit by arbitrage could be made by purchasing gold bullion in the market and having it coined at the mint:
To say that money is more valuable than bullion or the standard, is to say that bullion is selling in the market under the mint price. It can therefore be purchased, coined, and issued as money, with a profit equal to the difference between the market and mint prices. The mint price of gold is 3l. 17s. 10½d. If, from increasing opulence, more commodities came to be bought and sold, the first effect would be that the value of money would rise. Instead of 3l. 17s. 10½d. of coined money being equal in value to an ounce of gold, 3l. 15s. 0d. might be equal to that value; and therefore a profit of 2s. 10½d. might be made on every ounce of gold that was carried to the mint to be coined.
(ibid)
An endogenous increase in the quantity of money ended thus the first stage of the adjustment. The quotation goes on:
This profit, however, could not long continue; for the quantity of money which, by these means, would be added to the circulation, would sink its value, whilst the diminishing quantity of bullion in the market would also tend to raise the value of bullion to that of coin: from one or both these causes a perfect equality in their value could not fail to be soon restored.
(ibid)
The increase in the quantity of money in circulation 
raised the market price of gold bullion 
hence according to (4.8) or (7.4) it lowered the value of money 
This fall in the value of money now caused a rise in all prices 
in the same proportion as the rise in 
. To this general effect of monetary origin was added a specific effect in the sole market for bullion: since part of the quantity available had been coined, the supply was reduced and consequently the market price of bullion increased more than all other commodity prices which were not subject to this market effect. In other words, the relative value of gold bullion in terms of all other commodities rose. It should be observed that such rise was of another nature than a rise occurring when the difficulty of producing gold increased for exogenous reasons: here the rise was endogenous to the adjustment process. In equation (7.4) this endogenous rise in the value of gold 
mitigates the fall in the value of money consequent upon the increase in its quantity. As emphasised by Ricardo, at the end of the second stage of the adjustment the equality between the value of money and the value of gold bullion was restored but at a higher level than the initial one (before the increase in the wants of commerce):
It appears then, that, if the increase in the circulation were supplied by means of coin, the value both of bullion and money would, for a time at least, even after they had found their level, be higher than before; a circumstance which though often unavoidable, is inconvenient, as it affects all former contracts.
(ibid)
In Proposals, Ricardo emphasised this inconvenience to advocate the substitution of notes for coins. This did not mean, however, that with a metallic circulation no return to the initial value of money was possible: as observed above by Ricardo, it remained at a higher level “for a time at least”. A third stage of the adjustment completed it, which was the consequence of the rise in the domestic value of bullion. If in the initial situation the value of bullion in terms of all other commodities was the same in the country and abroad, so that there was no inflow or outflow of it, the domestic increase in this value – supposing that the foreign value of gold remained unaltered – made it profitable to import it, as soon as the difference was greater than the cost of importation (see Chapter 8 below). This international adjustment had been described by Ricardo in 1810 in one of his three letters to the Morning Chronicle on the Bullion Report:
Before 1797, when the Bank paid in specie, increased commerce, and increased taxation might have required, precisely as they do now, an addition to the circulating medium, which the Bank might have supplied with their notes without causing any depreciation in their value as compared with gold; but if they had refused or neglected to do so, the increased demand for money would have raised the foreign exchange above par, and the mint price of gold above the market price; or in more popular language, the market price of gold would have fallen below the mint price, and would have so continued till the Bullion Merchants had availed themselves of the advantage attending the importation of gold at the favourable exchange, and the subsequent coining of it into money, and thereby supplied the demand for currency. The exchange would then have been at or about par, and the market and mint prices of gold at the usual level.
(III: 140)
Since at the end of the second stage the value of money was equal to the domestic value of bullion (at a higher level than initially and abroad), the market price of gold bullion was again equal to the legal price of gold in coin. The import of bullion sunk its market price below the legal price, and, as in the first stage, imported gold was coined until both prices were again equalised. The value of money consequently fell and equalised with the domestic value of bullion at the initial (and foreign) level.
The superiority of paper money over coins
With a metallic circulation, the adjustment to an increase in the wants of commerce thus finally occurred: the higher required quantity of money was coined and the value of money returned to its initial level.1 But this took time and in the interval the rise in the value of money raised the value of previous debt contracts and was accompanied with a deflation of prices that hurt activity in general.2 However, there was another way of accommodating an increase in the wants of commerce: the issuing of paper money. As Ricardo observed before starting to describe the adjustment of a metallic circulation:
It is the rise in the value of money above the value of bullion which is always, in a sound state of the currency, the cause of its increase in quantity; for it is at these times that either an opening is made for the issue of more paper money, which is always attended with profit to the issuers; or that a profit is made by carrying bullion to the mint to be coined.
(Proposals; IV: 56‒7)
The increase in the wants of commerce 
led owners of commercial bills to apply for notes at the discount window of the Bank of England, and this demand was “attended with profit” to the Bank. The actual quantity of money immediately increased accordingly 
and the inconvenience consequent upon the adjustment of a metallic circulation was avoided:
This inconvenience [a rise in the value of money] is wholly got rid of, by the issue of paper money; for, in that case, there will be no additional demand for bullion; consequently its value will continue unaltered; and the new paper money, as well as the old, will conform to that value. Besides, then, all the other advantages attending the use of paper money; by the judicious management of the quantity, a degree of uniformity, which is by no other means attainable, is secured to the value of the circulating medium in which all payments are made.
(ibid: 57‒8)
One should observe that for Ricardo the increase in the note issue did not require an additional quantity of coin or bullion in the reserve of the Bank to back it. There was no threat of conversion of these notes: the paper was not in excess since its increase responded to the wants of commerce. Whether with a metallic or a paper circulation, the quantity of money adjusted endogenously. With paper money it did so quicker and safer than with specie.
7.3 The adjustment to a discretionary increase in the convertible-note issue
“Forcing” Bank of England notes into circulation
I will here consider the following case: starting from a situation in which money conformed to the standard, the bank discretionarily increased its issues of convertible notes, while the wants of commerce remained unaltered 
so that according to (7.3) 
The notes became consequently in excess, in the sense defined above when discussing the notion of “wants of commerce”: the actual quantity of money (M) became greater than its conformable one 
The symmetrical case of a discretionary contraction of the note issue implied a deficiency in the quantity of money, if the wants of commerce were unaltered
The adjustment was thus the same as in the case of an increase in the wants of commerce, while the quantity of money remained the same 
and it has been analysed above.3
A discretionary increase in the convertible-note issue could occur when the Bank made advances to government but also when it “forced” its notes into circulation by lowering its discount rate below the rate in the money market (see Chapter 9 below). In a speech in Parliament on 1 July 1822, Ricardo emphasised that the only way the Bank could increase the demand for its notes and put them into circulation was to lower its discount rate:
He [Ricardo] was glad that the Bank had determined to reduce its rate of interest to four per cent.; indeed, they would have done wrong in declining to do so, as by that means only could they bring their notes into circulation. If the Bank had not reduced its interest to four per cent. the country must necessarily resort to a metallic currency, or else notes must be issued from some other quarter, as it would be impossible for the Bank of England to put a single pound note into circulation.
(V: 223 n1)
In contrast with the case of the adjustment in the value of money to an increase in the wants of commerce, there is no description in Proposals or later writings of the adjustment to a discretionary increase in the convertible-note issue. Two reasons may be conjectured for this absence. First, Ricardo had already given such a description in High Price, both for the part of the adjustment which was common to inconvertibility and convertibility and for that part which was specific to convertibility. We will see below that when notes were convertible, the adjustment of their value was the outcome of the adjustment of the market price of gold bullion to the legal price of gold in coin, and this adjustment was not affected by the theory of value of commodities developed in Principles because, in the country where gold bullion was the standard of money, it was not subject to the general adjustment process (gravitation) that applied to any competitively produced commodity. It was thus unnecessary for Ricardo to repeat himself on something which remained valid under his general theory of value. The second reason for the absence of a detailed analysis of the adjustment in the value of money to a discretionary increase in the convertible-note issue may be that, from Proposals onwards, Ricardo repeatedly advocated a monetary plan which, although embodying the convertibility of the note, substituted a policy rule – the regulation of the quantity of money by a central bank – for private arbitrage on the standard (see Chapter 9 below). His aim was consequently to show that such a plan could implement a nearly “perfect” currency, rather than to analyse how private arbitrage actually taking advantage of convertibility did the job in an imperfect way.
The analysis of the adjustment to a discretionary increase in the convertible-note issue raises two questions: (1) In a mixed system of coins and convertible notes, which part of the adjustment fell on the circulation of coins? And (2) Which part of the adjustment depended on the behaviour of the bank of issue (the Bank of England)? Let me consider these two questions successively.
The melting of circulating coins
The part of the adjustment which fell on the circulation of coins was common to both a convertibility and an inconvertibility regime and it is why it could be described by Ricardo when in High Price he considered the case of a mixed system of coins and inconvertible notes. The adjustment is exactly symmetrical with that of an increase in the wants of commerce in a purely metallic circulation: we have seen above that when the wants of commerce exogenously increased 
the circulation of coins increased through the coining of bullion; now that the quantity of money exogenously increased (∆M / M > 0), the circulation of coins was contracted through their melting into bullion. In other words, this part of the adjustment relied on the convertibility of coin into bullion, as the adjustment to the wants of commerce in a pure metallic circulation relied on the convertibility of bullion into coin.
In High Price, this part of the adjustment was described as follows:
It has been observed [by Francis Horner4] that an increase in the paper currency will only occasion a rise in the paper or currency price of commodities, but will not cause an increase in their bullion price. This would be true at a time when the currency consisted wholly of paper not convertible into specie, but not while specie formed any part of the circulation. In the latter case the effect of an increased issue of paper would be to throw out of circulation an equal amount of specie; but this could not be done without adding to the quantity of bullion in the market, and thereby lowering its value, or in other words, increasing the bullion price of commodities. It is only in consequence of this fall in the value of the metallic currency, and of bullion, that the temptation to export them arises; and the penalties on melting the coin is the sole cause of a small difference between the value of the coin and of bullion, or a small excess of the market above the mint price.
(High Price, III: 64; note; Ricardo’s emphasis)
There were three successive stages in the adjustment of the metallic circulation. Consider an initial situation in which money conformed to the standard, so that 
The Bank then discretionarily increased its note issue, so that the aggregate quantity of money increased at a rate 
while the legal price 
of gold in coin and the quantity of money 
required by the wants of commerce remained unchanged 
During the first stage of the adjustment, the excess of 
resulted in a rise of the market price of gold bullion 
According indifferently to equation (4.8) or (7.4), and supposing no exogenous change in the value of gold bullion 
the value of money fell 
and all prices 
rose in the same proportion as 
that is, at the same rate 
as the aggregate quantity of money. Now, with 
above 
(money was depreciated), a profit by arbitrage could be made by collecting gold coins in circulation at 
, melting them, and selling bullion in the market at the higher 
. Arbitrage was implemented as soon as the difference between and was greater than the melting cost 
, which may here be neglected, as Ricardo did. An endogenous contraction in the quantity of money, equal to the quantity of coins removed from circulation, ended thus the first stage of the adjustment: “the effect of an increased issue of paper would be to throw out of circulation an equal amount of specie”.
In the second stage, this contraction in the quantity of money in circulation 
sunk the market price of gold bullion 
hence according to (4.8) or (7.4) it increased the value of money 
This rise in the value of money now caused a fall in all prices 
in the same proportion as the fall in 
. To this general effect of monetary origin was added a specific effect in the sole market for bullion: because of the increased supply consequent upon the melting of coins the market price of bullion fell more than all other commodity prices (which were not subject to this market effect). In other words, the relative value of gold bullion in terms of all other commodities fell. The quotation goes on: “but this could not be done without adding to the quantity of bullion in the market, and thereby lowering its value”. This fall was endogenous to the adjustment process and should be distinguished from an exogenous fall consequent upon the discovery of a new productive mine. In equation (7.4) this endogenous fall in the value of gold 
mitigated the rise in the value of money consequent upon the contraction in its quantity. At the end of the second stage of the adjustment the equality between the value of money and the value of gold bullion was restored but at a lower level than the initial one (before the monetary shock).
The third stage of the adjustment started, which was the consequence of the fall in the domestic value of bullion. If in the initial situation the value of bullion in terms of all other commodities was the same in the country and abroad, so that there was no inflow or outflow of it, the domestic fall in this value – supposing that the foreign value of gold remained unaltered – made it profitable to export it, as soon as the difference was greater than the cost of exportation (see Chapter 8 below). The quotation goes on: “It is only in consequence of this fall in the value of the metallic currency, and of bullion, that the temptation to export them arises.” Since at the end of the second stage the value of money was equal to the domestic value of bullion (at a lower level than initially and abroad), the market price of gold bullion was again equal to the legal price of gold in coin (or slightly above the latter, by the “small difference” due to the melting cost). The export of bullion reduced its supply in the domestic market, so that its market price rose above the legal price, and, as in the first stage, coins were removed from circulation, melted, and the bullion sold in the market for profit, until both prices were again equalised. The value of money consequently rose and equalised with the domestic value of bullion at the initial (and foreign) level.5
Now the distinction between the cases of convertibility or inconvertibility of the Bank of England note entered the picture for the second question raised above: Which part of the adjustment depended on the behaviour of the Bank? Under inconvertibility, the adjustment through the melting of coins gathered in circulation and the exporting of the bullion thus obtained restored the initial situation – the increased issue of notes simply substituting for the coins melted and exported – but nothing could prevent the Bank from increasing again its issue. The outcome of the repeated adjustment process was the complete elimination of coins from domestic circulation, which became exclusively filled with inconvertible notes. This was the situation actually achieved in England in 1810. The lack of circulating coins obviated the operation of the adjustment process and the depreciation of money increased without limit (for a discussion of this case, see Section 7.5 below).
Such a fatal outcome could not happen under convertibility. It happened under inconvertibility when the reservoir of gold provided by the domestic circulation of coins was exhausted. Such exhaustion was excluded under convertibility for a simple reason: coins to be melted with a view to exporting the bullion were not gathered in circulation (at a cost equal to the commission of money-changers) but obtained by arbitragers from the Bank of England at no cost, when they returned to the Bank the notes issued by it. This difference apparently did not affect the adjustment process based on melting the coin and exporting the bullion:
While their notes were payable in specie on demand, [the Bank] could never issue more notes than the value of the coin which would have circulated had there been no bank. If they attempted to exceed this amount, the excess would be immediately returned to them for specie; because our currency, being thereby diminished in value, could be advantageously exported, and could not be retained in our circulation.
(High Price; III: 57)
At the end of the first stage of the adjustment process, the quantity of money was now reduced in its note component (because notes were returned by arbitragers to the Bank for coins) instead of its metallic component, but in the second stage the general effect on all prices and the specific market effect on the price of bullion operated as above, with the same consequences for the value of bullion and its exportation. However, the final outcome of the adjustment process now depended on the reaction of the Bank of England to the conversion of its notes consequent upon its discretionary increased issuing.
At the time of the Bullion Essays, Ricardo already contended that, if the note was convertible into coin, a fall in the value of money would only be temporary:
Whilst Banks pay in specie there can be no additions to the circulation which can permanently lower the value of money, – because they cannot permanently lower the value of gold and silver. Those metals would not have been of greater value now if no bank had ever been heard of.
(“Notes on Bentham’s ‘Sur les prix’”; III: 278)
According to Ricardo, the only permanent cause of a fall in the value of money, when the note was convertible, was a fall in the value of the standard, and such fall could only temporarily be caused by an excess issue (in contrast with an exogenous fall consequent upon the discovery of a new mine). This endogenous fall only occurred in the interval between the moment arbitrage between the Bank and the bullion market started and the moment when it stopped. However, if the Bank re-issued the notes returned to it, the adjustment process was again set into motion and with it the fall in the value of the standard. This is why the export of bullion – permitted by the fall in its value – could not by itself correct the excess issue and eliminate depreciation, as long as the Bank re-issued the notes in excess. The excess of notes generating their depreciation could only be eliminated if something prevented the Bank from renewing its increased issues, and this is precisely what convertibility did. It prevented the Bank from issuing in excess during a prolonged period of time, because of an effect borrowed by Ricardo from Smith and Thornton:
In this manner if the Bank persisted in returning their notes into circulation, every guinea might be drawn out of their coffers. If to supply the deficiency of their stock of gold they were to purchase gold bullion at the advanced price, and have it coined into guineas, this would not remedy the evil, guineas would still be demanded, but instead of being exported would be melted and sold to the Bank as bullion at the advanced price. ‘The operations of the bank,’ observed Dr. Smith, alluding to an analogous case, ‘were upon this account somewhat like the web of Penelope, the work that was done in the day was undone in the night.’ The same sentiment is expressed by Mr. Thornton:– ‘Finding the guineas in their coffers to lessen every day, they must naturally be supposed to be desirous of replacing them by all effectual and not extravagantly expensive means. They will be disposed, to a certain degree, to buy gold, though at a losing price, and to coin it into new guineas; but they will have to do this at the very moment when many are privately melting what is coined. The one party will be melting and selling while the other is buying and coining. And each of these two contending businesses will now be carried on, not on account of an actual exportation of each melted guinea to Hamburgh, but the operation or at least a great part of it will be confined to London; the coiners and the melters living on the same spot, and giving constant employment to each other.
‘The Bank,’ continues Mr. Thornton, ‘if we suppose it, as we now do, to carry on this sort of contest with the melters, is obviously waging a very unequal war; and even though it should not be tired early, it will be likely to be tired sooner than its adversaries.’
The Bank would be obliged therefore ultimately to adopt the only remedy in their power to put a stop to the demand for guineas. They would withdraw part of their notes from circulation, till they should have increased the value of the remainder to that of gold bullion, and consequently to the value of the currencies of other countries. All advantage from the exportation of gold bullion would then cease, and there would be no temptation to exchange bank-notes for guineas.
(High Price; III: 58‒9)6
What I will henceforth call the Penelope effect operated in two steps. First, the reserve in coins at the Bank shrank with the conversion of the notes and after a certain time the Bank had to purchase bullion in the market, in order to have it coined at the mint and replenish its reserve. The export of bullion consequently stopped – “guineas would still be demanded, but instead of being exported would be melted and sold to the Bank as bullion at the advanced price” – because the sale of bullion to the Bank in the London market did not expose its owner to the cost of transporting it to a foreign market. The demand for bullion by the Bank replaced the demand for exportation and sustained its market price above the legal price of gold in coin, “the coiners [the Bank] and the melters [the arbitragers] living on the same spot, and giving constant employment to each other”, as Thornton wrote. But this was for the Bank a loss-making activity, since it purchased bullion in the market above the price at which it was legally compelled to give coins against its notes. Then the second step of the Penelope effect started: after a certain time, “the Bank would be obliged therefore ultimately to adopt the only remedy in their power to put a stop to the demand for guineas. They would withdraw part of their notes from circulation.” The Bank did that, not because it finally understood that the high price of bullion was caused by the excess issue – Ricardo repeatedly complained that Bank directors were notoriously ignorant of this relation – but simply in the hope that, issuing fewer notes through discounting, fewer notes would be returned to be converted into coin. This reduction took place until arbitragers ceased to return notes to the Bank, that is, until the spread between the market price of gold bullion and the legal price of gold in coin fell below the melting cost (so that “there would be no temptation to exchange bank-notes for guineas”) – in other words, “till they [the Bank] should have increased the value of the remainder [of notes] to that of gold bullion”.
The same adjustment was mentioned in Chapter XXVII of Principles when Ricardo discussed the positions of Smith and of the editor of the Wealth of Nations, Buchanan, on the respective effects of the debasement of the coin and of an excess issue of notes on the Bank experiencing a loss when it was obliged to purchase bullion at a high price and to have it coined at the legal price. Ricardo’s object was then to defend “the principle of limitation of quantity” (I: 353) according to which a reduction in the quantity of money, whether debased coins or notes issued in excess, could elevate its value to that of the full-bodied coin, stopping the Penelope effect:
On the principle above stated, it appears to me most clear, that by not re-issuing the paper thus brought in, the value of the whole currency, of the degraded as well as the new gold coin, would have been raised, when all demands on the Bank would have ceased.
(Principles; I: 355)
And again in a letter to Mill of 18 December 1821 where he commented on pages of the latter’s Elements of Political Economy, Ricardo evoked the losing business of the Bank:
119 “But in these circumstances” &ca. The latter part of this paragraph is not clear. Why would the bank pay for their notes when they came back £4? Answer. Their stock of gold would inevitably be soon exhausted and they would be obliged to buy in the market at £4–what they sold at £3. 17. 10½ in order to replace that stock.
(IX: 128)
In conclusion on the adjustment to a discretionary increase in the convertible-note issue, one can make four remarks. First, in contrast with the case of inconvertibility, there could not be a permanent excess of convertible notes. Second, contrary to what is usually ascribed to Ricardo, the exportation of bullion, which under inconvertibility adjusted the aggregate quantity of money when it was in excess, only played a role at the beginning of the adjustment process under convertibility. As soon as the Bank purchased bullion to replenish its metallic reserve, the exportation of bullion was replaced by a domestic adjustment: the reduction of the Bank issue under the pressure of the Penelope effect.7 Third, such adjustment explains how the rise in the market price of gold bullion above the legal price of gold in coin eventually led to a contraction in the note component of the aggregate quantity of money which restored their equality. Fourth, although arbitrage was quickly triggered by such spread, the adjustment through the Penelope effect might take quite a long time, depending on the lag with which the Bank first purchased bullion to replenish its reserve, and second contracted the issue to stop its losses. During this interval, the market price of gold bullion remained above the legal price of gold in coin: money was depreciated. According to Ricardo, this was one of the reasons why the monetary system should be improved, instead of simply restoring the pre-1797 convertibility of the Bank of England note into coin (see Chapter 9 below).
7.4 A non-quantitative approach
Two conclusions may be drawn from equations (7.2) to (7.4) above. The first derives from (7.2): the introduction by Ricardo in 1816 (in Proposals) of the condition of conformity of money to the standard gave a theoretical foundation to the proposition that, not only the high price of bullion was “a proof” of the depreciation of bank notes (as stated in 1810 in the title of High Price), but it also indicated that the actual aggregate quantity of money M was in excess of its conformable level 
Ricardo’s tour de force was thus to provide a simple test – the comparison of the two observable magnitudes and 
– of whether the actual aggregate quantity of money was or not in excess (or deficient), without having to know what this level and the required (conformable) level were.
A second conclusion derives from (7.4), which reformulates the Money–Standard Equation: Ricardo’s mature theory of money, consistent with Principles and contained in the 1819‒1823 papers, is not a quantity theory of money. Of course, the quantity of money mattered, as emphasised in Principles:
There is no point more important in issuing paper money, than to be fully impressed with the effects which follow from the principle of limitation of quantity.
(I: 353)
The above analysis of how the value of money adjusts to an increase in the wants of commerce or in the discretionary issuing of convertible notes shows that the understanding of the link between the quantity and the value of money requires putting the standard of money centre-stage, as emphasised by the sentence in Proposals which is the epigraph of the present chapter:
The only use of a standard is to regulate the quantity, and by the quantity the value of a currency.
(IV: 59)
One should refrain from jumping on the second part of the sentence to conclude that Ricardo advocated a quantity theory of money. As appears in the first part of the sentence, the quantity of a currency only “regulates” its value since it is itself regulated by the standard. The above analysis of this regulation allows clarifying the difference between Ricardo’s theory of money and the so-called Quantity Theory of Money.
The Quantity Theory of Money is traditionally presented in the form of the Cambridge equation 
(with M the supply of money, K the proportion of the real volume of their planned transactions T which individuals wish to hold as real balances, and P the price level of the commodities transacted, so that KPT is the demand for money) or in the form of the equation of exchange 
in which the velocity of circulation v replaces its reciprocal K (see Patinkin 1956: 97). The value of money 
being defined as the reciprocal of the general price level P, both expressions lead to the following relation, with K (respectively v) supposed to be a given parameter reflecting the organisation of monetary transactions:
(i) |
Equation (i) leads to the well-known conclusion that the value of money 
falls (the general price level P rises) if the exogenous supply of money M increases faster than the volume of transactions T determined independently (money being neutral in respect to aggregate output).
Ricardo’s analysis of the relation between a change in the value of money and a change in its quantity is summed-up by equation (7.4):
(7.4) |
As already emphasised in Chapters 3 and 4 above, it is necessary to distinguish between what I will call the conformable state of the monetary system – in which “money conforms to the standard”, that is 
– and situations in which these equalities are not fulfilled; these situations may thus be conveniently called monetary disequilibrium (this term being only used here because it speaks for itself in the literature, without implying that the conformable state is an equilibrium state).
In the conformable state the value of money 
is independent of its quantity and only depends on the value of the standard 
and on its legal price 
when coined into money (equation (3.8)). If this state is maintained through time, (7.4) simplifies in:
(7.7) |
The conformable value of money varies with the value of the standard and only with it. As for the conformable quantity of money 
it is a function of its value 
– hence of the value of the standard 
– and of the aggregate value Y of the commodities which it circulates (equation (7.6)). It is thus endogenously determined. In the conformable state, there is no equilibrium value of money determined by the equality between an exogenously given supply of money and a demand for real balances, in contrast with the Quantity Theory of Money. In Ricardo, 
varies positively with the aggregate value of commodities to be circulated and inversely with the value of the standard:
(7.8) |
Disequilibrium occurs when the monetary system departs from its conformable state because 
The rate of change in the value of money 
as given by (7.4), is now affected by the rate of change in the actual quantity of money 
Combining (7.8) with (7.4) gives:
(7.10) |
Formally, this equation is analogous to equation (i) of the Quantity Theory of Money. However, it should be noted that, while (i) is an equilibrium relation, (7.10) is only valid in disequilibrium: the value of money varies inversely with its quantity only in disequilibrium. Moreover, as shown in Sections 7.2 and 7.3 above, such disequilibrium triggers an endogenous adjustment of the actual to the conformable quantity of money, what Ricardo calls in Principles “the principle of limitation of quantity” (I: 353). If 
(the actual quantity of money is deficient), this adjustment operates through new coining and / or discounting of bills for notes. If 
(the actual quantity of money is in excess), it operates through melting and forced reduction in the note issue, under the pressure of the Penelope effect. Finally, Ricardo’s theory of money is specific because of the channel of transmission of a change in the quantity to the value of money. This is concealed in equation (7.4) where, thanks to (7.3), 
substitutes for 
in the Money–Standard Equation (4.8). In disequilibrium, a change in the quantity of money affects its value indirectly, through a change in the market price of the standard. This contrasts with the Quantity Theory of Money in which a change in the quantity of money affects its value directly thanks to real-balance effects on the demand for money.
To sum up, Ricardo’s integration of money in his theory of value and the Quantity Theory of Money have in common that they assume the neutrality of money in respect to the relative prices of commodities (for a justification of this assumption see Section 6.2 of Chapter 6, and see Appendix 3 for Ricardo’s aborted attempt at non-neutrality when commodities were taxed on profits). A monetary disequilibrium (that is, an inequality between the actual quantity of money and its conformable level) has no real effects on the markets for commodities, which are supposed to remain in their natural state (a corollary is the specificity of the market for gold bullion as compared with the market for any other commodity; see Section 7.6 below). However, Ricardo’s theory of money is not a quantity theory of money for three reasons: (a) In the conformable state, the causal relation between the quantity and the value of money is turned upside down; (b) “the principle of limitation of quantity” only operates in disequilibrium, provided there is convertibility; and (c) it operates on the value of money through a change in the market price of the standard. A practical consequence ensues: the bank of issue should not target the rate of change in the absolute quantity of money but the rate of change that preserves the equalisation of the value of money with its conformable level, that is, the equalisation of the market price of the standard with the legal price at which the note is convertible into it (see Chapter 9 below):
The issuers of paper money should regulate their issues solely by the price of bullion, and never by the quantity of their paper in circulation. The quantity can never be too great nor too little, while it preserves the same value as the standard.
(Proposals; IV: 64)
In contrast with the Quantity Theory of Money, an exogenously given supply of money does not confront an independently given demand for real balances, but the actual quantity of money M endogenously adjusts to the quantity 
required by the wants of commerce. When the Bank discretionarily increases its note issue, commodity prices do not rise because there is a stable demand function for money, but do so thanks to a sequence of effects in which an excess of the actual quantity of money over its conformable level raises the market price of the standard, hence lowers the value of money, hence raises all prices (except the market price of the standard which is brought back to its legal convertibility level). There is indeed a self-adjusting process (the Penelope effect) which eliminates the excess issue of notes and restores the conformity of money to the standard, but it is not the adjustment to an equilibrium value of money determined by a supply-and-demand mechanism.8 Finally, one does not need to measure the variation in the money supply (with the related problems of choosing the appropriate monetary aggregate) to explain the variation in its value; it is enough to compare the market price of gold bullion with the legal price of gold in coin to conclude whether money conforms to the standard or not.
To conclude on this point, although changes in the aggregate quantity of money affect its value in disequilibrium, Ricardo’s mature theory of money, as it is embodied in the Money–Standard Equation, is not a quantity theory of money. Modern attempts at introducing real-balance effects, at linking the quantity of money and prices through the rate of interest, or at connecting Ricardo’s alleged quantity theory of money with Say’s Law are therefore irrelevant (see Appendix 7 below).
It is striking that the sentence in Proposals which best sums up Ricardo’s view of how the standard regulates the value of money is in a paragraph where he criticised a currency without a standard for its value being unstable. And it is noteworthy that this paragraph is one of the very few references to an author who was in the eighteenth century the most acute on this question of the standard of money: James Steuart (see Deleplace 2015d). Ricardo’s judgement on Steuart was ambivalent, mixing a critique in the main text with a laudatory appreciation in a footnote, and it is to be regretted that he did not develop this point more:
This idea of a currency without a specific standard was, I believe, first advanced by Sir James Steuart,* but no one has yet been able to offer any test by which we could ascertain the uniformity in the value of a money so constituted. Those who supported this opinion did not see, that such a currency, instead of being invariable, was subject to the greatest variations, – that the only use of a standard is to regulate the quantity, and by the quantity the value of the currency – and that without a standard it would be exposed to all the fluctuations to which the ignorance or the interests of the issuers might subject it.
*The writings of Sir James Steuart on the subject of coin and money are full of instruction, and it appears surprising that he could have adopted the above opinion, which is so directly at variance with the general principles he endeavoured to establish.
(Proposals; IV: 59)
Ricardo’s contention that the value of a money without a standard could not be “regulated” originated in the Bullion Essays, where it was linked to the double inconvertibility which characterised the English currency at the time. However, the question should be raised of whether Ricardo’s mature theory of money, from Proposals onwards, applies to such money.
It was Ricardo’s contention that from 1797 (when the convertibility of the Bank of England note was suspended) to 1819 (when it was resumed) money had no standard. As he wrote in 1822 in his pamphlet On Protection to Agriculture:
It is also forgotten, that from 1797 to 1819 we had no standard whatever, by which to regulate the quantity or value of our money. Its quantity and its value depended entirely on the Bank of England, the directors of which establishment, however desirous they might have been to act with fairness and justice to the public, avowed that they were guided in their issues by principles which, it is no longer disputed, exposed the country to the greatest embarrassment. Accordingly, we find that the currency varied in value considerably during the period of 22 years, when there was no other rule for regulating its quantity and value but the will of the Bank.
(IV: 222‒3)
This was not only a retrospective statement:9 at the time of the Bullion Report, when the question of the consequences of inconvertibility was hotly debated, Ricardo had already affirmed:
That gold is no longer in practice the standard by which our currency is regulated is a truth. It is the ground of the complaint of the [Bullion] Committee (and of all who have written on the same side) against the present system.
(Reply to Bosanquet; III: 255)
Indeed, there was still a legal price of gold in coin, and the difference between the market price of gold bullion and this legal price was for Ricardo “a proof” of the depreciation of bank notes, as he wrote in the title of High Price. Formally gold was still the standard of money, but it no longer actually “regulated” its quantity, hence its value, because paper money was no longer convertible into this standard. The above analysis allows understanding why it was so. With the suspension in 1797 of the convertibility of the Bank of England note into coin, the Penelope effect had disappeared, and the quantity of notes issued faced henceforth no other limit than “the will of the Bank”. Some kind of “regulation”, however, still survived because of the other (illegal but usual) convertibility, from coin into bullion (the melting pot). The more the Bank issued notes, the more coins were displaced from domestic circulation to be melted with a view to exporting the bullion. But this “regulation” only operated during the first ten years of suspension, until nearly all coins had disappeared. With the growing shortage of coins to be obtained from money-changers against notes, the cost of gathering them increased and the melting cost increased with it, pushing the market price ever and ever upwards, without limit. This was the situation in 1809‒1810 which motivated Ricardo’s first writings and the appointment of the Bullion Committee:
The same cause which has produced a difference of from fifteen to twenty per cent. in bank-notes when compared with gold bullion, may increase it to fifty per cent. There can be no limit to the depreciation which may arise from a constantly increasing quantity of paper. The stimulus which a redundant currency gives to the exportation of the coin has acquired new force, but cannot, as formerly, relieve itself.
(High Price; III: 78)
The excess of money consequent upon an ever-increasing issue of Bank of England notes could no longer be corrected by the contraction of the metallic circulation, because there were no longer any coins to be melted with a view to exporting bullion. In the absence of the convertibility of the note into coin, the convertibility of the coin into bullion was condemned to disappear sooner or later. This double inconvertibility made the regulation of money by the standard impossible, in contrast with the double convertibility which ensured it before 1797.
This impossibility meant that not only was “gold no longer in practice the standard by which our currency is regulated”, as quoted above, but also that gold had not been replaced by another standard, contrary to what Bosanquet contended. For the latter, the Bank of England note had become the standard because it was used in the relations between the State and the public. In the Appendix to his Reply to Bosanquet, Ricardo quoted him:
Let us hear what he says: “If a pound note be the denomination, it will, of course, be asked what is the standard? The question is not easy of solution. But, considering the high proportion which the dealings between government and the public bear to the general circulation, it is probable the standard may be found in those transactions; and it seems not more difficult to imagine that the standard value of a one pound note may be the interest of 33l. 6s. 8d. – 3 per cent stock, than that such standard has reference to a metal, of which none remains in circulation”.
(III: 255‒6)
As mentioned above, there remained one use of the standard under inconvertibility: measuring the depreciation of the bank note. But this disqualified the bank note as standard, because it could evidently not measure its own depreciation. For once, Ricardo’s irony was ferocious:
So then we have a standard for a pound Bank note, it is the interest of 33l. 6s. 8d. – 3 per cent stock. Now, in what medium is this interest paid? because that must be the standard. The holder of 33l. 6s. 8d. stock receives at the Bank a one pound note. Bank notes are, therefore, according to the theory of a practical man, the standard by which alone the depreciation of Bank notes can be estimated!
A puncheon of rum has 16 per cent. of its contents taken out, and water poured in for it. What is the standard by which Mr. Bosanquet attempts to detect the adulteration? A sample of the adulterated liquor taken out of the same cask.
(ibid: 256; Ricardo’s emphasis)
The Money–Standard Equation and inconvertibility
The question is simple: does Ricardo’s mature theory of money, as it is embodied in the Money–Standard Equation, apply to inconvertibility? Since this equation, whether in the form of (4.8) or (7.4), is derived from the condition (3.8) of conformity of money to the standard, this question may be reformulated as follows: what is the meaning, if any, of this condition of conformity in the inconvertibility case?
Under convertibility, this condition has both a normative and a positive meaning. It must be fulfilled if money is to be “perfect” – to use Ricardo’s word – and it is fulfilled as the consequence of a self-adjusting process. Under inconvertibility, the normative meaning subsists but the positive one is ruled out. Although the melting of circulating coins and the exportation of the bullion thus obtained provide a corrective process to the excess issue of notes, their operation faces an obvious limit: the disappearance of all coins from domestic circulation. At this point, depreciation starts to increase without limit, and money can no longer conform to the standard. In this sense, there is no self-adjusting process under inconvertibility. The reason is obvious: the institutional set-up which guarantees its operation under convertibility does not exist under inconvertibility. Under the convertibility of the Bank of England note into coin, this set-up was made of three aspects: (a) a legal price of gold in coin; (b) convertibility both ways between bullion and the coin; and (c) convertibility one-way of the note into the coin. As it was practised after the suspension of 1797, inconvertibility preserved (a) but not (c): the Penelope effect, which constrained the issuing behaviour of the Bank of England, was discontinued. And this led after several years to the disappearance of (b): there were no longer circulating coins to be converted into bullion, hence no further correction of the excess issue of notes by the exportation of bullion.
The persistence of the normative meaning of the condition of conformity might nevertheless suggest another role for it than being the outcome of a self-adjusting process: to provide a target for a policy rule. In the absence of the full convertibility set-up, the bank of issue could mimic its working by varying the quantity of notes inversely with the sign of the difference between the market price of gold bullion and the legal price of gold in coin. Ricardo himself observed in Proposals (IV: 67) that this was “the criterion which I have so often mentioned”, and this “judicious management of the quantity” of money (ibid: 57) was one of the two pillars of his Ingot Plan. As will be developed in Chapter 9 below, this policy rule substituted for actual convertibility by making it superfluous: an excess quantity of notes was immediately corrected by the central bank when it observed a rise of the market price of bullion, before arbitrage was triggered and notes returned to the bank. However, this does not mean that for Ricardo this policy rule could dispense from convertibility. When he wrote:
On these principles, it will be seen that it is not necessary that paper money should be payable in specie to secure its value; it is only necessary that its quantity should be regulated according to the metal which is declared to be the standard.
(Principles; I: 354)
he did not contend that convertibility was unnecessary but that convertibility into coin was unnecessary and should be replaced by convertibility into bullion. With the Ingot Plan the clerks at the gold window of the Bank of England would be condemned to idleness only if convertibility forced the Bank to apply the policy rule. Only its interest – provided it was correctly understood by the directors of the Bank – could induce it to do so, and its interest was to avoid being exposed to the Penelope effect. Even the independent commissioners of the public bank advocated by Ricardo as a substitute for the Bank of England (see Chapter 9 below) would implement the policy rule if they knew it was the only way to avoid the shrinking of a metallic reserve voluntarily maintained at a low level. The stabilisation of the market price of the standard at the level of its legal price could not simply be the target of a policy rule without convertibility forcing the implementation of this rule.
I may now conclude on the question raised above: does Ricardo’s mature theory of money, as it is embodied in the Money–Standard Equation, apply to inconvertibility? The answer should be that it does not, whether as analysis of the self-adjusting property of the value of money – absent under inconvertibility – or as the theoretical foundation of a monetary policy rule – implausible under inconvertibility. The only use of Ricardo’s mature theory of money for the analysis of inconvertibility is to show that the latter is neither self-adjusting nor practicable. Nothing can be said beyond that, something which is not to be regretted since for Ricardo a money without a standard was not viable.
7.6 Conclusion: the specificity of the market for gold bullion
According to Ricardo, in a mixed system of coins and notes convertible into coin, and in the absence of a policy rule applied to the issuing of notes – such as the one advocated by him in his two plans – the quantity of money adjusted thanks to private arbitrage between the market for gold bullion and the bank of issue, whether in the case of a change in the wants of commerce or in the discretionary issuing by the bank. It is thus useful to sum up the specificity of Ricardo’s theory of money as regards the relation between the quantity and the value of money and the specificity of the market for gold bullion.
Ricardo’s analysis, although it assumed the neutrality of money in respect to the relative prices of commodities, was not a quantity theory of money. The integration of the value of money in Ricardo’s general theory of value relied on the distinction between a real cause of change – when the relative value of the standard in terms of all other commodities varied for exogenous reasons – and a monetary cause of change – when the actual quantity of money departed from its conformable level and affected the market price of the standard. Whatever the cause of change in the value of money a self-adjusting process was at work. If the cause was real (such as a change in the difficulty of production of gold bullion), the value of money settled permanently at a new level and the prices of all commodities except the standard adjusted accordingly in the same proportion. If the cause was monetary (such as a discretionary change in the note issue, not required by the wants of commerce), the value of money departed temporarily from its initial (conformable) level, before returning to it after the excess or deficiency in the quantity of money had been endogenously eliminated. This corrective process first gave rise to an exportation or importation of bullion but ended up with the Bank changing its issues appropriately for reasons of interest.
The self-adjusting process of the quantity, hence of the value, of money depended on the working of the market for gold bullion, but because gold bullion was the standard of money, its market was very peculiar. Arbitrage was practised on it as on every other market, but the conditions of this practice were unique, because arbitrage was not only with its foreign markets but also with domestic monetary institutions: the mint, where bullion was converted at a legal price into coin, and the Bank of England, where notes were converted at par into coins to be melted into bullion. The outcome of this domestic arbitrage was the equalisation of the market price of bullion with the fixed legal price at which bullion and money were convertible into each other (taking into account the costs of this convertibility). This meant that the market price of the standard was affected by the laws governing these monetary institutions and by their behaviour, independently of what happened to the market prices of all other commodities. In other words, as soon as gold bullion became the standard of money, its market price did not gravitate around its natural price – as was the case for all commodities to which the competition of capitals applied – but was stabilised thanks to an adjustment process sui generis. As quoted in the epigraph of Chapter 3 above, “In a sound state of the currency the value of gold may vary, but its price cannot” (V: 392).
From a theoretical point of view, the commodity acting as the standard of money thus resembles all other commodities in that it has two prices, one regulating (in Ricardo’s parlance) the other. For any commodity produced in competitive conditions there exists a natural price which regulates its market price (according to the so-called Smithian “gravitation”). For the standard of money there exists a legal price, to which the market price adjusts. However, this resemblance conceals a crucial difference. The natural regulating price of any commodity can be determined independently of the market process, that is, independently of any theory of the market price. In the modern Classical theory of prices to be found in Sraffa (1960) and derived from Ricardo, the determination of the price system requires the knowledge of the technical methods of production and of one exogenous distribution variable, under the assumption of a uniform rate of profit, but no particular theory of the market prices: the only assumption is that the system of (natural) prices is “adopted by the market” (Sraffa 1960: 3), whatever the market process that makes it to be so. Nothing of the kind for the legal price of the standard of money: this institutional datum only acquires an economic meaning because of a specific process which adjusts the market price of the standard to it. Far from being of secondary importance, the adjustment process of the market price of gold bullion makes the legal price effective and therefore constitutes it as an economic magnitude: for want of such adjustment, this datum would simply remain outside of economic theory, and money with it.10
It is the reason why two symmetrical errors usually stand in the way of a proper understanding of Ricardo’s monetary theory: either treating the standard of money (gold bullion) like any other commodity (its market price gravitating around its natural price) or assuming its market price as constant because it is imposed by law. Either way is equally irrelevant since it evacuates the singularity of the standard of money as commodity: the determination of its market price by two apparently contradictory adjustment processes. As a competitively produced commodity, gold bullion is subject to the competition of capitals and its market price is regulated by its cost of production in the least productive mine, which varies exogenously with the discovery of new mines and new mining techniques. The only particular assumption required by Ricardo’s analysis of this adjustment is (in modern Sraffian parlance) that gold bullion should be a non-basic commodity so that it affects the rate of profit neither in the gold-producing nor in the gold-importing country (see Chapter 5 above).11 As the standard of money, gold bullion is subject to arbitrage with domestic monetary institutions and its market price is regulated by the fixed legal price at which bullion is convertible into money and money is convertible into it, independently of its cost of production. A conclusion emerges: this double regulation of the market price of gold bullion may only be non-contradictory if one assumes that gold bullion is produced outside of the country in which it is used as the standard of money. The market price of gold bullion in the foreign producing-country (regulated by the cost of production) and its domestic market price (regulated by the legal price) are made coherent with each other through the exchange rate (equation (5.9) in Chapter 5 above).
Being produced abroad was thus for gold bullion not a point of fact, consequent upon the absence of mines in England, but a point of theory: the “sound state of a currency” – that is, the self-adjustment of a monetary system endowed with a standard – required this standard to be: (a) non-basic, and (b) produced abroad.12 In other words, the commodity used as the standard of money in an economy should not be part of the system of production of commodities that determines the relative prices and the distribution of income in this economy.
The determination of the exchange rate which linked the domestic market price of gold bullion and its price in what Ricardo called its “world market” was consequently part of the theory of money. And appropriate monetary rules could stabilise the market price of the standard at the level of its legal price – thus ensuring the conformity of money to the standard – better than a mixed system of coins and notes convertible into coin. These questions will respectively be the objects of the next two chapters.
Appendix 7: The effects of a change in the money supply: real balances, the rate of interest, and Say’s Law
Chapters 6 and 7 above have analysed the bullion-price channel of transmission of a monetary shock to the value of money hence to commodity prices, which contrasts with other transmission channels attributed to Ricardo in modern literature, based on the real-balance effect or the rate of interest. The rejection of an interpretation of Ricardo’s monetary theory in terms of the Quantity Theory of Money also leads to consider the role of Say’s Law in that theory as irrelevant.
The origin of the idea that in Ricardo a change in the money supply affected commodity prices thanks to a real-balance effect is to be found in Patinkin’s discussion of “Neoclassical monetary theory”, that is, “the once widely-accepted body of thought which organized monetary theory around a transactions or cash-balance type of equation, and which then used these equations to validate the classical theory of money” (Patinkin 1956: 96). In this context, the real-balance effect was defined by Patinkin as follows:
The most persuasive formulations of this [classical quantity] theory were developments of the following tripartite thesis: an increase in the amount of money disturbs the optimum relation between the level of money balances and the individual’s expenditures; this disturbance generates an increase in the planned volume of these expenditures (the real-balance effect); and this increase creates pressures on the price level which push it upwards until it has risen in the same proportion as the amount of money.
(ibid: 97)
Although “only Wicksell and Fisher provided complete, systematic statements of this thesis” (ibid):
[I]ndeed, the basic fact underlined by the foregoing thesis – that the causal relationship between money and prices is not at all a mechanical one, but is instead the economic consequence of the prior effect of changes in the amount of money on the demand for commodities – was already a commonplace of the classical quantity-theory tradition of Cantillon, Thornton, Ricardo, and Mill.
(ibid: 98)
Patinkin, however, adds a restriction:
It must be emphasised that, in contrast to the neoclassical ones, none of these earlier expositions of the quantity theory should be regarded as having recognized the real-balance effect in the fullest sense of the term; for none of them brought out the crucial intermediary stage of the foregoing thesis in which people increase their flow of expenditures because they feel that their stock of money is too large for their needs. Instead, in a Keynesian-like fashion, these expositions more or less directly connected the increased outflow of money expenditures with the increased inflow of money receipts.
(ibid; Patinkin’s emphasis)
In a note on “the mechanism of the quantity theory in the earlier literature”, Patinkin repeats that
[E]xamples of early expositions of the quantity theory which make it clear that an increase in the quantity of money raises prices through its prior effect in increasing demand are provided by Richard Cantillon, Henry Thornton, David Ricardo, and John Stuart Mill.
(ibid: 375)
Concerning Ricardo, Patinkin refers to four instances in the Bullion Essays and adds: “As against the purely mechanical exposition of the quantity theory in his Principles, we have the other passages already cited that show that he fully understood the effect of an increase in the quantity of money in increasing demand”, the reference to Principles being to Chapter XXVII “On currency and banks” with the mention “see in particular” its first page. It may be useful to quote the four passages in the Bullion Essays where, according to Patinkin, “the quantity of money raises prices through its prior effect in increasing demand”:
I do not dispute, that if the Bank were to bring a large additional sum of notes into the market, and offer them on loan, but that they would for a time affect the rate of interest. The same effects would follow from the discovery of a hidden treasure of gold or silver coin. If the amount were large, the Bank, or the owner of the treasure, might not be able to lend the notes or the money at four, nor perhaps, above three per cent.; but having done so, neither the notes, nor the money, would be retained unemployed by the borrowers; they would be sent into every market, and would every where raise the prices of commodities, till they were absorbed in the general circulation. It is only during the interval of the issues of the Bank, and their effect on prices, that we should be sensible of an abundance of money; interest would, during that interval, be under its natural level; but as soon as the additional sum of notes or of money became absorbed in the general circulation, the rate of interest would be as high, and new loans would be demanded with as much eagerness as before the additional issues.
(High Price, III: 91)
What would be done with the gold by the owner of the mine? It must be either employed at interest by himself, or it would finally find its way into the hands of those who would so employ it. This is its natural destination; it may pass through the hands of 100, or 1000 persons, but it could be employed in no other manner at last. Now if the mine should double the quantity of money, it would depress its value in the same proportion, and there would be double the demand for it. A merchant who before required the loan of 10,000l. would now want 20,000l.; and it could be of little importance to him whether he continued to borrow 10,000l. of the Bank, and 10,000l. of those with whom the money finally rested, or whether he borrowed the whole 20,000l. of the Bank. The analogy seems to me to be complete, and not to admit of dispute. The issues of paper not convertible are guided by the same principle, and will be attended with the same effect as if the Bank were the proprietor of the mine, and issued nothing but gold. However much gold may be increased, borrowers will increase to the same amount, in consequence of its depreciation; and the same rule is equally true with respect to paper. If money be but depreciated sufficiently, there is no amount which may not be absorbed, and it would not make the slightest difference whether the Bank with their notes actually purchased the commodities themselves, or whether they discounted the bills of those who would so employ them.
(Reply to Bosanquet, III: 217)
Suppose the paper in circulation not convertible into specie to be 20 millions, and I have credit sufficient with the bank to get a bill discounted at the Bank for £1000 – wishing to extend my business to that amount. Suppose too that all the other trades possess equal facilities and that by these various bills being discounted a million is added to the circulation. Now the possessors of this additional million have not borrowed it to let it remain idle but for the purpose of extending their different trades. The distiller goes to the corn market with his portion; the cotton manufacturer, the sugar baker &c. with theirs; the quantity of corn, sugar, and cotton in the country remaining precisely the same as before. Will not the effect of this additional million be to raise the prices of all commodities 1/20 or 5 pct. that being the proportion in which the currency is increased. These borrowers of the Bank will succeed in their object of increasing their trade, but by rendering the 20 millions which was before in circulation less efficient there will be a corresponding loss in the trade of those who were before possessed of this sum. As no addition had been made to the quantity of the corn, the sugar or the cotton but only to the prices of those commodities there would be no increased trade but a different division of it. If another million were added to the circulation by new demands for discounts, the same effects would again follow. There can be no limits to the depreciation of money from such repeated additions.
(“Notes on the Bullion Report”, III: 362‒3)
The same cause, namely, an increase of money which will induce him [the merchant who buys to sell again] to give a higher price for goods, will secure him one proportionally higher when he sells again. A competition of capitalists keeps down prices, but money is not capital. An increase of capital is attended with all the benefits enumerated; an increase of money to be retained in circulation is unattended with any benefit whatever.
(“Notes on Trotter”, III: 390)
These quotations – all from the Bullion Essays – convey an impression of ambivalence. The first and the third seem to validate the real-balance effect interpretation: a given demand for real balances – dependent on the aggregate value of commodities to be circulated – faces an exogenous increase in the money supply, and the commodity prices adjust upwards, through the spending of undesired money balances. It is thus no surprise that, referring to Patinkin, Hollander (1979: 480‒1) also used these quotations as proofs of “a clear account of the effect on commodity demand of increased money supplies” (ibid: 481n).
In the second quotation, however, the picture is different. Leaving aside the “analogy” between a mine and the Bank (see Chapter 4 above), commodity prices increase as a consequence of the fall in the value of money which is itself caused by an increase in its supply. The demand for money adjusts to commodity prices, not the other way round. As for the fourth quotation, it is unclear whether the merchant “give[s] a higher price for goods” because he agrees to do so (to deplete his undesired money balances) or is forced to do so (because the seller has passed the fall in the value of money on to price).
One should notice that no quotation consistent with the real-balance effect was found by Patinkin or others in writings beginning with Proposals for an Economical and Secure Currency (1816). On the contrary, there are in them numerous examples of statements along the line of the second quotation, as in Principles:
The demand for money is not for a definite quantity, as is the demand for clothes, or for food. The demand for money is regulated entirely by its value, and its value by its quantity.
(I: 193)
In addition, Ricardo’s assumption in Principles that the system of relative prices of commodities is unaffected by changes in the value of money would require, if the real-balance effect were introduced, that the spending of undesired money balances left relative prices unaltered, an assumption difficult to admit. In contrast, as noted in Chapter 6 above, the competition of capitals for the general rate of profit forced all sellers of commodities to pass the fall in the value of money on to their prices in the same proportion, leaving relative prices unaltered.
In any case, the contention in the present book that Ricardo’s mature theory of money was not a “quantity theory” – in Patinkin’s sense – makes it irrelevant to search for a real-balance effect in it.13
2. The rate of interest and changes in commodity prices
The fact that in the instances above the quantity of money increases through discounts by the Bank of England has led some commentators to emphasise the role of the rate of interest in the transmission of a change in the money supply to commodity prices.
Inquiring about the relation between the rate of interest and the rate of profit in Ricardo, Panico (1988) emphasises the relation between the market for loans and the markets for commodities:
As soon as the market interest rate is below that natural level ‘solely’ determined by the rate of profits (see Ricardo, III, pp. 143 and 376), the demand for loans increases. The opposite happens if the market rate is above the average rate. According to Ricardo, this mechanism enforces the gravitation of the market around the average rate. The increased (decreased) demand for loans will be attended by an increased (decreased) demand for commodities, which creates a rise (fall) in the price level. This process will go on up to when the market interest rate adjusts to the average interest rate.
(Panico, 1988: 17; Panico’s emphasis)
The above quotation from High Price is then used to show that “the length of the variation of the market interest rate around the average interest rate […] depended upon the time required by the borrowers to spend in the commodity market the money obtained through new loans” (ibid: 18).
Davis (2005) leans on the same quotation to emphasise the role of the rate of interest in the rise of the price level:
Ricardo envisioned the process as follows. An expansion of the money supply lowers the rate of interest. The fall of interest stimulates demand in all markets and thereby raises prices. As prices rise, the demand for loanable funds increases until such time as interest rates return to normal levels.
(Davis 2005: 9)
As the quotation from Ricardo makes clear, the rise in prices was envisaged by him to contend that an increase in the amount of money did not permanently lower the rate of interest, which ultimately depended on the rate of profit – a statement repeated over and over again from the Bullion Essays to the 1819‒1823 papers. Although Ricardo admitted that an increased note issue by the Bank of England temporarily depressed the rate of interest, it is doubtful that any functional relation according to which “the fall of interest stimulates demand in all markets” might be found in his writings. The fall in the rate of interest and the rise in prices were two parallel effects of the increase in the money supply, including when there was a speculative boom. Being examined on 26 March 1819 by the Lords’ Committee on the Resumption of Cash Payments and asked “Has not the increase of prices during the progressive depreciation of paper a tendency to produce over-trading, and excessive speculation?” Ricardo answered:
I think over-speculation has rather been encouraged by the facility with which speculators have been enabled to raise money upon discount, in consequence of the progressive increase of paper issues.
(V: 446)
The question of the effect of a change in the quantity of money on general prices through a change in the rate of interest could be clarified by a comparison between Ricardo and Wicksell, who in his Lectures on Political Economy (originally published in 1901‒1906 and translated into English in 1934‒1935) elaborated his own theory by starting from a detailed and critical analysis of Ricardo’s. The following aspects are only an indication of the direction that such a comparison could take. As mentioned in Chapters 5 and 7 above, Ricardo contended that any quantity of money could “perform the same functions in circulation” depending on its value, because “the demand for money is not for a definite quantity, as is the demand for clothes, or for food. The demand for money is regulated entirely by its value, and its value by its quantity” (Principles; I: 193). This idea was echoed in Wicksell:
That a large and a small quantity of money can serve the same purposes of turnover if commodity prices rise or fall proportionately to the quantity is one thing. It is another thing to show why such a change of price must always follow a change in the quantity of money and to describe what happens.
(Wicksell 1934‒1935, Vol. II: 160; Wicksell’s emphasis)
Ricardo’s analysis and what Wicksell calls “the positive solution” to the problem of money (ibid: 190 sq.) have in common to consider an indirect link between the quantity of money and general prices, to describe the adjustment triggered by a monetary disequilibrium and restrict it to a system with convertibility (Ricardo excluding it under inconvertibility and Wicksell in “a pure credit economy”), and to suggest a policy rule that allows stabilising the value of money at the required level without having to know what this level is. However, they differ on the exact content of each of these aspects: the transmission channel of the quantity of money to prices is the market price of gold in Ricardo and the bank rate of interest in Wicksell; the adjustment to a discretionary increase in the quantity of money restores the initial level of prices in Ricardo and only stops their rise in Wicksell (prices remaining at the new equilibrium level); the policy rule enforces in Ricardo the fulfilment of the equality between the value of money and the (unknown) value of the standard and in Wicksell that of the equality between the bank rate of interest and the (unknown) natural rate. The general perspective is also different: while Wicksell wished to rescue the Quantity Theory of Money from its shortcomings, Ricardo’s analysis, as shown in Chapter 7 above, was not a quantity theory of money.
The usual link between the role of Say’s Law in Ricardo’s macroeconomics and his alleged quantity theory of money wanders from the subject if, beginning with Proposals on Economical and Secure Currency, such theory is not to be found in Ricardo. The conventional view is, however, endorsed by King (2013: 120): “There is a very clear link between Ricardo’s position on Say’s Law and his approach to monetary theory, and here again there is strong support for the conventional interpretation”, an interpretation which King summarises as follows:
[Say’s Law] induced Ricardo to deny the existence of ‘general gluts’, and with them any possibility that output might be constrained by deficient effective demand. This also led Ricardo to endorse a strict version of the Quantity Theory of Money and thereby also to accept the neutrality of money.
(ibid: 112‒13)
In his “Note on Say’s Law”, Patinkin had discussed “the question as to whether Say, Ricardo, Mill – both senior and junior – and other deniers of the possibility of a ‘general glut’ did or did not think in terms of this [Say’s] identity” (Patinkin 1956: 472). For Patinkin, Keynes and Lange subscribed to the interpretation of the Classicals as advocates of Say’s Law as an identity, while Becker, Baumol, and Schumpeter:
[M]aintain that Say’s Law was not intended as an identity; that classical economists had reference only to the long-run ability of the economy to absorb any increase in output; and that this is attested by their explicit recognition of the possibility of short-run oversupply in the market.
(ibid: 474)
Concerning Ricardo, Patinkin mentioned several illustrations tending to give credit to the latter interpretation but nevertheless concluded: “Even if we accept this secular interpretation of Say’s Law – and the evidence in favour of it is convincing – we must again emphasize that classical economists failed to specify the market mechanism which makes this law valid” (ibid: 475).
Taking up this question Hollander (1979) devoted a whole chapter to “The quantity theory and the law of markets”, which concluded that:
[M]y analysis of the quantity theory earlier in this chapter has demonstrated a conscious allowance on Ricardo’s part for depressions generated by sudden contraction of the money supply, and a conscious treatment of the effects on prices generated by money supply increases in terms of an initial pressure on commodity demand in commodity markets.
(Hollander 1979: 512)
According to Hollander, an illustration of the fact that “it is ‘Say’s Equality’ to which Ricardo ascribed” (ibid) was provided by Ricardo’s position in the post-war depression.
Following Hollander, there has been a vast literature which thus concerned both the analytics of Say’s Law in Ricardo and its use in his understanding of facts. Peach (1993) and O’Brien (1995) converge in considering that Ricardo was guilty of Schumpeter’s famous Ricardian-Vice accusation of applying “to the solution of practical problems” a method in which “the desired results emerged almost as tautologies” (Schumpeter 1954: 473). For Peach:
[Say’s Law] was also much more than a purely abstract doctrine. The ‘law’, which subsumed the principles of rationality and near-perfect market knowledge, was evidently thought to capture the modus operandi of the real economic system; and, contrary to S. Hollander’s account, the nature of the doctrine and, more importantly, the manner in which it was applied by Ricardo, are alone sufficient to convict him on the charge of the ‘Ricardian Vice’. This judgment stands regardless of his adherence to the ‘equality’ version of the doctrine (an interpretation which, I argue, is weakly preferable to the ‘identity’ alternative). For, although Ricardo did make limited allowance for temporary maladjustments, his emphasis always remained firmly, and characteristically, on ‘equilibrium’ tendencies. Moreover, when he was confronted by an external world which confounded his ‘law’-based expectations, his response was to blame reality, not his model.
(Peach 1993: 15, Peach’s emphasis; see also: 140‒1)
For O’Brien (1995: 52), Schumpeter’s Ricardian Vice implied:
[A]nother vice which may be called the ‘Ricardian telescope’ […] Ricardo habitually concentrated on the long-run equilibrium position and ignored the process of adjustment. Economists who concentrate only upon the relationship between the long-run equilibrium values of macroeconomic variables are guilty of the vice of employing the Ricardian Telescope.
This evaluation was left unchanged twelve years later when this article was included as Chapter 5 of O’Brien (2007: 81‒2).
Other authors, however, have introduced qualifications to such a charge. Before Hollander, Gootzeit had tried to show that “a more careful formulation of its basic elements would have saved the Ricardian system from one of its most fundamental weaknesses – the lack of integration between the monetary and real sectors of the economy” (Gootzeit 1975: ix). For him, it is true that in Principles Ricardo “did not think it necessary to include a carefully coordinated analysis of his monetary system, which was more descriptive than theoretical” (ibid: 9). However, “Ricardo’s monetary writings [the Bullion Essays] exhibited a ‘sophisticated’ version of Say’s Law because his analysis of the market clearing relationship in a growing inflationary economy concentrated on short run maladjustments generated from the money to the output market” (ibid: 19). It should be noted that this “integration” suggested by Gootzeit takes for granted that Ricardo advocated a quantity theory of money.
Ahiakpor (1985: 19‒20) also emphasises that:
[W]ith regard to the variations in the quantity of money and their possible effect on relative prices, Ricardo offered three reasons for the non-neutrality of money: […] lagged adjustment [of nominal prices], income distributional effects, […] and the role played by taxes on prices (or profits).
For Davis (2005: 185):
Ricardo’s preference for stable prices above all other monetary objectives does not imply that he was blind to the effects of money on output and employment. His ingot plan, his allowance for devaluation if the currency was highly depreciated, and his criticism of the Bank’s return to gold all demonstrate that he recognized the nonneutrality of money.
Focusing on Ricardo as “an empirical economist”, Davis maintains that “Ricardo’s account of the postwar period relied on a distinction between temporary and permanent causes of distress” (ibid: 41) with monetary shocks being listed in the temporary causes, and that he “attributed postwar economic crises to a series of exogenous shocks [to which] Britain responded quickly”, a diagnosis confirmed by historical facts (ibid: 37). Recognising the link in Ricardo between his quantity theory of money and short-run adjustment does not, however, necessarily lead to exonerate him from the toxic influence of Say’s Law, as illustrated by Green:
This group [the bullionists “led by Ricardo”] maintained that exogenous changes in the money supply would be reflected in corresponding variations in the price level, which followed from their assumption of a fixed level of output and monetary velocity. In other words, their short-run quantity ‘theory’ of money was no theory at all but simply a logical outcome of assuming Say’s Law. The inflationary process was seen as the transitional mechanism by which monetary deviations were corrected: ‘That commodities would rise or fall in price, in proportion to the increase or diminution of money, I assume as a fact which is incontrovertible’ (Ricardo, Works, III, p. 193fn).
(Green 1998: 137; my emphasis)
As shown by equation (7.4) in Chapter 7 above, Ricardo’s “incontrovertible” conclusion neither requires a quantity theory of money nor does it derive from Say’s Law.
There is no doubt that the existence of maladjustments in the employment of capital was considered by Ricardo as a possible cause of economic crisis. Sraffa quotes a witness of a discussion between Malthus and Ricardo in December 1820 writing in his diary:
Mr Ricardo and Mr Malthus came and entertained us for two hours and upwards with an argument in defence of their respective theories on Political Economy. Mr Malthus contending that the present distress arose from unemployed capital and Ricardo from misemployed capital which would soon assume its proper channels.
(quoted in VIII: 334n)
A little earlier, Ricardo had written to Malthus on 9 October 1820:
With abundance of capital and a low price of labour there cannot fail to be some employments which would yield good profits, and if a superior genius had the arrangement of the capital of the country under his controul, he might, in a very little time, make trade as active as ever. Men err in their productions, there is no deficiency of demand.
(VIII: 277)
However, as mentioned in Chapter 7 above, Ricardo’s method was to analyse monetary disequilibrium – defined by the divergence between the actual and the conformable quantities of money – in reference to a real economy in its natural state. The study of maladjustments in the employment of capital appears thus separate from the study of monetary disequilibrium.14
In any case, despite contrasting views in modern literature about the link between Ricardo’s alleged quantity theory of money and his use of Say’s Law, this question loses a great part of its importance in the absence of such theory.
Notes
1 Unless the demand for import addressed to the world market for bullion induced the opening of less productive mines and consequently raised the value of bullion, hence the value of money. This might occur if the wants of commerce increased simultaneously in all countries, but was unlikely if it was the case in only one.
2 The “great distress” suffered by the economy in the interval before the coining of imported bullion was also emphasised by Ricardo in the analytically identical case of an import of bullion generated by a discretionary contraction in the note issue by the Bank of England. See the speech in Parliament on 12 June 1822 referred to in the following note.
3 In a speech in Parliament on 12 June 1822, Ricardo emphasised the hurtful consequences for the economy of a discretionary contraction in the note issue by the Bank of England, because there would be an interval of “great distress” before imported gold would be coined and fill the gap in the required quantity of money:
His hon. friend had said, that whilst the Bank was obliged to pay its notes in gold, the public had no interest in interfering with the Bank respecting the amount of the paper circulation, for if it were too low, the deficiency would be supplied by the importation of gold, and if it were too high, it would be reduced by the exchange of paper for gold. In this opinion he did not entirely concur, because there might be an interval during which the country might sustain great inconvenience from an undue reduction of the Bank circulation. Let him put a case to elucidate his views on this subject. Suppose the Bank were to reduce the amount of their issues to five millions, what would be the consequence? The foreign exchanges would be turned in our favour, and large quantities of bullion would be imported. This bullion would be ultimately coined into money, and would replace the paper-money which had previously been withdrawn; but, before it was so coined, while all these operations were going on, the currency would be at a very low level, the prices of commodities would fall, and great distress would be suffered. Something of this kind had, in fact, happened.
(V: 199‒200)
4 In an article published in 1802 in the Edinburgh Review and commenting on Thornton’s Paper Credit (see III: 64 n3). At that time there were still coins in circulation side by side with inconvertible Bank of England notes, and it is the reason why Ricardo discussed this case, although when he did so (in 1810) nearly all coins had disappeared.
5 The role of the specific effect in the market for bullion in lowering its value and prompting its exportation was a contrario emphasised by Ricardo when he described the actual case (in 1810): all coins having disappeared from domestic circulation, the supply of bullion in the market could no longer be increased by their melting, hence the value of bullion did not fall. The above-quoted note in High Price continues:
When the circulation consists wholly of paper, any increase in its quantity will raise the money price of bullion without lowering its value, in the same manner, and in the same proportion, as it will raise the prices of other commodities, and for the same reason will lower the foreign exchanges; but this will only be a nominal, not a real fall, and will not occasion the exportation of bullion, because the real value of bullion will not be diminished, as there will be no increase to the quantity in the market.
(High Price, III: 64; note; Ricardo’s emphasis)
6 In the Appendix to the fourth edition of High Price, Ricardo used an expression very similar to the one he quoted from Thornton about “the coiners and melters”: with his plan of convertibility of the Bank of England note into bullion, “there would be no temptation to melt the coin, and consequently the labour which has been so uselessly bestowed by one party in recoining what another party found it their interest to melt into bullion, would be effectually saved” (III: 127).
7 Comparing the inconvertibility case with two monetary systems, one with “a mixed currency of coin and notes, convertible into specie on demand, and freedom of metallic exportation” and one “where (convertible) notes alone circulate”, Hollander (1979: 439‒40) concludes:
As far as concerns the mechanism of bullion movement, the two cases of convertibility are analytically identical with the inconvertible case where a mixed currency circulates, for in this case too an addition to the note issue will generate corrective outflows.
(Hollander’s emphasis)
It appears that for the “corrective outflows” of bullion convertibility and inconvertibility are not “analytically identical”: under convertibility of the Bank of England note, the reduction of the note issue under the pressure of the Penelope effect sooner or later substitutes for the exportation of bullion as corrective process.
8 Marcuzzo and Rosselli also reject the traditional interpretation of Ricardo’s monetary theory as a quantity theory of money, and they contrast the notion of “natural” quantity of money with the usual notion of “equilibrium” quantity of money: “It is our contention that the definition of the natural quantity of money in Ricardo is given by analogy with the definition of natural wages and natural prices” (Marcuzzo and Rosselli 2015: 374). This “analogy” is based on two common aspects: first, “the natural quantity of money, like other natural magnitudes, puts to rest the forces that determine its changes”; second, a distinction applies between variations which “may be permanent, that is, reflecting changes in the structure of the economy that require a different natural level for the quantity of money, or temporary” (ibid). These two aspects are obviously linked: the “temporary” deviations from the “permanent” level of “natural” magnitudes are corrected by a self-adjusting process, and the question is thus whether the analogy may hold for it. This seems dubious for two reasons. First, the quantity of money permanently required by the wants of commerce is endogenous, in contrast with natural quantities of commodities which are exogenous. Second, the adjustment process is specific to money: one thing is to say that the adjustment of the quantity of money is market-driven because it responds to profitability conditions in the bullion market; another thing is to draw an analogy with the adjustment in the markets for all other commodities. Arbitrage in the domestic market for gold bullion is permitted by the existence of an institutional set-up (convertibility at a legal price) which makes this particular commodity the standard of money and has no equivalent for any other commodity. On the specificity of this market, see Section 7.6 above.
9 See also in the manuscript “Draft on Peel” of December 1821:
He supposed that the reverting from a currency regulated by no standard, to one regulated by a fixed one, the greatest care would be taken to make the transition as little burthensome as possible, but the fact is that if the object had been to make the alteration from the one system to the other as distressing to the country as possible no measures could have been taken by the Bank of England so well calculated to produce that effect as those which they actually adopted.
(V: 519)
10 In modern parlance, one would say that for all other commodities produced in competitive conditions the question of the existence of the price system can be solved in the absence of a definite analysis of its stability – the price system analysed in Sraffa (1960) is an example of this method – while for the commodity acting as the standard of money, the existence of its legal price as an economic magnitude depends on a specific analysis of its stability. In other words, the Ricardo–Sraffa tradition on the price question leaves stability (the determination of price in disequilibrium) outside the theory, while Ricardo’s theory of money puts it centre-stage: the legal price of the standard actually rules the roost insofar as its market price is subject to a definite adjustment process.
11 “The criterion is whether a commodity enters (no matter whether directly or indirectly) into the production of all commodities. Those that do we shall call basic, and those that do not, non-basic products” (Sraffa 1960: 8; Sraffa’s emphasis). The consequences of this distinction are as follows:
These [non-basic] products have no part in the determination of the system. Their role is purely passive. If an invention were to reduce by half the quantity of each of the means of production which are required to produce a unit of a ‘luxury’ commodity of this type, the commodity itself would be halved in price, but there would be no further consequences; the price-relations of the other products and the rate of profits would remain unaffected. But if such a change occurred in the production of the opposite [basic] type, which does enter the means of production, all prices would be affected and the rate of profits would be changed.
(ibid: 7‒8; Sraffa’s emphasis)
12 In this sense, Ricardo’s image of a gold mine being discovered on the premises of the Bank of England was not only inappropriate to the understanding of an increased note issue (see Chapter 4 above) but also unfortunate for the understanding of the nature of the standard. This may be the reason why Ricardo only used this image in the Bullion Essays, not in his mature theory of money, which put the standard centre-stage.
13 Discussing Ricardo’s “failure to tease out the implications for the demand for money of a bad harvest or of foreign remittances”, Glasner (2013: 20) remarks that:
[T]he demand for money was not part of Ricardo’s basic theoretical repertoire, although he was able to recognize cases in which there was an exceptional demand for money, e.g. during a financial crisis. The idea of a functional change in the demand for money was just slightly beyond the limits of his prodigious theoretical reach.
The absence of a “functional” demand for money was rather an advantage than a shortcoming in Ricardo’s theory of money, and his analysis of a shock on the exchange rate could dispense with it (see Chapter 8 below).
14 For a different view see Martin (2008).
