329. Harrod et al. to F. D. Roosevelt , 20 November 1933 [a] , [1]


[Answered by 341 ]

20 November 1933 [2]

Private and confidential.

To the President of the U.S.A.


We, the signatories of this letter, occupy official positions as teachers of economics, in the University of Oxford. We wish first to state our grounds for venturing to write to you.

We are all interested in the question of monetary policy and have, as a matter of professional duty, followed the events of the recent depression closely. We feel strongly that this problem is not to be solved by the method of waiting for it to solve itself. [3] We believe that a deliberate policy of expansion [4] is desirable, both to hasten the amelioration of present troubles and also to lay the foundations of policy designed to prevent the recurrence of similar troubles in future.

We have expressed in public our views regarding the application of these ideas to British policy. [5] We have throughout the depression remained convinced that counteracting measures of expansion should be adopted, on principles which have in our judgment remained fundamentally unchanged.

We believe that positive measures are even more important in the U.S.A. than here. We therefore welcomed with enthusiasm the indications which came to us that an expansionist policy was likely to be pursued and that a programme of Public Works was to play a prominent part in it. We welcomed with enthusiasm your statement to the World Economic Conference of July 1 which rendered a joint attempt to buttress up the old system out of the question. [6] We also welcomed with enthusiasm your statement of October 22 that a resolute attempt would be made to raise commodity prices in the U.S.A. to a specified level and thereafter to prevent fluctuations in the commodity value of the dollar. [7] We believe that the policy of which these statements contain the promise might if wisely pursued usher in a new era of prosperity, and that some such policy is a matter of urgency if western civilization is to be maintained. We think that this is a great opportunity for the U.S.A. to assume leadership in these matters.

It is our intention in the following paragraphs to outline as briefly as possible the means by which such a policy would in our judgment be rendered effective.

We [8] believe that inflation of a kind likely to increase output and raise prices [9] can only be produced by an increase of incomes which is not accompanied by an equivalent increase in the costs of producing the goods which these incomes are used to buy. [10] We lay stress on the increase of incomes by contrast with (1) an increase of money in circulation and (2) an increase in the liquidity of corporation assets.

We suggest four ways in which such an increase of incomes may be brought about.

1. An increased production of capital goods increases the incomes and expenditure of those engaged in their production without increasing the quantity of consumable goods on the market. It thus tends to raise the price level of those consumable goods.

The increased production of capital goods may be stimulated by low interest rates, especially a low long-term rate. [11] For this reason it is important that in their open market policy the Federal Reserve Banks should be encouraged to purchase long-dated securities. The stimulus given by a low rate constitutes an overwhelming argument in favour of an easy money policy.

Unfortunately in certain circumstances the stimulus may be ineffective. It will be ineffective if there is no confidence in the future course of prices and also if there is such an excess of productive capacity in the form of capital plant already existing, that the demand for new capital plant cannot be much expanded by a low interest rate. In the depth of a depression an easy money policy is not enough. It requires the assistance of other measures. [12]

2. We believe that the principal weapon in raising prices should be a great campaign of Public Works. [13] In addition to the direct employment given, Public Works also give employment in the first instance to the trades producing the necessary raw materials and capital equipment.

But they have a further repercussion, which is of the first importance in considering a policy of raising commodity prices. The wages, salaries etc., paid out to those engaged in the employment already mentioned will be largely expended on consumable goods. Thus the desired effect of a rise in the purchasing power devoted to consumable goods, unaccompanied by an increase in the gross costs of producing them is secured. Thus a force is set up tending to raise commodity prices and to expand the output of consumable goods. In our opinion this measure should take priority of all others and the whole strength and ingenuity of the Administration should be put behind an immediate attempt to give effect to it.

3. Another means of raising purchasing power without increasing costs, is to distribute incomes by way of unemployment relief and other social services with borrowed money. It is essential that these schemes should not be financed by taxation. If they are, the income that is put into one pocket is taken out of another, and there is no net increase of purchasing power nor tendency for prices to rise. We therefore recommend a great drive for relieving distress through the agencies of the federal, state or municipal administrations on borrowed money.

4. [14] All methods of encouraging private enterprise in constructional works should be explored. We may mention as instances (a) special assistance in railway construction (b) the guarantee of interest on new issues for approved purposes and (c) subsidies for these purposes.

5. We believe that it is important that the borrowings incurred for purpose 2, 3 and 4 should not be allowed to raise the long-term interest rate. It is desirable that the Federal Reserve Banks should make such purchases of long-dated [15] securities as are necessary to prevent rates rising and ensure that funds are available for government borrowing. This conforms to the principles laid down in our first recommendation.

We also feel it desirable to refer to certain expedients that cannot be relied on as the principal weapons in the specific task of restoring the volume of output and employment. We do not wish to affirm that there is nothing to be said in favour of these expedients or that they have no positive merits of their own.

1. We do not believe that relief should be sought in the first instance by measures designed to increase the quantity of money. [16] We do not believe that resort to the printing press in the literal sense would be of value. [17] Nor do we believe that there is much to be gained through the further injection of credit by open market operations of the Federal Reserve System on a large scale. This is not to deny that the purchases in the spring of 1932 were highly beneficial. [18] It is essential that money should be kept as easy as possible and the indebtedness of the member banks to the Federal Reserve Banks should be kept at a low level. If sufficient purchases are made to satisfy these two conditions, further additional purchases would not of themselves necessarily have an expansionary [19] effect.

It is to be observed, however, that both the issue of additional currency and further purchases by the Federal Reserve System may subsequently become necessary again, as ancillaries to the other measures of internal expansion [20] which were recommended above. Much use of the printing press does not, however, seem likely to be required.

2. We regard with profound respect your intentions in instituting the industrial codes. We appreciate that the new deal with labour was conceived as a great piece of humane and progressive statesmanship and that it may be regarded by future historians as an important landmark in American progress towards more rational relations between employers and employees, between the rich and poor. [21]

We are bound to say, however, while recognizing the great merits of this legislation so conceived, that it is not specifically conducive to quickening economic activity.

If a rise in money wage rates is initiated, it is essential that it should be accompanied by a sufficient rise of prices to cover the increase of costs consequential upon it. [22] But it must be noted that the rise of wage rates cannot be relied on to secure this, in the absence of other measures for increasing purchasing power.

Any labour measures tending to raise employers' costs give rise to similar difficulties. Here again their presence makes it all the more necessary to execute simultaneously other policies of expansion. [23]

3. We do not think that loans to corporations by such bodies as the Reconstruction Finance Corporation, [24] even if financed by "inflation money", necessarily tend to quicken activity. The test here is whether they are used to finance the output of new capital goods. If they are merely used to enable corporations to liquify their assets, they may have little or no effect in reviving activity. Such loans may often be justifiable. But there is danger that too much "inflationary money", the creation of which must after all be of a finite amount, may be used for this purpose, and too little left over for use in ways which would cause an actual expansion of activity.

4. We do not condemn schemes for regulating the output of raw products which have fallen to specially low values. But we point out that this cannot take us far in solving the general problem. [25] Such schemes may enable farmers to keep in line with the restrictive policy of manufacturers and so improve their relative position. But the main objective is a reversal of restriction all round. Moreover it is doubtful if much can be done, through restriction, to increase the purchasing power of farmers except with international co-operation. Farmers are, however, directly benefited in so far as restriction among manufacturers gives place to expansion, since the demand for their products is thereby improved.

5. We do not believe that the purchasing power of the dollar can be effectively reduced, by depreciating its value either in terms of other currencies or of gold.

It does not follow that dollar devaluation in terms of other currencies must be ruled out. On the contrary it may be desirable. But what is essential is that it should follow as a corollary of internal inflation and not be used as a lever for producing internal inflation. If used in the latter way it is likely to be ineffective in the U.S.A. and detrimental elsewhere. [26] But if the U.S.A. embark on measures of genuine internal inflation of the kind mentioned above, and if the other countries do not fall into line with parallel policies, then some external depreciation of the dollar is to be expected and welcomed.

Professor Irving Fisher many years ago propounded a scheme for stabilizing the commodity value of the dollar by varying its gold content. [27] This scheme has been held to be on right general lines in the past by many distinguished economists. [28] But we do not think it would be effective in present circumstances. Two conditions are necessary for its effectiveness. (i) Action must be taken as soon as any tendency to deflation (or inflation) appears, so that the forces of slump or boom may be counteracted before they have gathered strength. (ii) The rest of the world, or most of it, must be on the gold standard.

If the rest of the world is not on the gold standard, raising the dollar price of gold is not likely to have much effect on the general level of dollar prices.

If a few other countries are on the gold standard, raising the dollar price of gold is tantamount to depreciating the dollar in terms of the currencies of these countries. The consequence of this is that the prices of commodities of a stable and internationally tradable character must be adjusted accordingly. This adjustment may consist of a rise of American prices or a fall of gold prices or partly of both. If the gold countries are of less economic importance than the U.S.A., the fall in them is likely to be greater than the rise in the U.S.A. Thus the rise in the price of gold has but a weak effect in raising dollar prices and a stronger effect in reducing gold prices. It plunges the gold countries into greater difficulties and is likely, if pursued à outrance to drive them off the gold standard. If this result is achieved, the gold policy becomes quite ineffective in raising American prices.

The consideration advanced in the last paragraph has a broader application. Any policy of dollar exchange depreciation has a weakening effect on the outer world unless accompanied by a vigorous internal inflation of American purchasing power. Now it is essential from the world and the American points of view that any revival of activity in the U.S.A. should be allowed to have a stimulating effect on the rest of the world.

External depreciation unaccompanied by internal inflation, lowers the (foreign) prices at which Americans are prepared to offer their goods for export without directly increasing her demand for world goods. This must deepen foreign depression and may nullify, so far as the effect on America is concerned, the relative improvement of her position as an exporter directly caused by dollar devaluation.

Finally we should like to emphasize the positive policy outlines in paragraphs 2 and 3 on pages 2 and 3 [29] as a matter of urgency and as the only effective method of reducing the commodity value of the dollar [30] and increasing the level of activity in the country.

We have the honour to be

D H Macgregor, Drummond Professor of Political Economy.

L. M. Fraser, M.A., Fellow of Queen's Coll[ege].

W. M. Allen, M.A. Fellow of Balliol College.

E. L. Hargreaves, M.A. Fellow of Oriel College.

R. F. Bretherton, M.A. Fellow of Wadham College.

J. E. Meade M.A. Fellow of Hertford College

E G Dowdell, M.A. Lecturer of St. John's College

Redvers Opie M.A. Fellow of Magdalen College.

R. L. Hall M.A. Fellow of Trinity College

R. F. Harrod, M.A. Student of Christ Church, University Lecturer in Economics

  1. 1. Four versions of this letter survive. The oldest is an undated mimeographed Ts, six pages with a few corrections in Harrod's hand, which was probably circulated among the Oxford economics fellows: the copy filed in HCN 4.27.1, in fact, is signed by E. G. Dowdell, one of the signatories of the final version. The second version consists in very extensive amendments, deletions, additions and movements of text in Harrod's hand on the basis of version 1. This is filed in HCN 4.27.2; two additional leaves with additions to be inserted are filed in HCN 4.27.3. The third version is a mimeographed Ts of version two (six pages); a copy of it was found among Joan Robinson's papers, and is filed in JVR/vii/328/1-6. The fourth version, reproduced here, is the one actually sent to the president of the United Staes; it differs from the third only in a couple of passages. The significant changes are listed in the following notes to this letter.

    2. The letter was transmitted to the president by Felix Frankfurter, accompanied by a letter sent from Oxford and dated 23 November 1933. Frankfurter so explains the origin of this letter:

    • A leading Oxford don, speaking for an influential group of Oxford economists who are deeply sympathetic with your financial and economic objectives and look to you as a stimulus to action by the British Government, came to ask me whether it would be deemed impertinent for them to express their views to you in a private communication. I replied that one of your chief characteristics is accessibility of mind to suggestions, particularly from those in general agreement with your purposes. He then asked whether there were any likelihood of such a communication reaching you, and I volunteered to undertake its transmission.

      [...] I felt I ought to serve as the post-office, as it were, for this letter because the writers have been strong supporters of your policies, have refused to be worshippers of the golden calf theory of currency and have vainly struggled for action by their government along your expansionist lines. (Frankfurter to Franklin D. Roosevelt, 23 November 1933, in FDR PPF 1820, Box 1; Frankfurter's letter is printed in Roosevelt and Frankfurter. Their Correspondence 1928-1945, annotated by M. Freedman, London and Toronto: The Bodley Head, pp. 167-68.)

    3. This sentence substitutes "We all believe that laissez-faire is no solution of the economic problems which confront us" from the second version.

    4. The words "of expansion" substitute the words "of planned expansion or inflation" from the second version.

    5. See, in particular, the collective letters to The Times "Restoration of Prices" (July 1932: Harrod 1932:4 ) and "More Money in Circulation" (March 1933: Harrod 1933:1 ), and Harrod's contributions on "Monetary Policy" ( 1932:2 and 1932:3 ), on reflation ( 1932:7 ), on exchange policy ( 1933:4 and 1933:5 ) and on the Treasury's policy ( 1933:6 ).

    6. Roosevelt's "Wireless to the London Conference Insisting upon Larger Objectives Than Mere Currency Stabilization", delivered on 3 July 1933, is printed in The Public Papers and Addresses of Franklin D. Roosevelt, vol. 2: The Year of Crisis, New York: Random House, 1938, pp. 264-65.

    7. Roosevelt's commitment to raising the price level, in particular of agricultural products, was expressed in the fourth "Fireside Chat", 22 October 1933, and is reproduced as "We Are on Our Way, and We Are Headed in the Right Direction" in The Public Papers and Addresses of Franklin D. Roosevelt, vol. 2: The Year of Crisis, New York: Random House, 1938, pp. 420-27.

    8. In the first draft, the paragraphs including the present one down to the one numbered 5 below (that is, the "positive recommendations") were originally at the end of the letter, after the "list of negatives".

    9. From the second version, the words "increase output and raise prices" substitute the words "stimulate activity".

    10. In his letter to Meade of 7 November 1933 (cited in note 5 to letter 324 R) Kahn suggested to stress this point.

    11. From the second draft, the remainder of the paragraph substitutes the sentence "This constitutes an overwhelming argument in favour of an easy money policy."

    12. The last two sentences were added in the second version, while the following paragraphs were crossed out at this point:

    • We are thus in a vicious circle. Consumers' purchasing power can only be increased by an increased production of capital goods. But the tendency to produce new capital goods will be very weak until the consumers' demand for consumable goods is greatly expanded.

      Thus in the depth of a depression the easy money policy is not enough. This is not to say that it is unimportant. For as soon as a genuine revival has got going, the demand for new capital goods will become effective once more. This is likely to have much bigger effects in assisting the climb back to normality in its later stages than in its earlier stages. Indeed in the later stages it may do all the work without the assistance of other measures. But in the early stages such assistance is [b] indispensable.

    13. Stress on public works was suggested by Kahn in his letter to Meade of 7 November (the passage is cited in note 5 to letter 324 R).

    14. This section did not appear in version one.

    15. The words "long-dated" do not appear in the first three versions, and may have been inserted as a consequence of Keynes's comment, as reported by Kahn on 18 November 1933: letter 328 on [jump to page] .

    16. From the second version onwards, the words "of money" substitute the words "of actual money in circulation".

    17. This point was also emphasized by Kahn: see his letter to Meade of 7 November, cited in note 5 to letter 324 R.

    18. In February 1932 the Glass-Steagall Act enable the Federal Reserve to hold government securities attains Federal Reserve Notes and widened the circumstances under which the member banks could borrow from the Federal Reserve. The Fed immediately undertook expansive open market operations, and within six months it increased its holdings of government securities by $1.1 billion.

    19. In the first version, the word "inflationary" stood instead of "expansionary".

    20. In the second version, the word "expansion" substitutes "inflation".

    21. Labour-management relations policy was a major preoccupation of the New Deal administration. The National Recovery Act (1933) gave employees the right to bargain collectively without employers interference in any labour organization. Later labour relations were codified in the National Labour Relations Act of 1935 and the Fair Labour Standards Act of 1938, complemented by the Social Security Act of 1935.

    22. Kahn particularly insisted on this point in his letter to Meade of 7 November (see note 5 to letter 324 R).

    23. The last two paragraphs substitute the following passage in the first draft:

    • We do not believe that an increase of purchasing power (among wage earners) which entails a pari passu increase in the costs of production, has in and by itself any effect in stimulating recovery.

      It does not follow that we condemn this policy or would wish your work undone. We merely regard it as irrelevant to the objective about which we are writing this letter. It is necessary to particularize.

      A rise of money wage rates may do no harm if it is accompanied by a sufficient rise of prices to cover the increase of costs consequential upon it. Such a rise of prices may be made to occur, if simultaneously some other measures likely to cause a rise of prices can be put into effect. But it must be noted that the rise of wage rates cannot be relied on to secure, in the absence of these other measures, a sufficient rise of purchasing power (e.g. if the rise in wage rates led to a decrease of employment). And, if the rise of purchasing power does not proceed at the same time, the rise in costs may be positively harmful to recovery. The existence of the codes makes it all the more necessary therefore, to execute other measures simultaneously.

      Labour measures resulting in a rise in costs otherwise than through an increase of wages are likely to be injurious to recovery. For instance a shortening of the working day may make it difficult for an employer to handle his labour force as efficiently as before and so may cause his money costs to increase more than the money purchasing power of his employees increases. This is likely to have a restrictive effect. The labour measures may be highly desirable on general grounds and we do not recommend their reversal. Their injurious effect may be counter-weighed by the natural progress of industrial efficiency. But here again their presence may make it all the more necessary to execute other simultaneous measures for increasing purchasing power.

      We do not recommend the shortening of the working day, unless justified by its own merits on grounds of social welfare, as a remedy for unemployment. We think it likely to slow down and not to quicken economic activity.

    24. The Reconstruction Finance Corporation was created as part of a comprehensive legislative program enacted by the Congress on President's Hoover request at the end of 1931. It aimed at offering government credit to a variety of financial institutions in an attempt to liquify their assets. See, for instance, J. S. Olson, Herbert Hoover and the Reconstruction Finance Corporation, 1931-1933, Ames: The Iowa State University Press, 1977, in particular chapter 4.

    25. In the first version, the sentence concludes: "... a partial solution of the problem"; versions 2 and 3 substitute this expression by "... a temporary remedy of the problem".

    26. Kahn had insisted that the dangers of a low exchange rate policy should be stressed: see his letter to Meade of 7 November, cited in note 5 to letter 324 R.

    27. I. Fisher, The Purchasing Power of Money (1925 edition, chap. XIII, §5); Stabilizing the Dollar, New York: Macmillan, 1920.

    28. The words "many distinguished economists" substitute the words "many economists, including ourselves" in the first version.

    29. In the first three versions the sentence runs as follows: "... to use the whole weight of our authority in pressing the policy indicated in 2 and 3 upon you as a matter ...".

    30. The first three versions conclude at this point; the concluding remark might thus have been suggested by Kahn or Joan Robinson (see letter 326 R on [jump to page] ).

    1. a. TLS, seven pages on seven leaves, in FDR PPF 1820, Box 1. Details concerning the three other surviving drafts are given in note 1 to this letter. The letter is printed in Roosevelt and Frankfurter. Their Correspondence 1928-1945, annotated by M. Freedman, London and Toronto: The Bodley Head, pp. 168-73.

      b. Ts:«in».

Welcome page

top of page

Return to index of this section

Go to previous page

Go to next page