730. R. G. Hawtrey to Harrod , 24 December 1937 [a] , [1]

[Replies to 700 ]

18, Elm Park Gardens, [London] S.W. 10.

24 December 1937

Dear Harrod,

I am sure you will not mind that I have delayed so long my reply to your letter of the 1st October (which I return for reference).

1) Stocks. [2] The fact that your sample business men all denied taking the rate of interest into account does not surprise me. I note that they were all engaged in industry and not in dealing. Originally I formulated my theory without including producers at all, but a few years ago it struck me that it applied in principle to them, though there would be many who in practice would be little if at all amenable to the rate of interest. Manufacturers' stocks are often of primary products or materials in an early stage of manufacture subject to big fluctuations of price compared with which the charge for interest would be trifling. Their needs also depending usually on actual forward contracts for their finished output are more definitely calculable than those of dealers. And many manufacturers keep clear of bank advances altogether, facing the inconvenience of occasional superfluous idle cash rather than be dependent on a banker.

Presumably some of your sample manufacturers do rely on bank advances. But it is more than five years since Bank rate changed, [3] and they would probably have no very clear recollection of the way dear money used to affect them. The purchase of goods for stock is usually a matter of routine, which has not been disturbed by this cause for five years.

But if you ask a manufacturer whether he minds being charged 8 per cent. for his overdraft, he will reply with no uncertain voice. If you further press him to say whether, if he were, he would take steps to reduce his overdraft, and if so how, you would very soon find that postponing the purchase (or rather the delivery) of goods for stock is the only effective method readily open to him. He might be rather reluctant to say anything at all about the subject, if it is associated in his mind with unwelcome hints from his banker which he would rather forget and which he regards as altogether exceptional and as a highly confidential matter.

You say, "It would be rash to deny some effect of rate changes, but when you reduce the matter to quantitative terms and speak of a 10 per cent. change, I say on the strength of this evidence non credo." [4] But from the point of view of the trader a 10 per cent. change is quite a negligible matter. Whether he lays in 90 days' or 80 days' supplies at a time makes no appreciable difference to him. And when he orders 90 days' supplies it makes very little difference to him whether he calls for the whole lot to be delivered in a week or asks for deliveries to be in two or half-a-dozen instalments. If he stipulates for half-a-dozen he will pay interest on an average of days' supply in excess of his minimum stock; if for two, on ; if for a single delivery on 45. Even if his minimum stock is 30 days' supply (a high estimate) the average stock varies on these suppositions, from to 75 days' supply. Yet an average variation of 10 per cent., or even 5, carried into effect by traders in general in a period of three months would have a very considerable effect on producers' order books.

You say that you "can hardly believe that the majority of retailers are sufficiently sophisticated to be very responsive." But no sophistication is required. The retailer's capital consists practically of his stock in trade and nothing else. The extent to which he can prudently supplement it by bank advances or credit from wholesalers is a matter constantly in his mind. When he does so supplement it, the charge for interest is by no means negligible in comparison with his income. He may, as you suggest, buy at regular intervals, or he may buy whenever his stocks need replenishment. But even if he buys at regular intervals, his calculation of what will be "sufficient to meet his customers' needs in the interval" [5] is subject to a wide margin of error. If, without expressly intending to shorten the interval, he keeps his estimate of requirements down to a minimum, he may have to make an intermediate purchase. And even if he does not cut down the amount of his periodical purchases he can keep down his overdraft by spreading out deliveries.

You "stressed the danger of holding redundant supplies," because you "imagined the inconveniences of going below normal were more obvious." But the point is that it is possible to make a large reduction in average stocks without making any reduction in the minimum, and with only a very slight sacrifice of convenience. Whenever the minimum is reached, a new consignment (large or small) is ordered.

With regard to Tinbergen's failure to find a correlation between the rate of interest and trade fluctuations, I do not know to what periods or countries he applied his analysis. Since rising prices cause a high short-term rate and the high short-term rate causes falling prices the absence of any correlation would not in itself be very surprising. But as a matter of experience the effect of rising prices (or expanding demand) in raising the rate and of falling prices (or contracting demand) in lowering the rate so greatly predominates that there is a correlation which requires no mathematical analysis to discover it. A comparison of the table of average rates of discount in Pigou's Industrial Fluctuations (p. 399) with that of unemployment (pp. 381-2) will readily satisfy you of that. [6]

2) Deepening of Capital. You describe the deepeners as "people who have thought out paying plans of reconstruction for which from time to time they need finance." [7] That is not the only form deepening takes. Often the plan is for producing and putting on the market some new form of labour-saving plant which can be acquired by those who use it without any reconstruction at all. Sometimes reconstruction is required, and in that case the labour-saving device can only be introduced in replacements and new installations. Sometimes not merely reconstruction is required but a complete scrapping of the plant formerly used.

In any case the rate of progress is conditioned by finance. You say that "the fact that there is a gap to be filled does not affect them (the deepeners) one tittle." But if, as I think you would agree, widening is the first charge on the resources of the investment market, then what is available for deepening [b] is the residue or "gap", supplemented by whatever temporary borrowing may be practicable.

I think you under-estimate the difficulty experienced by anyone who has a remunerative project ready for exploitation, in raising the necessary capital. Direct access to the market through a public issue is usually only available to an existing concern with a past record of profit-making, and such flotations are primarily for widening, though there is often an admixture of deepening. For a project aiming at the production of a new labour-saving contrivance the projector has to get in touch with some individual capitalist or a small group who can be got to take an interest in his proposal. They must be people engaged in some kind of business more or less related to it, otherwise they will not be capable of launching it or even of exercising a responsible judgment in regard to it. In order to finance him, they must draw upon their own surplus funds, that is to say, their investments outside their own business, plus what they think it prudent to raise by pledging their credit.

The resources available for new projects through these channels are limited. As soon as any project is adopted, it ties up the funds provided for it for some years. When it is firmly established as a going concern, there may be a public issue of shares, which puts the capitalists who sponsored it once again in a position to take up something else. The interval that must pass before this happens depends on the demand for the new labour-saving plant. That demand depends in turn on the facilities for raising capital available to those who are to use it, whether in the course of widening or renewing capital or independently.

These facilities will take the form partly of depreciation allowances and reinvested profits, partly of flotations in the investment market, partly of advances from banks, in anticipation of profits or of flotations. These resources are limited, and the margin available for deepening will depend on the extent to which they are being used up in widening. If there is any special pressure it will result in enlarged bank advances, but the extent to which the banks will allow this depends on the general credit situation. If they are unwilling to let credit expand, they may either restrict advance for extensions of plant or for carrying long-term securities or they may compensate them by diminishing their advances for other purposes.

An adverse capital market will delay the exploitation of deepening projects, and will defer the time at which the funds tied up in them can be released by public flotations. And even when such projects are ripe for public flotations, an adverse capital market may result in postponement.

Theoretically it is of course possible that, even though the resources available for exploitation are thus limited, still the deepening projects might be exhausted. That is to say, there would be no projects for further intensifying the capitalisation of economic activity, which would promise a sufficient margin of profit to attract the promoter. As this state of satiety approached, I should think that the promoters would come to be content with much narrower margins of profit, though there would always remain a substantial excess of the marginal yield of instruments over the long-term rate of interest. If the current needs of widening were insufficient to use up current savings, the available deepening projects would be completed out of the surplus and deepening would then be exhausted. If the annual surplus were 286 million units, this stage would be finished a little earlier than if it were 246 million units. But the difference is not very important (unless of course the extra time just allows for a series of new inventions sufficient to restart the deepening process).

You say that the fringe of unsatisfied issues "eventually must be used up." [8] I do not think that is necessarily so; the rate of progress of deepening might be so retarded by various causes that it would fail not merely to outstrip new invention but even to keep pace with it. But if the possibilities of deepening were exhausted, and current needs of widening were insufficient to absorb current savings, that would be a far more fundamental problem than the trade cycle.

In my letter of 24th April I pointed out (p. 6) [9] that, if your theory of the trade cycle were true, the first symptom of a depression would be a fall in the long-term rate of interest. To that you have made no answer.

3) Absorption of Cash. You say you would "only speak of an absorption of cash as a cause of trouble if it followed from a change on balance in the desire of people to hold cash." [10] Here you are confusing the absorption of cash with hoarding. The absorption of cash includes not only the accumulation of cash balances but also the repayment of bank advances. But that does not materially affect your point. Just as a man may hold more cash because he has more cash to hold, so he may repay more bank advances because he receives more cash. In either case trouble may result, and it is not clear to me why, if "a shift of cash from those who desire to use it to those who desire to hold it", or to pay off bank advances, causes a shortage of demand with all its consequences, you should be unwilling to say so. If you want to distinguish this case from that where the cause of trouble is a change in the desire of people to hold cash, why not call this a change in the desire of people to hold cash?

The absorption and release of cash are of fundamental importance because it is through them that there [c] arises an inequality of output and sales. If, in the notation used in Chapter VI of my "Capital and Employment", [11] there is to be any difference between , sales to final purchasers, and A, output, there must on balance be an absorption or release of cash on its way from the consumers who receive incomes to the amount of B (=  ) to the final stage of sale, . The absorption or release of cash may occur among consumers, or in the investment market, or in the banks, or in industry. Any theory which traces fluctuations to an excess or deficiency of demand in comparison with output requires the occurrence of an absorption or release of cash somewhere. Your theory depends on it occurring in the investment market, Keynes' on it occurring among consumers. I attach special importance to the absorption of cash by industry induced by a rise in the short-term rate of interest. But these are all particular applications of the same general principle, differing as to the motives at work, and the groups of people affected by them. If, in working out any theory of this type, the absorption or release of cash is supposed not to occur, the theory simply collapses.

Yours very sincerely

R G Hawtrey

  1. 1. This letter was accompanied by a note of congratulations on Harrod's marriage, which Hawtrey could not attend (24 December 1937, in HP IV-452).

    2. Refers to letter 700 , [jump to page] .

    3. The bank rate was reduced from 6 to 2 per cent between April and June 1932, and kept at a that level till the war.

    4. Letter 700 , [jump to page] .

    5. Letter 700 , [jump to page] .

    6. Tables XVII and I respectively in A. C. Pigou, Industrial Fluctuations (1929).

    7. Letter 700 , [jump to page] .

    8. Letter 700 , [jump to page] .

    9. Letter 662 , [jump to page] .

    10. Letter 700 , [jump to page] .

    11. Hawtrey, Capital and Employment (1937), p. 137.

    1. a. TLS with autograph corrections, nine pages on nine leaves, numbered from the second, in HP IV-451.

      b. Ts: «deepening,».

      c. Ts: «that arises».


Welcome page

top of page

Return to index of this section

Go to previous page

Go to next page