urgently required to prevent a fall in its power to buy not only gold but also other commodities in general, seems ill-judged on the part of an authority who desires stability of prices.

It will perhaps be said that Mr. Keynes' rejection of currency limitation in favour of credit limitation is an obiter dictum unnecessary for his main purpose, which is to urge that we should continue to use a paper standard, but should regulate its value so as to keep it stable in purchasing power over commodities, allowing gold and foreign exchanges to go hang. Whether this would be desirable or not depends on our estimation of a number of probabilities, such as the likelihood (very small surely) of many other countries adopting the paper standard arrived at by the Government of Great Britain and Northern Ireland, and the possibilities of great changes in the productiveness of gold-producing. But before discussing whether we should make our pound stable in terms of commodities or in terms of dollars or gold, we may as well make sure that we know how to do it. We shall certainly make a mess of any scheme of regulation if we refuse to face the elementary fact that currency is no exception to the general theory of value, but, like other things, is cheapened by increased supply and made dearer by increased demand, or if we blind ourselves to the fact that the British Treasury is the only body which can supply Currency Notes and which can afford to burn them, or finally, if we imagine that under existing circumstances it is anything but the will of the Treasury to raise money in other ways which determines whether any fiscal year shall end with more or with less Currency Notes outstanding than it began.

EDWIN CANNAN

A COMMENT ON PROFESSOR CANNAN'S ARTICLE

SINCE Professor Cannan quotes and criticises a sentence from my book, but does not quote, criticise or allude in any way to the arguments which led up to my conclusion, I will re-state them briefly.

I criticised the old-fashioned policy of looking to the volume of legal-tender money in circulation as the regulator of the standard of value mainly on the two grounds: (1) that, by concentrating too much on one factor in the quantity equation to the exclusion of the others, it was theoretically unsound; and (2) that, used as a criterion for compensatory action through the bank-rate or otherwise, it gave the signal too late and was therefore practically inefficient.

(1) While the possibility of variations in the factor of the quantity equation which, in my book, I called the "volume of real balances," but which corresponds to what is called by others the "velocity of circulation of bank deposits" (which seems to me an inconveniently artificial conception), has always been admitted by sensible writers on Monetary Theory, it has been frequently assumed that, except over relatively long periods, these variations are not very large. We now know, however,— at least I think we do-that these variations can be both very large and very rapid, and, indeed, that, where we are concerned with the period of the credit cycle, it is they which are at the root of the trouble.

Let me repeat the quantity equation in the form in which I stated it in A Tract on Monetary Reform (p. 77):

=

n = p(k + rk')

where n number of units of "cash" in circulation (defined on p. 83 as being, in the case of Great Britain, "note circulation plus private deposits at the Bank of England ").

=

price of each " consumption unit," or in other words the index-number of prices.

knumber of consumption units, the monetary equivalent of which the public find it convenient to keep in 66 cash."

k' ditto which the public find it convenient to keep in bank balances available against cheques.

r = the proportion of their potential liabilities (k') to the public which the banks keep in "cash."

No. 133.-VOL. XXXIV.

F

Now, the old-fashioned doctrine used to be that if n could be kept reasonably steady, all would be well. My object was to point out that if k and k' were capable of violent fluctuation, steadiness of n might be positively harmful and must be reflected in an extreme unsteadiness of p,-this being, in fact, what has generally happened in booms and depressions of trade. As an illustration, I showed (p. 84) that if, beginning with October 1922, k and k' were to fall back to their values in October 1920, prices would rise 30 per cent. without any change whatever in the volume of cash or the reserve policy of the banks." In short, Professor Cannan's policy could not save us from a very violent oscillation. I proposed, therefore, that when there were signs of a tendency of k and k' to change, this should be counteracted partly by appropriate movements of bank-rate and partly by direct action on the magnitude of n,-though this latter would take the form, at least in the first instance, of varying that part of n which consists, not of bank-notes, but of private deposits at the Bank of England.

In short, the policy of fixing the value of n by law is unsound, because the right value of n is not always the same but is constantly fluctuating. The same volume of note-issue, which is violently deflationary at one phase of the credit cycle, may be violently inflationary at another phase. This is equally true whether the primary object of our monetary policy is to keep the standard of value steady in terms of gold or in terms of commodities.

(2) But, furthermore, in a modern community with a developed banking system, an expansion in the circulation of legal-tender money is generally the last phase of a lengthy process. A tendency towards inflation can operate for a long time before it eventuates in a demand for more money in circulation; and by the time this point has been reached the thing may be out of hand and practically uncontrollable. To prescribe an effective maximum to the note-issue and to place one's reliance on that is like prescribing that a patient's temperature shall not exceed 99° and leaving him uncared for until the thermometer registers that figure, by which time, maybe, nothing on earth can prevent his temperature from rising very much higher.

To explain completely why this is the case would involve a lengthy analysis. But I may instance one point in particular which is often overlooked. The current price quotations of wholesale staple commodities are those at which goods are being exchanged at the moment, or, rather, in the majority of cases,

at which contracts are being concluded for the exchange of goods at a later date. Now these transactions constitute only a very small proportion of the demand for bankers' services on the date in question. The bulk of the bankers' advances and the bulk of the cheques changing hands on that day relate to transactions which had been arranged some little time, possibly some months, previously. Thus there is an appreciable interval of time, perhaps from three to six months according to circumstances, before a change in wholesale prices produces its full effect on the balance sheets of the banks. Moreover, there may be a yet further interval before we experience the full effect of the above on the demand for Currency Notes. Currency Notes are largely required for the payment of wages, and the volume in circulation is mainly governed by the level of money wages and of retail prices. In the long run these are likely to move more or less in the same proportion as wholesale prices. But the causal connection between the two is very far from being instantaneous. Thus to allow prices and then credit to expand or to contract until the effect is felt in a demand for more or less money in circulation is to court disaster. By that time innumerable contracts will have been entered into which cannot be cancelled, and the volume of money in circulation can only be kept in check at the expense of bankrupting the business world,-a course often followed in former days when Professor Cannan's doctrines still held the field.

For these reasons those responsible for monetary policy must keep their weather eye on almost everything except the volume of the note-issue; that is to say, so long as they are doing their job properly. For the note-issue tells much more about how they have been acting in the past than about what they should do in the future; and a big movement in it proves, not that the moment has just arrived for changing the course, but that the navigation is already at fault and that they are on the rocks. To depend on the volume of the note-issue as the criterion of action, is like navigating with nothing whatever but a plumbline and with one's eyes closed to the skies and the horizon.

There is nothing in this, of course, to conflict with Professor Cannan's doctrine "that due limitation of the amount of a currency is necessary for the maintenance of its purchasing power," or with the view that a limitation of note-issue may be required as a check against very gross abuse; though even in this case the method is not always efficient, because, when once matters are out of hand, it is impossible to enforce the limitation,

and it is idle to cry after the event, that, if only it had been enforced, subsequent disasters would have been avoided.

I also agree most strongly with Professor Cannan's satire against those who lay stress on a revival of trade "causing a legitimate demand for currency,"-"I am not forcing my currency on anyone." This theory generally leads to contracting the basis of credit just when it ought to be expanded, and expanding it when it ought to be contracted. Some endorsements of this most erroneous theory in this year's speeches by chairmen of the "Big Five" indicate a real risk of a period of inflation ahead of us.

On the other hand, there is much else in his article which I believe to be confused and wrong. Professor Cannan is unsympathetic with nearly everything worth reading-as it seems to me-which has been written on monetary theory in the last ten years. Yet the almost revolutionary improvement in our understanding of the mechanism of money and credit and of the analysis of the trade cycle, recently effected by the united efforts of many thinkers,1 may prove to be one of the most important advances in economic thought ever made. The ideas are new. They are only just beginning to be capable of complete or clear expression. It is natural that middle-aged bankers should feel shy. But it is not natural that Professor Cannan should write as though none of all this existed, as though his own subject were incapable of development and progress, and as though the last word had been said years ago in elementary text-books.

J. M. KEYNES

1 Mr. Bellerby has lately assembled in his Control of Credit, published by Messrs. P. S. King, (38.) for the International Association on Unemployment, an impressive collection of opinions from many sources.