View PDF

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

CENTRAL BANK
COOPERATION
1924-31
STEPHEN V. O. CLARKE

FEDERAL RESERVE BANK OF NEW YORK




CENTRAL BANK COOPERATION:
1924-31

STEPHEN V. O. CLARKE

FEDERAL RESERVE BANK OF NEW YORK










To
John Henry Williams

Foreword

T

HIS monograph is another in the series published by the Federal Reserve
Bank of New York on monetary policy, institutions, and techniques. It
deals with a segment of history that has special fascination for students of monetary affairs: the efforts of American, British, French, and German central bankers
to reestablish and maintain international financial stability in 1924-31 and the
frustration of those efforts during the financial crisis at the end of that period.
This is a story whose broad outlines are familiar, but the author has used the
historical records of this Bank and unpublished papers of various prominent
Americans to bring new insight to his description of the dramatic events of those
years.
Mr. Clarke's monograph surveys central bank cooperation as it was practiced
in an era when the international economy differed in many respects from that
of today. In the twenties and early thirties, the world was still feeling the repercussions of the political and economic upheaval generated by World War I.
Reparations and war debts were souring relations among the great powers. The
United States, having rejected membership in the League of Nations, had retired
by and large into isolation. At the same time, the Western democracies were
experiencing a reaction against government management of their economies—of
which they had had their fill during the war—and a renewal of confidence in the
functioning of free market forces. As governments withdrew from economic
management, central banks were left to carry virtually the whole responsibility
of maintaining economic stability.
Although the contrasts are many, striking parallels with today are also to
be found. Mr. Clarke's account of the efforts of the authorities in the United
States and abroad to reconcile domestic and international policy objectives has
a familiar ring, as does his discussion of the complications that arose from conflicts in the foreign policy objectives of the great powers. This book also focuses
on the perennial problem of the policy maker who must be on the alert to adjust
economic policies and techniques to changing circumstances. In fact, the study
suggests that the failures of 1929-31 can be traced basically to the slowness of
the Western democracies to realize that the problems at hand were radically




(Please turn page.)

different and more complicated than those of the midtwenties and that their

handling thus demanded major new departures in economic policy and technique.
The problem of benefiting from the lessons of the past, while at the same time
discerning where previous events are no longer relevant, is as old as the study of
history. How far the experience of 1924-31 can be of help in dealing with
present-day international financial problems, and how far it is only a reminder
of turbulent days gone by—these are intriguing questions posed by Mr. Clarke's
study. At the very least, however, a review of the period provides a vivid
reminder of the pitfalls into which an earlier generation fell, and a warning to
governments and central banks today.

ALFRED HAYES
President

New York City
January 1967




Acknowledgments
Many persons have given invaluable assistance to the author in the preparation
of this monograph. Associates in this Bank, especially Francis H. Schott, have
supplied unfailing encouragement and constructive criticism throughout the
various stages of the study. The author has benefited from the comments of staff
members of the Board of Governors of the Federal Reserve System as well as
from the scholarly suggestions of Benjamin H. Beckhart of Columbia University,
Arthur I. Bloomfield of the University of Pennsylvania, Lester V. Chandler of
Princeton University, and Herman E. Kroos of New York University. Sir Theodore Gregory and Sir Otto Niemeyer made numerous valuable suggestions based
on their intimate knowledge of the events of 1924-31. The author is indebted to
those who arranged for him to use several important collections of private papers:
Dr. Everett Case who arranged for the use of the Owen D. Young Papers, located
in Young's former office in Van Hornesville, New York; Thomas S. Lamont for
the use of his father's papers, deposited in the Baker Library of the Graduate
School of Business Administration of Harvard University; and Richard H.
Logsdon, Director of Libraries of Columbia University for the Papers of George
L. Harrison. He owes an especially large debt to Mrs. Evelyn Knowlton, who
searched through not only these private collections but also this Bank's records
and the Library of Congress for the key historical materials on which this study
is so heavily based. Mrs. Knowlton also prepared the Dramatis Personae, the
Chronology of Important Events, 1919-31, and the index. The author wishes
to thank Mrs. Ann A. Goldenweiser for permission to quote from her late husband's papers, deposited in the Library of Congress, and the Macmillan Company
of London for permission to quote from Sir Henry Clay's Lord Norman. Valuable
assistance in the collection and checking of statistical materials was provided
by Miss Vincenzina Santoro and Mrs. Rona Lupkin. Many thanks are also due
to Miss Gloria Topper who designed the cover of the book, Miss Abigail M.
Cantwell and Miss Stella E. Walsh for their painstaking efforts in the final editing
and preparation of the manuscript for publication, to Mr. John H. Hendrickson
and Mr. Sigurds Vidzirkste for the preparation of the charts, and to Mrs. Marilyn
Daniels and Miss Geraldine Barry for patiently providing indispensable secretarial assistance.




Contents:
Dramatis Personae

Page
10

1- Introduction

15

2. The Problems of 1924-31

18

Economic Difficulties
Political Tensions
Temporary Solutions
Trouble Spots
The Economic Breakdown

18
21
22
23
24

3. The Role of Central Bank
Cooperation

Page
5. Britain's Return to Gold

71

Introduction
71
The Objectives of Anglo-American
Cooperation
72
The Strategy of Cooperation
75
Timing
78
Implementing the Strategy
81
The Federal Reserve and
Morgan credits
81
Coordination of monetary policy .. 85
Consolidating Sterling's Position
96
The Unfinished Task of International Economic Cooperation
105

27

Introduction
The Gold Standard Conception of
Central Bank Cooperation
Central Banks as Instruments of
National Policy
The Debate on Central Bank
Cooperation
The Genoa proposals
Strong's critique of the
Genoa proposals
A Pragmatic Approach to Central
Bank Cooperation

27
27
29
33
34
36
40

4. The Bankers' Role in the Stabilization of the German Mark: 1924
45

Introduction
Cooperative Aspects
Political and economic conditions
for the Dawes Loan
The transfer committee and
the agent general
Competitive Aspects:
Dollars or Sterling?
The Gold Discount Bank
The Reichsbank's international
reserves
Final Negotiations for the
Dawes Loan




45
48
50
56
58
58
60
67

6. The Defense of a Key Currency:
1926-28

Introduction
Payments Surpluses Complicate
Monetary Policy
Germany
France
The Shift toward Gold
Germany
France
British-French Accommodation
The Central Bankers' Meeting of
July 1927
The Easing of Federal Reserve
Policy
The Payments Balance Shifts in
Favor of Europe
The Liquidation of the Federal Reserve's Sterling Balance and the Shifting of the Demand for Gold to New
York
The International Financial System
Becomes More Rigid
Inadequate Efforts to Buttress
Sterling
Conclusion

108

108
109
109
110
112
112
115
119
123
124
127

130
134
139
141

Page
7. Mounting Strains
Declining Cooperation

Appendix
144

The Reparation Settlement
Strains During the Boom:
July 1928-October 1929
International imbalance
Difficulties of cooperation
The role of the Bank of France . . .
False Hopes: November 1929June 1930
Strains During the Depression:
July 1930-May 1931
,
Deterioration
Central bank cooperation: much
activity; little accomplished

173

8. The 1931 Crisis

182

Introduction
Germany at Loggerheads with
France
Central Bank Cooperation: Half
Measures to Support Austria and
Germany
Supporting the Austrian schilling..
The failure of efforts to support
the reichsmark
The Attack on Sterling
The first attack
Britain's response
Central bank credits to the
Bank of England
The bankers' loan to the
British government
The final onslaught
Epilogue

182

209
213
218

9. Conclusion

220




Page
222

145
147
147
150
164
168
171
171

183
185
186
189
201
202
203
204

Index

228

CHARTS AND TABLE

Chart 1: Foreign Exchange Rates in
New York, June 1914 and 1920-31...

19

Table: The Dawes Loan

70

Chart 2: Short-term Interest Rates:
New York and London, 1924-26 . . . .

87

Chart 3: Wholesale Price Indexes:
United States and United Kingdom,
1924-26

91

Chart 4: Arbitrage Opportunities
between New York and London,
1924-26

94

Chart 5: Arbitrage Opportunities between Paris and London, April 1927June 1928
122
Chart 6: Arbitrage
Opportunities
between New York and London,
January 1927-June 1928
129
Chart 7: Central Bank Discount Rates:
United States, United Kingdom, Germany, and France, December 1927April 1931
149
Chart 8: Central Bank Discount Rates:
United States, United Kingdom, Germany, and France, May-December
1931
190
Chart 9: Supporting Sterling in Crisis:
A Partial View, July 13- September 21,
1931
208

Dramatis Personae*
Balfour, Lord Arthur James, British Prime Minister, 1902-06; Foreign Secretary, 1916-19;
Lord President of the Council, 1919-22, 1925-29; British Representative, League of
Nations, 1920; Chief British Delegate, Disarmament Conference in Washington, 1921-22.
Bradbury, Sir John (later Lord), with British Treasury, 1913-19; British Delegate, Reparation Commission, 1919-25.
Bruning, Heinrich, German Chancellor, 1930-32.
Case, James Herbert, Deputy Governor, 1917-30, Chairman and Federal Reserve Agent,
1930-36, Federal Reserve Bank of New York.
Churchill, Winston (later British Prime Minister), First Lord of the Admiralty, 1911-15;
Chancellor of the Exchequer, 1924-29.
Clementel, Etienne, French Minister of Finance, 1914, 1924-25.
Coolidge, Calvin, Vice President of the United States, 1921-23; President of the United
States, 1923-29.
Crane, Jay E., Manager, Foreign Department, 1919-27, Assistant Deputy Governor, 1928-29,
Deputy Governor, 1929-35, Federal Reserve Bank of New York.
Cunliffe, Lord Walter, Director, 1895-1920, Deputy Governor, 1911-13, and Governor, 191318, Bank of England.
Dawes, Charles G., President, 1902-21, and Chairman, 1921-25, Central Trust Co. of Illinois;
American Member and Chairman, First Committee of Experts on Reparations, 1924;
Vice President of the United States, 1925-29; American Ambassador to Great Britain,
1929-32.
Doumergue, Gaston, French Premier, 1913-14, 1934; President of the Republic of France,
1924-31.
Gilbert, S. Parker, Assistant Secretary of the Treasury of the United States, 1920-21; Under
Secretary of the Treasury, 1921-23; Partner, Cravath, Henderson & deGersdorff, 192324; Agent General for Reparation Payments, 1924-30; Partner, J. P. Morgan & Co.,
1931-38.
Goldenweiser, E. A., Assistant Statistician, 1919-24, Assistant Director, 1925, and Director,
Division of Research and Statistics, 1926-45, Federal Reserve Board.
Hamlin, Charles S., Governor, 1914-16, and Member, 1914-36, Federal Reserve Board.
Harjes, Henry Herman, American partner in the French banking house, Morgan, Harjes &
Co., 1898-1926.

* This list, which is provided for reference purposes, includes only the important posts of the more
prominent figures mentioned in this monograph.

10




Harrison, George L., Assistant General Counsel, 1914-18, and Counsel, 1919-20, Federal
Reserve Board; Deputy Governor, 1920-28, Governor, 1928-36, and President, 1936-40,
Federal Reserve Bank of New York.
Harvey, Sir Ernest M., Comptroller, 1925-28, Director, 1928-29, and Deputy Governor,
1929-36, Bank of England.
Henderson, Arthur, British Secretary of State for Foreign Affairs, 1929-31.
Herriot, Edouard, French Premier, 1924-25, 1926, 1932.
Hoover, Herbert, Director, United States relief program in Europe, 1914-19; Secretary of
Commerce of the United States, 1921-28; President of the United States, 1929-33.
Hughes, Charles E., Associate Justice, 1910-16, and Chief Justice, 1930-41, United States
Supreme Court; Secretary of State of the United States, 1921-25; Partner, Hughes,
Rounds, Schurman & Dwight, 1917-21, 1925-30.
Jay, Pierre, Chairman and Federal Reserve Agent, Federal Reserve Bank of New York,
1914-26; American Member, Transfer Committee, and Deputy Agent General for
Reparation Payments, 1927-30.
Kemmerer, Edwin W., Professor, Princeton University, Princeton, New Jersey, 1912-43;
Financial Adviser to many governments between 1917 and 1934.
Keynes, John Maynard (later Lord), Fellow, King's College, Cambridge University, Cambridge, England, 1910-46; Editor, Economic Journal, 1912-46; Secretary, Royal Economic Society, 1913-46; Author, The Economic Consequences of the Peace (1919), The
Economic Consequences of Mr. Churchill (1925), and The General Theory of Employment, Interest, and Money (1936); Member of the Macmillan Committee, 1929-31.
Kindersley, Sir Robert (later Lord), Director, Bank of England, 1914-46; Governor,
Hudson's Bay Co., 1916-25; Chairman and Managing Director, Lazard Brothers & Co.,
Ltd.; British Member, First Committee of Experts on Reparations, 1924.
Lacour-Gayet, Robert, Financial Attache at the French Embassy in Washington, 1924-30;
Director of Economic Research, Bank of France, 1930-36.
Lamont, Thomas W., Partner, J. P. Morgan & Co., 1911-40; Director, 1940-48, and Chairman, 1943-48, J. P. Morgan & Co., Inc.; Alternate American Member, Committee of
Experts on Reparations, 1929.
Laval, Pierre, French Premier, 1931-32, 1935-36.
Logan, James A., Unofficial American Delegate, Reparation Commission, 1923-25.
Lubbock, Sir Cecil, Director, 1909-42, and Deputy Governor, 1923-25, 1927-29, Bank of
England.
Luther, Hans, German Minister of Finance, 1923-25; German Chancellor, 1925-26; President, Reichsbank, 1930-33; German Ambassador to the United States, 1933-37.
MacDonald, J. Ramsey, British Prime Minister, 1924, 1929-35.
Macmillan, Lord Hugh Pattison, Chairman, Committee on Finance and Industry, 1929-31.




11

May, Sir George (later Lord), Secretary, Prudential Assurance Co., 1915-31; Chairman,
Committee on National Expenditure appointed by the British Prime Minister, 1931.
McDougal, James B., Governor, Federal Reserve Bank of Chicago, 1914-34.
McGarrah, Gates W., American Member, General Council of the Reichsbank, 1924-27;
Chairman and Federal Reserve Agent, Federal Reserve Bank of New York, 1927-30;
President, Bank for International Settlements, 1930-33.
McKenna, Reginald, British Chancellor of the Exchequer, 1915-16; Chairman, Midland
Bank, 1919-43; British Member and Chairman, Second Committee of Experts on
Reparations, 1924.
Mellon, Andrew W., Secretary of the Treasury of the United States, 1921-32.
Meyer, Eugene, Member, Federal Farm Loan Board, 1927-29; Governor, Federal Reserve
Board, 1930-33.
Mills, Ogden L., Under Secretary of the Treasury of the United States, 1927-32; Secretary
of the Treasury of the United States, 1932-33.
Moreau, Emile, Director General, Banque de l'Algerie, 1906-26; Governor, Bank of France,
1926-30; President, Banque de Paris et des Pays Bas, 1930; French Member, Committee
of Experts on Reparations, 1929.
Moret, Clement, with French Ministry of Finance, 1908-28; Deputy Governor, 1928-30,
and Governor, 1930-35, Bank of France.
Morgan, J. P., Partner, 1891-1940, and Senior Partner, 1913-40, J. P. Morgan & Co.; Chairman, J. P. Morgan & Co., Inc., 1940-43; American Member, Committee of Experts
on Reparations, 1929.
Morrow, Dwight W., Partner, J. P. Morgan & Co., 1914-27; American Ambassador to
Mexico, 1927-30; Member, United States Senate, 1930-31.
Niemeyer, Sir Otto E., with British Treasury, 1906-27; Comptroller, 1927, Director, 1938-52,
Bank of England; British Member, Financial Committee, League of Nations, 1922-37.
Norman, Montagu C. (later Lord), Director, 1907-44, Deputy Governor, 1918-20, and Governor, 1920-44, Bank of England.
Platt, Edmund, Member, Federal Reserve Board, 1920-30.
Poincare, Raymond, President of the Republic of France, 1913-20; French Premier,
1912, 1922-24, 1926-29.
Quesnay, Pierre, General Manager, Bank of France, 1926-30; General Manager, Bank for
International Settlements, 1930-37.
Rist, Charles A., Professor, University of Paris, 1914-26; Deputy Governor, Bank of France,
1926-29; Financial Counsellor, National Bank of Rumania, 1929.
Sackett, Frederic M., United States Senator, 1925-30; American Ambassador to Germany,
1930-33.

12




Salter, Sir Arthur (later Lord), General Secretary, Reparation Commission, 1920-22;
Director, Economic and Financial Organization, League of Nations, 1922-30.
Schacht, Hjalmar H. G., President, Reichsbank, 1923-30, 1933-39; German Minister of
Economics, 1934-37; German Member, Committee of Experts on Reparations, 1929.
Siepmann, Harry A., Assistant to Finance Member, Executive Council of the Governor General of India, 1922-26; Head, Central Banking Section, 1926-36, and Director, 1945-54,
Bank of England.
Snowden, Philip (later Lord), British Chancellor of the Exchequer, 1924, 1929-31; Lord
Privy Seal, 1931-32.
Sprague, Oliver M. W., Professor, Harvard University, Cambridge, Massachusetts, 1913-41;
Economic Adviser, Bank of England, 1930-33.
Sproul, Allan, Assistant Federal Reserve Agent and Secretary, 1924-30, Federal Reserve
Bank of San Francisco; Assistant Deputy Governor and Secretary, 1930-34, Assistant
to Governor and Secretary, 1934-36, Deputy Governor, 1936, First Vice President,
1936-40, and President, 1941-56, Federal Reserve Bank of New York.
Stewart, Walter W., Director, Division of Research and Statistics, Federal Reserve Board,
1922-25; Vice President, 1926-27, Chairman, 1930-37, and President, 1937-38, Case,
Pomeroy & Co.; Economic Adviser, Bank of England, 1928-30.
Strong, Benjamin, President, Bankers Trust Co., 1914; Governor, Federal Reserve Bank of
New York, 1914-28.
Swope, Gerard, President, General Electric Co., 1922-39.
Theunis, Georges, Belgian Premier, 1921-25, 1934-35; Governor, National Bank of Belgium, 1942-44.
Vissering, Gerard, President, De Nederlandsche Bank, 1912-31.
Warburg, Paul M., Member, Federal Reserve Board, 1914-18; Chairman, International
Acceptance Bank, 1921-32.
Young, Owen D., Chairman, General Electric Co., 1922-39; Chairman, Radio Corporation
of America, 1919-29; American Member, First Committee of Experts on Reparations,
1924; American Member and Chairman, Committee of Experts on Reparations, 1929;
Interim Agent General for Reparation Payments, 1924; Director, Federal Reserve Bank
of New York, 1923-40.
Young, Roy A., Governor, Federal Reserve Bank of Minneapolis, 1919-27; Governor,
Federal Reserve Board, 1927-30; Governor, 1930-36, and President, 1936-42, Federal
Reserve Bank of Boston.




13




Central Bank Cooperation:
1924-31

1. Introduction
Like many historical studies, this monograph is the result of an attempt to gain
insight into present-day problems by exploring those of the past. The study was
initiated in a period when the authorities of the North Atlantic countries were
cooperating closely to handle the problems of the international financial system,
and was written in order to fill the need to understand the accomplishments and
shortcomings of central bank cooperation from 1924 to 1931, the only earlier
period during which this type of cooperation was undertaken on a significant
scale.
Heretofore no detailed study of central bank cooperation during the 1920's
and early 1930's has been published although several authors have discussed
various aspects of the subject, sometimes at considerable length, in the course
of biographical and other more general economic works. This lack was due
partly to the absence, until recent years, of any pressing interest in the subject
and partly to the inadequacy of accessible information on the relations between
central banks during the period. However, the revived interest in central bank
cooperation has been accompanied by a great enlargement in the availability
of relevant historical materials. A major portion of the diary of Emile Moreau,
Governor of the Bank of France, 1926-30, was published in 1954.1 Thereafter

1
Emile Moreau, Souvenirs d'un Gouverneur de la Banque de France (Paris: Librairie de Medicis,
1954), hereafter referred to as Moreau Diary.




15

autobiographies of Hjalmar Schacht2 and Hans Luther,3 who presided over the
Reichsbank, 1923-39, were published. Equally important, official archives and
the letters and papers of bankers and statesmen have become accessible. Of
particular interest for this study are the papers of Benjamin Strong and George
L. Harrison, who headed the Federal Reserve Bank of New York from 1914
through 1940, of Charles S. Hamlin, a member of the Federal Reserve Board,
1914-36, and of E. A. Goldenweiser, director of the Board's research staff,
1926-45, and the minutes and other records of the Open Market Investment Committee of the Federal Reserve System. In addition, valuable insights into the
negotiations to stabilize the German currency in 1924 and to settle the reparation
problem are provided by the papers of Thomas W. Lamont of J. P. Morgan
Co. and of Owen D. Young, who among numerous other positions served for
many years as a director of the Federal Reserve Bank of New York.
Scholars have already begun to exploit this wealth of material. Included among
the recently published works are excellent biographies of Montagu Norman,
Governor of the Bank of England, 1920-44,4 and of Benjamin Strong,5 a monumental monetary history of the United States,6 and a fascinating history of
Germany's role in the 1931 financial crisis.7 Close study of these recent publications as well as of the abundant literature published earlier on the more general
economic and political problems of the interwar years has of course been indispensable in the writing of the present monograph. One important source of
information has not been explored: the archives of foreign central banks. The
monograph thus tells the story of central bank cooperation on the basis of the
voluminous but nevertheless incomplete historical materials available in the
United States.
The study is less than comprehensive in another sense too. It focuses primarily

2

Hjalmar Schacht, My First Seventy-Six Years (London: Allan Wingate, 1955).

3

Hans Luther, Vor dem Abgrund, 1930-1933 (Berlin: Propylaen Verlag, 1964).

4

Sir Henry Clay, Lord Norman (London: Macmillan & Co., Ltd., 1957).

5

L. V. Chandler, Benjamin Strong, Central Banker (Washington, D. C: The Brookings Institution,
1958).
6

Milton Friedman and Anna J. Schwartz, A Monetary History of the United States, 1867-1960
(Princeton: Princeton University Press, 1963).

7
Edward W. Bennett, Germany and the Diplomacy of the Financial Crisis, 1931 (Cambridge:
Harvard University Press, 1962).

16




on those developments that are of central importance in the stabilization of the
exchange rate structure during the mid-1920's and in the efforts to maintain that
structure until the breakdown in September 1931. In these developments the
central banks of Britain, France, Germany, and the United States played the
major roles. The monetary authorities of these countries did not, of course,
discuss their problems only among themselves. There was no exclusive club of
the big four. The leaders of each of the major central banks corresponded and
exchanged visits with a wide circle of other central bankers. The central bankers
of most European and some other overseas countries participated at one time
or other during 1924-31 in the granting of stabilization or other credits or were
themselves the recipients of such credits. The cooperative activities of the smaller
central banks certainly contributed significantly to the degree of international
monetary stability that was achieved. Yet any one or several of these cooperative
efforts could have failed—as at least one did—without threatening the international financial edifice.
This was not true of the cooperative efforts that involved the major powers.
The main pillars of the 1924-31 monetary order were the stabilizations of the
currencies of Germany in 1923-24, Britain in 1925, and France in 1926-28.
These pillars were based in turn upon the fixed gold parity of the dollar,
unchanged from the figure at which specie payments had been resumed in 1879.
So long as the stability of the major currencies' exchange rates was maintained,
the essential features of the international monetary order were preserved. But,
when the pressures loosed in the 1931 crisis overwhelmed both national and
international cooperative efforts and forced, first, a resort to exchange controls
to protect the reichsmark and, shortly thereafter, the depreciation of sterling,
the international monetary system of the 1920's was shattered beyond repair.
The vital role of the major currencies was well understood by the leaders
of the American, British, French, and German central banks. Events so moved
as to bring these central bankers together, sometimes face to face, more
often in correspondence by letter and cable or on the telephone, in repeated
attempts to resolve international monetary problems in which some or all of
them were vitally concerned. It is primarily the story of their efforts to establish
and then maintain the stability of the international monetary system that this
monograph aims to tell.




17

2. The Problems of 1924-31
In the history of 1924-31, the relations among the central banks of Britain,
France, Germany, and the United States form a relatively small but significant
part. To a certain extent the central banks themselves influenced the environment in which they functioned, and part of the story of what they did will be
related in subsequent chapters. But to a very large extent the central banks'
policies and their relations with each other were determined by international
political and economic developments over which they had little control. A brief
outline of these developments is set forth in this chapter.

ECONOMIC DIFFICULTIES

The world of the 1920's had changed in many of its fundamentals from the one
familiar to the pre-1914 generation. That generation, to which the heads of the
major central banks owed their basic outlooks, had known a world economy that
was unified to a unique degree. For the most part, exchange rates were stable
and the major currencies convertible into gold. Movements of merchandise and
capital were on a multilateral basis and were impeded to only a relatively minor
extent by tariffs and other barriers. At the center of this system was Britain, still
a leading industrial power. London was the world's major financial center, and
sterling the world's currency.
To this orderly, integrated, and expanding system, the world of the 1920's
bore only a superficial resemblance. In the early 1920's indeed any resemblance
was hard to find. War and revolutions had weakened both the economies and
currencies of Europe. The Austro-Hungarian Empire had been dismembered.
Much of continental Western Europe was devastated. The French franc was
quoted in the exchange markets at a fraction of its prewar value, the German
currency was entirely destroyed, and even sterling, whose stability in terms of
gold had been the keystone of the prewar exchange rate structure, was fluctuating
at substantial discounts from the 1914 parity (see Chart 1). Of the major currencies, only the dollar remained at the same gold value as before the war.
These monetary changes reflected deeper economic ones. The war had led to
major shifts in economic power and competitive strength. British industry, whose
primacy had been surpassed even before the war by the United States and
Germany, lost further ground during the 1914-18 hostilities as well as in
the decade that followed. What had been British overseas markets were now
18




CHART 1

FOREIGN EXCHANGE RATES IN NEW YORK
JUNE 1914 AND 1920-31

Note: Monthly averages of daily market buying rates in New York at noon
for cable transfers.
*The monthly average of daily rates for the mark dropped below 0.6 cents
in November 1921. A year later it was 0.0147 cents. In the fall of 1923, the
rate was stabilized at 1 trillion marks = 23.8 cents. In October 1924, the
old currency was replaced by the reichsmark whose parity in terms of
gold was established at the equivalent of 23.82 cents.
Source: Board of Governors of the Federal Reserve System, Banking and
Monetary Statistics (Washington, D. C, November 1943).

increasingly supplied by newly established local industries or by industrial
competitors abroad. Among the latter, the foremost was the United States,
whose international competitive strength had been greatly increased through the
development of new products and highly efficient productive techniques. Growing
rapidly, the United States had become preponderant in the world economy.




19

It accounted for almost half of world industrial production in 1925-29; its
imports of the nine principal raw materials and foodstuffs in 1927 and 1928
accounted for 39 per cent of total imports of such commodities into fifteen
major trading countries. As a source of capital, its weight was even greater—
foreign loans floated in the United States during the decade ended in 1929
exceeded the total of similar loans floated in all the other capital lending
countries put together.1 Even in the second rank, Britain experienced sharp
challenges from Japan and, once their currencies were stabilized, also from Germany and France. But Britain—its youth decimated by the war, its business
leadership complacent, its industry facing stagnant domestic markets and encumbered by heavy taxation, and its labor force often recalcitrant—responded to
these challenges only slowly.
Britain's adjustment difficulties were associated not only with rigidities in its
internal economic structure but also with the structure of exchange rates adopted
by the European countries during the mid-1920's. In Britain, the conviction
was widespread that the reestablishment of sterling at its prewar gold parity was
a prerequisite for the restoration of London's position as the major international
financial center. On the Continent, in contrast, there was less preoccupation with
financial status and more with the need to maintain or strengthen the international
competitive position of trade and industry. In many Continental countries, moreover, prices had risen so much that it was unrealistic to try to force them back
to their prewar levels. Therefore, while Britain accepted strict financial disciplines
in order to restore and then maintain sterling's 1914 gold parity, the Continental
countries stabilized their currencies at exchange rates that represented heavy
discounts from their prewar values.
The international economy was fundamentally changed not only by the relative
decline in Britain's competitive strength but also by the development out of the
war of a burdensome structure of reparations and intergovernmental indebtedness.
Efforts led by the British to cancel these obligations came to naught on the
insistence of the United States Government that the European Allies honor their
contractual obligations, and the demands of the devastated Allied countries that
Germany pay an amount sufficient not only to offset Allied payments to the United
States but also to finance at least part of the cost of reconstruction. Accordingly,

1
United States Department of Commerce, The United States in the World Economy (Washington,
D. C: United States Government Printing Office, 1943), pages 29-31.

20




Germany made reparation payments to the European Allies equal to about 12
per cent of its total exports in 1925-30, while the United States obtained receipts
on account of war debts equal to 5 per cent of its total imports, the principal
effects being to weaken the international position of Germany and to strengthen
the already powerful position of the United States.

POLITICAL TENSIONS

These major changes in the international economy were not necessarily fatal.
In more propitious circumstances, any one or all of them could have been
handled through some combination of economic adjustment and cooperation
among the major powers. In the 1920's, however, the basis for such a solution
was weakened by sharp conflicts in the great powers' national aims and
also by the bitterness and mistrust that arose among them as the result of the
war. To these political difficulties, reparations and war debts of course contributed. A large segment of the German population felt that reparations and
the apparatus of foreign control erected to ensure their payment were not only
economically but morally wrong. Even those Germans who worked for the
fulfillment of the reparation obligations did so only in the belief that this would
hasten the day when the Allies would give up their claims. Across the Atlantic,
Americans resented references in Allied countries to "Uncle Shylock", while the
Allies, bled white by the war, grew increasingly bitter as they felt themselves
squeezed between a United States that insisted self-righteously that loans contracted be repaid and a Germany that was determined to escape the reparation
burden.
At least as important as the problem of intergovernmental debts in its detrimental effect on international cooperation was the conflict between Britain and
France over security against Germany. For France, security meant a Germany
confronted on all its frontiers by the allies of France—a Germany economically
enfeebled and thus powerless to wage another war. Such a policy was unacceptable to the British government, because it would help perpetuate the dominance
of France on the Continent and would hinder the reconstruction of trade between
Britain and the markets of Central and Eastern Europe. Lacking Britain's support,
France was not in a position to enforce its policy on Germany but had to accept
compromises which, to the extent that they fostered the recovery of Germany's
political and economic strength, were regarded by many French leaders as involving the sacrifice of French security. Moreover, French fears were only increased




121

as it became clear that the German government had a long list of aims, beginning
with the evacuation of the Ruhr and including among other things the dismantling
of the Allied control apparatus, the evacuation of the Rhineland, revision of the
frontier with Poland, and a customs union with Austria—for the attainment of
which it was prepared to press when it felt that the time was ripe. For its part,
the United States, disillusioned by Europe's quarrels, eager for "normalcy", and
the enjoyment of its rising prosperity, retired by and large into isolation. In these
circumstances, there were large obstacles indeed to great power cooperation to
deal with the grave economic difficulties of the 1920's.

TEMPORARY SOLUTIONS

Despite these many and fundamental difficulties, it was possible to believe, once
the worst of the postwar political upheavals in Central and Eastern Europe had
passed, that a stable international monetary system could be reestablished. Such
a belief was nurtured in 1924 by the stabilization of the German currency, the
tentative settlement of the reparation problem, the successful flotation of the
Dawes Loan, and the agreement of France and Belgium to evacuate the Ruhr,
and in 1925 by the stabilization of sterling and six other currencies in terms of
gold. Moreover, as Europe's political quarrels temporarily abated and the venerated symbols of normalcy and stability were reestablished, investors began to
look forward to a new era of international economic growth and prosperity.
The outlook appeared particularly bright in the primary producing countries and
also on the continent of Europe, especially in Germany, where the demand for
capital was huge and prospective rates of return on investment were far above
what might be expected in the relatively depressed British economy or even in
the United States whose economy in 1925-27 was still expanding relatively slowly.
The American economy responded vigorously to the hopeful outlook abroad.
A rising outflow of capital swung the United States balance of payments from
a surplus of almost $270 million in 1923 into a deficit of more than $1 billion
in 1927.2 The movement of gold to the United States, which had been a dis-

2

Throughout this monograph, a surplus or deficit in the United States balance of payments is defined
as equivalent to the total of (a) the changes in the United States gold stock and (b) the change in
foreign short-term dollar claims. The data used are from the United States Department of Commerce,
Historical Statistics of the United States (Washington, D. C: United States Government Printing
Office, 1960), pages 562-64.

22




turbing feature of the earlier postwar years, was reversed. In mid-1928 the
United States gold stock was almost $400 million lower than it had been four
years earlier. Moreover, after sterling's return to gold, Britain too resumed its
traditional role as a lender of long-term funds.
These mutually reinforcing political and economic developments had striking
effects. The economies of Europe were rapidly restored: primary commodity
output expanded, and there was a strong revival of international trade.
Under the exhilarating influence of market incentives, some essential adjustments
were taking place, notably in Germany whose export volume almost doubled in
the six years ended in 1929.3

TROUBLE SPOTS

While the world economy appeared sound as long as the large outflows of United
States and British capital continued, new trouble spots that were to complicate
the underlying difficulties began to develop. A significant proportion of the inflow
of foreign capital to Germany financed local government projects that could not
be expected appreciably to increase the country's export capacity. At the same
time, there were signs that foreign investment in certain primary producing
countries was being overextended, with the consequence that many important
food and raw material prices declined during 1926-29. Financial difficulties also
appeared. Short-term inflows formed a substantial part of the foreign capital
invested in Germany; they also, in effect, financed much of the long-term capital
outflow from Britain. In both these countries, short-term liabilities to foreigners
at the end of the 1920's exceeded by several times the gold and foreign exchange
reserves of their central banks. Above all, trouble was brewing in the United
States where the final stage of the long economic upswing of 1921-29 was accompanied by a variety of speculative excesses, including a stock market boom that
saw call money rates in New York averaging 6 to 10 per cent in each of the
twelve months ended in October 1929 and occasionally drove rates for overnight
money as high as 20 per cent.
The New York stock market boom was only the first of a series of misfortunes
that was to lay bare the faulty foundations of the economic structure that had

3

Angus Maddison, "Growth and Fluctuation in the World Economy, 1870-1960", Quarterly Review
(Banca Nazionale del Lavoro, June 1962), page 186.




23

been so hopefully erected in the mid 1920's. The lure of speculative gains and
the high level of interest rates in New York led, not only to a sharp decline in
new foreign issues in the United States during the fifteen months ended in
October 1929, but also to some flow of foreign balances to America. Central
banks abroad countered these flows by tightening credit in their own markets
but, despite these defensive measures, foreign countries lost almost $240 million
in gold to the United States in the fifteen months ended in October 1929, a
sharp reversal from their gold gains of the four years ended in June 1928.

THE ECONOMIC BREAKDOWN

After the New York stock market crash at the end of October 1929, the strains
in the world economy intensified. The difficulties resulted from the interaction
of many elements of which four major ones will be briefly discussed: (1) the
"real" contraction, i.e., the decline in production, trade, and employment;
(2) the undermining of the financial positions of individuals, firms, and governments; (3) the renewed disequilibrium in the world economy; and (4) the
search for liquid and stable assets that increasingly developed as confidence in
banks and currencies declined. The interaction of these elements was not of
course in chronological, or any other, order but may best be conceived as part
of a deflationary spiral, in each twist of which all four elements operated as
both cause and effect.
The contraction of production and trade in the two years ended in June 1931
was the sharpest on record. Industrial production in some countries, notably
France, was fairly well maintained, but it declined in Britain by onefifthand
in Germany and the United States by one third. In the three latter countries,
one out of every four or five members of the labor force was unemployed in
mid-1931. With the contraction in incomes and production came a sharp drop
in the demand for food and raw materials, which aggravated the preexisting
decline in commodity prices. By mid-1931 many of these had fallen to three fifths
or less of their June 1929 levels. With both production and prices declining,
international trade also contracted sharply, the contraction being aggravated by
self-defeating governmental attempts to protect domestic markets through the
erection of tariff and other barriers to trade.
The contraction of economic activity had serious repercussions on the financial
positions of individuals, firms, and governments. Individuals who had contracted
mortgages and borrowed against securities, firms with bank loans and long-term
24




market obligations, and governments with fixed charges including domestic and
foreign debt—all were confronted with serious declines in incomes or revenues.
Moreover, as the economic outlook darkened, private borrowers found it increasingly difficult to obtain additional credit to cover the gap between income and
expenditure or even to refinance outstanding debt. Banks and capital markets
became highly selective in their lending. Retrenchment consequently became a
byword: individuals pruned consumption, firms laid off workers and reduced
investment outlays, creditors decreased their outstanding loans, and governments
attempted to cut expenditures in the futile hope of balancing their budgets.
But these retrenchment measures only aggravated the economic contraction
and in the end worsened the financial problems that they were designed to
alleviate. Consequently, financial obligations of all types went into default, from
home mortgages to government long-term obligations to foreigners, which led
in turn to a further freezing-up in borrowing facilities. Indeed, during the year
ended in June 1931 the markets for new foreign issues in both London and
New York became almost dead. For the primary producing countries whose
borrowing during the 1920's had saddled them with large fixed service charges
and for Germany which in effect had met its reparation obligation by borrowing
abroad, the difficulties were especially severe.
These problems were compounded by a search for liquidity that became
increasingly intense. The gyrations of exchange rates in the early 1920's and
the destruction of the German currency were remembered all too well. As the
depression deepened, fears of yet another period of currency chaos increased.
Confidence was shaken by currency devaluations in primary producing countries.
In the industrial countries the financial difficulties of individuals and businesses
resulted in a rising number of bank failures, particularly in the United States.
These failures in turn led to a disturbing increase in the hoarding of gold and
currency that reflected a growing uneasiness about the soundness of the banking
system. In the international sphere, the increased uncertainty and the search
for liquidity led to disequilibrating movements of balances from countries whose
currencies were suspect to centers such as New York and Paris whose positions
were regarded as impregnable.
The situation in the early 1930's was thus the reverse of that in the mid-1920's.
Just as the expansive optimism of the earlier period had led to a great outflow
of American capital, so the deflation and pessimism of the early 1930's slowed
down and finally reversed the outward capital flow. The disequilibrium between
the United States and the rest of the world that had been bared initially by the




25

1928-29 boom was now aggravated by the ensuing depression, the consequence
being that in the twenty months ended in June 1931 the United States drew an
additional $514 million in gold from abroad, largely from primary producing
countries. At the same time, capital moved in substantial volume to France,
initially for investment in its relatively prosperous economy and subsequently
for security. The resulting gold movement to Paris came largely from London
and was even greater than that to the United States.
The world economy was thus badly out of balance and one of its major
financial centers, London, was enfeebled, when a severe liquidity crisis struck
at the weak spot in the network of international financial claims. As already
noted, the revival of Central Europe in the mid-1920's had depended heavily
on borrowing from abroad, much of it short term. By the early 1930's the shortterm component had, if anything, increased. Especially was this true of the
Austrian and German banking systems, the bulk of whose short-term obligations
to foreigners was held by London and New York. Amsterdam, Paris, Zurich, and
other centers in turn maintained large short-term claims on London and New
York. Hence the announcements in the spring and early summer of 1931 that the
largest Austrian commercial bank and a major German bank were insolvent
precipitated a scramble for liquidity which dealt devastating blows successively
at the international financial positions of Vienna, Berlin, and London. In each
instance the withdrawal of foreign balances, together with the flight of domestic
capital, swept away a large part of the gold and foreign exchange that was
available to the authorities to support their respective currencies and forced
Austria and Germany to adopt exchange controls and moratoria on foreign
obligations and Britain to abandon sterling's gold parity.

26




3. The Role of Central Bank Cooperation
INTRODUCTION

In dealing with the problems of 1924-31, the authorities in the major countries
were hobbled by international political conflicts and a dim perception of how
much the world had been changed by the wars, revolutions, and inflations of
1914-23. They were also handicapped by fundamental conceptual difficulties. The
goal toward which they strove and the norm by which they judged economic
behavior were defined by gold standard conceptions that purported to describe
the way in which the world economy had operated prior to 1914 and that had
a unity and simplicity whose appeal was widespread and powerful. Supported
by the great weight of opinion in their countries the major central bankers
worked, sometimes with almost moral fervor, to realize the gold standard
ideal. But the economic clay was recalcitrant. The vision of a unified world
economy, functioning according to accepted rules, could not be fitted to the
circumstances of the 1920's. Yet, so influential was the gold standard vision
that the few attempts to develop alternative conceptions of the working of
the world economy were dismissed or ignored by the authorities. Rather than
discard the cherished vision, each central banker sought to reinterpret the traditional theory to suit his country's particular needs. However, as these needs
were seldom similar and often diametrically opposed, the reinterpretations resulted in conflicting views about the manner in which the gold standard should
operate. The result was confusion and misunderstanding and in the end a failure
to deal effectively with the economic difficulties of 1924-31.

THE GOLD STANDARD CONCEPTION OF CENTRAL BANK COOPERATION

Under the gold standard conception to which the central bankers adhered in the
early 1920's, the role of central bank cooperation was distinctly limited. Indeed,
there would normally be little need for it. This followed because the countries
adhering to the system were expected to behave, for the most part, like components of a unified economy. The currencies were to be roughly stable in terms
of both gold and each other, fluctuating only within the gold points. Balance-ofpayments disequilibria were to be speedily corrected by economic adjustments
within the affected countries. The rules of the game by which adjustments were
to be achieved and exchange rate stability maintained had been described in
countless textbooks and official reports and were well understood by the central




27

bankers. An illustration of the way in which they expected the gold standard
to operate is found in a letter that Benjamin Strong wrote in May 1924 to Secretary of the Treasury Andrew Mellon.1
Prior to the war the regulation of both domestic and world general price levels was
a more or less automatic affair. The payment of the small net balance of indebtedness between the nations resulting from all trade and services by shipments of gold
had the effect of depleting bank reserves in the country where prices had advanced
too rapidly and overtrading and speculation had developed, thus forcing advances
in the rate of discount of the bank of issue and of interest rates generally, which
in turn induced borrowers to liquidate stocks of goods in order to pay loans, so
again reducing prices and restoring world price equilibrium. Comparatively slight
but rapid advances of prices in one country were automatically corrected by these
means and a fairly stable level of world prices resulted. No arbitrary human agency
had to be employed. There was no one charged with direct responsibility for
regulating prices and the extent to which discretion or judgment entered into the
processes of adjustment was very slight. Even the judgment which was exercised
was forced upon those responsible by movements of gold which made changes of
interest rates imperative and hardly a matter of discretion.

In this conception, the flexibility, the adjustability, of the economic components of the international system made cooperation almost unnecessary. Or
viewed from another angle, the adherence of the central banks to agreed rules
of behavior, and especially to the principle that the maintenance of exchange rate
stability should take precedence over all other economic objectives, was itself
a form of cooperation that normally made other forms of cooperation redundant.
In exceptionally severe financial crises, it is true, a central bank could legitimately
turn to its counterparts abroad for emergency credits. But such credits would
be short term in character and would be supported by vigorous monetary
measures to restore equilibrium. The unusual character of such cooperation was
underlined by the very few occasions on which the British central bank received
special credits from abroad during the seventy years following the enactment of
the Bank of England Act of 1844.2

1
Strong letter to Andrew Mellon, May 27, 1924, pages 2-3, historical records of the Federal
Reserve Bank of New York. All unpublished materials, not identified as from other collections of
papers, are from this Bank's historical records.
2

R. G. Hawtrey, The Art of Central Banking (London: Longmans, Green and Co. Ltd., 1932), page
229; Arthur I. Bloomfield, Monetary Policy Under the International Gold Standard: 1880-1914
(New York: Federal Reserve Bank of New York, October 1959), pages 56-57.

28




An important corollary of gold standard theory was that governments normally
played an extremely limited role in economic management. They were to provide
for the national defense and maintain domestic order so that market forces could
have free play. Budgets were to be balanced. While a government might legitimately combat combinations in restraint of trade, the notion that it might adopt
policies to counteract the swings in business cycles or to hasten the international
adjustment process had yet to be conceived. Overall economic policy, to the extent
that it existed at all, was a matter for central banks acting virtually alone.

CENTRAL BANKS AS INSTRUMENTS OF NATIONAL POLICY

The actuality of the central bankers' practice during the 1920's and early 1930's
bore little resemblance to the gold standard precepts to which they adhered.
In many ways their position was the reverse of that depicted. Whereas the
literature assumed that all gold standard countries followed agreed rules and
gave priority to exchange rate stability over all other policy objectives, the
actuality was that the central bankers disagreed continually about the way in
which the international financial system should operate and usually gave priority
to domestic objectives over external ones.
Indeed, whatever their hopes about the gold standard might be, Strong,
Norman, Moreau, and Schacht were first and foremost national central bankers.
Although each of their institutions was, in varying degrees, legally independent
of its government, this independence in no way reduced the central banker's
drive to serve his national interest as he saw it. Sometimes, as with Norman,
the view taken of the national interest was a very long-term one that identified
the interest of his own country—some would say of his own financial community
—with that of the international financial system.3 In other instances, the central
bankers devoted themselves more overtly to national objectives as Schacht did
in his perennial efforts to achieve a scaling-down of reparations or Moreau in
his attempt to mark out spheres of financial influence in Eastern Europe and
thus counter what he felt to be the excessive influence of the Bank of England
in that area. Even when they did not fully share the views of their respective
governments, prudence demanded that these views be taken into account by

3
L. V. Chandler, Benjamin Strong, Central Banker (Washington, D. C: The Brookings Institution,
1958), page 261. See also discussion in next section.




29

the central bankers. A letter that Strong wrote to Norman in February 1922
illustrates the point.4
In the face of a powerfully organized antagonism in Congress, the Federal Reserve
System must, to a considerable extent, rely for its protection against political attack
and interference upon the present administration . . . .
We cannot afford, practically or politically, to embark upon a course which ignores
the policy of the administration, which would possibly antagonize the administration and place us in the position where we would be quite helpless to resist the
repeated efforts which have been made in Congress to effect important and possibly
vital modifications in the underlying principles of the Federal Reserve System.

Each central banker thus formulated his policies within the framework of his
country's needs and political objectives. Within the whole range of these national
aims, the maintenance of the international gold standard, however defined, was
only one objective among many and one that could not always be easily reconciled with others to which vital importance was attached. In each country the
problem that confronted central bankers was essentially the same but took
different forms. In Britain there was a perennial conflict between those who
advocated an easing of financial policy to relieve the domestic economic stagnation and those who advocated a strict policy to support sterling in the exchange
markets. Sometimes in Germany, and more often in France and the United States,
the opposite side of the problem appeared in the form of a conflict between the
need to ease monetary policy in order to support sterling and the need to tighten
it in order to maintain domestic monetary stability.
Although the resolutions of these conflicts varied widely, they had one common
element—each central banker sought solutions that were consistent with the
needs of his own country as he saw them. When Moreau demanded in May 1927
that the Bank of England raise its discount rate to restrain a heavy flow of
sterling to Paris, Norman's response was that, given the condition of the British
economy, he could not do so "without causing a riot".5 The focus of Strong's
concern was at least as clear. His views are vividly expressed in a letter written
to Norman early in 1923 when the United States economy was expanding rapidly

4

Strong letter to Norman, February 18, 1922.

5

Confidential minutes of meeting of Friday, May 27, 1927, page 10, sent to Strong by Moreau on
June 2, 1927. The background and outcome of this episode are discussed at length in Chapter 6.

30




and was at the same time attracting a substantial inflow of gold. Since, contrary
to the traditional "rules", Federal Reserve discount rates had been increased to
levels well above the Bank of England's, Strong felt the need to explain:6
Never, I suppose, have the factors which should move us in our rate policy been so
carefully examined and considered, as recently. The results convinced me that our
action was required, and that with our excessive gold stock we must entirely ignore
any statutory or traditional percentage of reserve, and give greater weight to what
is taking place in prices, business activity, employment, and credit volume and
turnover.
Of course we must not close our eyes to the bearing this may have upon Europe
Enlightened opinion on this score must appreciate that should we dissipate our
credit resources in speculation and price boosting, in the long run Europe will
suffer. The advantage—a very temporary one—of a high price market in which
to sell us goods would be more than offset by the ultimate disorders of readjustment . . . .
You may be sure that inflation has no charms which have not been analyzed by
Reserve Bank men and rejected as spurious.

A few months earlier Strong had stated his position in more general terms when
he commented on a proposed statement of principles for central bank cooperation
that had been sent to him by Norman:7
no "surrender of sovereignty" should be attempted under the guise or through
the formalities of this expression of principles. I think you realize, as we have here,
and as in later years we may have occasion to realize it even more strongly, that
the domestic functions of the bank of issue are paramount to everything and that
anything in the nature of a league or alliance, with world conditions as they are,
is necessarily filled with peril.

As it evolved during the 1920's, therefore, the international financial system
lacked the unity, the tendency to move toward equilibrium, that was implicit
in the gold standard conception. On the contrary, given the criteria for policy
accepted by the authorities, there was no necessary reason, short of the exhaustion
of a deficit country's international reserves and devaluation, why disequilibrium
could not persist indefinitely. If adjustment policies could not be implemented
in deficit countries because of industrial rigidities and unemployment, or in

6

Strong letter to Norman, February 22, 1923, page 2.

7

Strong letter to Norman, July 14, 1922, page 3. Italics added.




31

surplus countries because of the threat of inflation, an impasse was reached.
Yet the chances of avoiding such an impasse were greatly reduced because the
authorities viewed international financial developments through the distorting
lenses of the gold standard conception and of their individual national needs
and aspirations. To Norman, viewing his difficulties toward the end of 1926,
the international machine was out of gear because the American authorities
were acting on gold standard principles different from the orthodox ones on
which he felt European financial policy was based.8 Similarly, the Bank of France
absolved itself of responsibility for the heavy flow of gold from Britain to France
that threatened the stability of sterling in the third quarter of 1929 by attributing
the movement to the free play of market forces under the gold standard.9 In
Berlin, London, Paris, and New York equally, it was always easy both to advise
other countries about how their economic policies could be improved and also
to find reasons why nothing more could be done at home.
Nowhere is the failure to achieve a consensus on the operation of the international financial system in the interwar years better reflected than in Norman's
testimony before the Macmillan Committee early in 1931. He had been asked
to discuss the Bank for International Settlements (BIS) and had stated that its
monthly meetings had led to extraordinarily good relations among the central
bank governors:10
Nevertheless, a Central Bank Governor, when he comes to Basel, is necessarily
to a certain extent dominated by the particular state of his own fiscal and economic
situation at home, of the mind of his public, of the mind of his Government, and
it is a matter therefore, not, as I say again, of establishing personal friendship—
which we have done to a marvellous degree, I think, almost intimacy—but of
taking up questions which are international and regarding them, as one would say,
on their merits. No two or three countries can really regard an international question on its merits. There is no such thing as merits when you come to that. To get
the affairs of a country, and possibly the needs of a country, considered on their
merits is a thing which has not been achieved, and, I believe, will not be achieved
for a very long time. The moment the position of certain countries is mentioned
you get a reaction for or against, for particular reasons applicable to the individual
in whom the action takes place. There will be a new Europe before we get away
from that.

8

Sir Henry Clay, Lord Norman (London: Macmillan & Co., Ltd., 1957), page 224.

9

Infra, page 166.

10

Committee on Finance and Industry, Minutes of Evidence (London: H. M. Stationery Office,
1931), Vol. 11, pages 298-99 (hereafter referred to as Macmillan Evidence).

32




THE DEBATE ON CENTRAL BANK COOPERATION

The conflict between the gold standard ideal and the actuality of national central
banking led to much confusion in the debate over the appropriate forms of central
bank cooperation. To be sure, each of the major central bankers supported at
one time or other the principle of cooperation. Moreover, Strong and Norman
at least came to understand that the international financial system could not be
left to run itself, that there was no automatic means by which exchange rate
stability could be maintained, and that, on the contrary, the international system
required management which could be successful in the long run only if there
were cooperation among the major countries.11
But while the central bankers were agreed on the principle that cooperation
and management were necessary, it was less easy to agree on any particular plan
because each proposal that was officially discussed was designed either to serve
specific national interests or at least to be consistent with such interests. Unfortunately for the clarity of the discussions, the proposals were not always put forward in terms of the national interest but in terms of how they would restore or
maintain some undefined form of the gold standard. Certainly this was true of the
major proposal supported by the British authorities, discussed below. By the same
token, it was all too easy for the central bankers to employ the gold standard ideal
as a debating weapon against proposals they found unacceptable. Strong did this
when he testified against legislation that would have made price stabilization a
specific responsibility of the Federal Reserve:12
I keep getting back to the subject of the gold standard, Mr. Chairman, because I
have great confidence that when the time comes to conduct these things as they
were in former years, a lot of the need for the type of management which has to be
applied in the present situation will be eliminated. It will be more automatic. We
won't have to depend so much on judgment, and we can rely more upon the play
of natural forces and their reaction on price . . . .

Norman employed similar arguments to discourage consideration of a proposal
sent to him by Strong on February 7, 1922 for an international stabilization fund

11

Strong letter to Norman, October 19, 1927.

12

United States Congress, House Committee on Banking and Currency, Stabilization,
ings, 69th Congress, 1st Session., on H.R. 7895 (Washington, D. C, 1927), page
testimony was given on April 9, 1926. See also Strong's testimony on March 19, 1928,
Congress, House Committee on Banking and Currency, Stabilization, Hearings, 70th
Session, on H.R. 11806, page 18-21.




Part 1, Hear379. Strong's
United States
Congress, 1st

33

that would have given the Federal Reserve a central role in the effort to restore
the international financial system. Norman wrote:13
Generally speaking, I do not believe that any artificial means for the stabilisation
of Exchange would, if ever, be practicable until the [intergovernmental war] debts
have been settled, the Reparations adjusted, and free Gold Markets have again
become much more general than they are now. And is it not true that when these
things shall have happened stability in the Exchanges will be looking after itself
in the old-fashioned way and artificial stabilisation will hardly be necessary?
THE GENOA PROPOSALS. Although Norman's arguments against the stabilization
fund proposal are hard to reconcile with his position as a major proponent of
international currency management, his view is understandable when account is
taken of the fact that four days earlier he had sent to Strong drafts of the resolutions that were to be presented by the British delegation to the Genoa Conference14 and that aimed, as William Adams Brown, Jr., has said, to restore
by agreement and management as much as possible of the London-centered international financial system that had existed prior to 1914.15 It is unlikely, however,
that Norman would have admitted any inconsistency between the grounds on
which he rejected Strong's stabilization proposal and those on which the Genoa
resolutions were based. Looking back on the Genoa proposals it is easy to see
how they served the interests of the London financial community and of the
Bank of England in particular. But from this distance in time, it is also easy to
forget that in the early 1920's London was the center to which the world was
looking for financial leadership. The United States had withdrawn into isolation.
France and Germany were devastated by war and inflation. London had been
the world's financial center before 1914 and, with its famed institutions and
skills intact, it sought to resume its historic role. Indeed, central bankers around
the world urged Britain to do so and waited with impatience for it to act. In
their minds almost as much as in Norman's, the restoration of sterling to its pre-

13

Norman letter to Strong, February 27, 1922, pages 1-2.

14

This conference, the most notable of a series dealing with international economic and financial
problems after World War I, was called by the governments of Britain, France, Italy, Belgium, and
Japan and met between April 10 and May 19, 1922. Representatives of thirty-three governments
participated, including those of nearly all European countries. The United States was represented
only unofficially.
15
W. A. Brown, Jr., The International Gold Standard Reinterpreted, 1914-1934 (New York: National
Bureau of Economic Research, Inc., 1940), Vol. 1, pages 345-46.

34




war parity and that of the gold standard were an identity. Given this viewpoint, it
required no great intellectual agility for Norman to reject Strong's stabilization
proposal because it was "artificial" while at the same time holding that the Genoa
proposals were a legitimate means to reestablish the gold standard.
The proposals, which were presented to the Genoa Conference by the British
delegation and were adopted by the Conference's Financial Commission, reflected
by far the most ambitious official attempt of the 1920's to organize central bank
cooperation.16 In addition to enumerating various economic and political prerequisites for a return to gold, the resolutions contained three major, interrelated,
and controversial proposals. These were to the effect that central banks (1)
should so conduct their credit policies as to prevent undue fluctuations of
domestic prices in terms of gold, (2) should cooperate continuously with each
other, and (3) should be divided into two categories: (a) those in center countries
which would hold their international reserves entirely in gold and (b) those of
other (unspecified) countries which would hold their reserves partly in gold and
partly in foreign exchange, i.e., in short-term claims on the center countries.
Those who drafted the Genoa proposals were aware that they contained many
difficulties. Nevertheless, they hoped that, despite the deep-seated political disagreements of Europe, central bankers could rise above national preoccupations
to cooperate for the attainment of common monetary objectives. For this purpose
the Genoa proposals set forth what, seen from the British viewpoint, appeared
to be minimum requirements. So long as prices in the major countries fluctuated
as wildly as they had in the earlier postwar years, there was little hope for the
successful restoration of an international gold standard. Price stability was surely
an apolitical aim to which apolitical central bankers could agree. Furthermore,
if existing or newly established central banks required foreign financial support
to stabilize their currencies in terms of gold, surely it was not unreasonable to
expect them to hold a substantial part of their international reserves in the
centers from which the support was supplied. In addition, specifically British
interests were at stake. Between its peak in May 1920 and the low reached
in January 1922, the index of United States wholesale prices had dropped
49 per cent. Over the same period the corresponding British index had declined
50 per cent, but the sterling-dollar exchange rate in January 1922 averaged only

16

The resolutions are printed in the Federal Reserve Bulletin, June 1922, pages 678-80.




35

$4.22, still well below the $4.867 parity at which the authorities were aiming.
Any further decline in American prices—or, to use the terminology of the Genoa
resolutions, any further rise in the gold value of the dollar—would further
complicate Britain's problems in reestablishing the prewar parity. Britain's
problem would be complicated still further if other countries that were attempting
to stabilize employed the proceeds of their stabilization credits, or deflated their
prices, in order to build up their gold holdings because, given the nature of the
international gold market, most if not all of the metal would be taken from
London.
Norman pursued one variant or another of the Genoa proposals throughout
most of the rest of the 1920's. Time and again he sought means by which to
manage gold flows so that the movement of the metal might be made to support,
or at least not interfere with, the aims of British monetary policy.17 When international commodity prices began to decline in 1926-27, Norman gave at least
tacit support to a League of Nations study of whether the decline was the result
of a scramble for gold by countries that were stabilizing their currencies and when
the results of the study were published he almost certainly approved of the
League's reiteration of the Genoa recommendations on the economization of gold
and particularly on the gold exchange standard.18 Norman also made repeated
efforts to convene the central bankers' meeting that had been proposed at Genoa.
Although the Genoa objective was not attained, Norman was instrumental in
bringing the central bankers together to discuss their problems face to face,
especially after the establishment of the BIS in 1930 provided them with a
monthly occasion on which to meet.
A major condition for the successful implementation of the Genoa proposals was of course the support of the
United States monetary authorities. Actually the initial reaction of the Federal
Reserve was not unfavorable. Strong secured the approval of the Harding Administration to attend a conference of central bankers that was to be called by
Norman, but the conference was postponed again and again because of difficulties
STRONG'S CRITIQUE OF THE GENOA PROPOSALS.

17
Strong, "A Bank Meeting", memorandum of July 22, 1925; Clay, op. cit., pages 224-25. Infra,
page 114.
18
League of Nations, Interim Report of the Gold Delegation of the Financial Committee (Geneva,
1930), page 20; Report of the Committee on Finance and Industry (London: H. M. Stationery
Office, July 1931), pages 122-23 (hereafter referred to as the Macmillan Report).

36




surrounding the settlement of European war debts to the United States and also
because the central bankers who were to participate had numerous other commitments. Moreover, as time went by, Strong came to have serious doubts about
the substance of the proposals themselves. Although at times he expressed his
misgivings in vague gold standard terminology, the basic ground for his objection
was that the proposals might impair the ability of the Federal Reserve to fulfill
its national central banking responsibilities.
Strong's misgivings centered at first on the proposal that credit policy should
be directed to the avoidance of undue fluctuations in the purchasing power of
gold, a proposal whose implementation, he believed, could run directly counter to
the central bank's obligation to maintain monetary stability within its own
market. This proposal, he wrote, had19
an ominous sound to me.
Expressing it from one point of view, the regulation of credit for the purpose of
maintaining the purchasing power of gold or the parity of currencies would imply
that the nation which had a discount on its currency should undertake, through its
bank of issue, to bring about a contraction of credit and currency; or, as in the
present case, the United States, with its currency at a premium the world over,
should undertake, through the Reserve Banks, to so regulate credit policies as to
expand credit and currency to a point where the value of our currency would
decline and consequently other currencies would approach the value of ours . . . .
From the standpoint of this country, we must be assured that we are not suggesting
or recommending a plan which if adopted would be anything in the nature of
handing a blank check to some of the impoverished nations of the world, or to their
banks of issue, and especially to those whose government finances are in complete
disorder and quite beyond control. This is one reason for caution from the standpoint of your country and ours, or at least your bank and ours, in accepting
principles which might have far reaching effects in practice.

Strong's opposition to any proposal that central banks should direct their
policy to the stabilization of prices in terms of gold became more vehement as
time passed. When, as mentioned above, the League reopened the subject in
1927-28 he used his influence to discourage the proposed study of whether the
decline in commodity prices was in any way connected with a gold shortage.
Strong indicated his skepticism about any such shortage, present or impending,

19

Strong letter to Norman, July 14, 1922, pages 5-6.




37

but he agreed that a purely factual study of the question might be useful. However, his serious concern was that the study would revive criticism along the oftrepeated line that the Federal Reserve was sterilizing gold and might recommend
that the United States should follow a policy of monetary ease at a time when
the Federal Reserve was struggling to restrain the culminating boom of the
1920's. It was with these concerns in mind that he told Sir Arthur Salter, the
director of the League's Economic and Financial Organization and an advocate
of the study, that neither the League nor any group meeting under its auspices
was in a position to advise central banks on policy and that, in particular, any
advice to the Federal Reserve would be very badly received in America.20
Strong's reaction against the Genoa proposals on the gold exchange standard
did not come until the late 1920's. He did not touch on the subject in his correspondence with Norman during the spring and summer of 1922, it being taken
for granted apparently that, in the turbulent conditions of the time, New York
and London would be the only centers in which other countries would hold the
foreign exchange component of their international reserves. As late as the spring
of 1926 Strong did Norman the favor of testifying, albeit somewhat backhandedly,
to the effect that India would better serve the cause of international monetary
stability if its currency were based, not on gold alone, but on the gold exchange
standard.21
Indeed, Strong did not turn against the gold exchange standard until he had
digested the details of the acute difficulties that arose between the central banks
of England and France in the spring of 1927. Thereafter, however, his letters
and memoranda expressed frequent misgivings.22 He was concerned that the gold
exchange standard facilitated a pyramiding of credit on small gold holdings in
center countries; that the conversion of foreign exchange balances into gold
might force the center countries sharply to increase their discount rates and so
bring about the very deflation that the advocates of the League study feared; and
that the gold exchange standard was, in some undefined way, undermining the

20

O. Ernest Moore, memorandum
of conversation between Governor Strong and Sir Arthur Salter,
May 25, 1928. See also Strong letters to Harrison, December 24, 1927 and July 8, 1928 (Harrison
Papers).
21

The Royal Commission
on Indian Currency and Finance,
Stationery Office, 1926), Vol. 5, pages 274-330.
22

Minutes of Evidence (London:

Strong letter to Norman, August 30, 1927, pages 2-3; Strong letter
11, 1928, pages 6-7; and Strong letter to Norman, September 19, 1927.

38




to Owen D . Young,

H. M.
June

gold standard. However, what lay at the bottom of these objections was Strong's
shock at the discovery that the Bank of France, in attempting to exercise its undoubted right to convert its large London balances into gold, had brought the
Bank of England to the point where it had to choose between severely deflating
the British economy or abandoning sterling's gold parity.23 To be sure, he foresaw no immediate danger that the Federal Reserve would face any similar
threat.24 Its holdings of gold above the minimum legal requirements were considered adequate to meet any likely foreign demands for the metal. Even so,
Strong could never approve of a system that might place in the hands of foreign
central bankers such power over Federal Reserve policy as had been wielded by
the Bank of France over the Bank of England. His conclusion was25
that every one of the nations should reestablish the gold standard, restore its
gold reserve or create a new one where needed, making special arrangements for
doing so where necessary, and reestablish its own domestic autonomy in monetary
matters without any such dependence upon other markets as implied by the gold
exchange standard. Important balances in London or New York of course may be
highly desirable, but the most desirable thing is autonomy and self-reliance and
good conduct on sound monetary principles at home.

As Strong's interest in the other Genoa proposals cooled, so did his willingness
to attend the proposed conference of central bankers. Almost from the beginning
he had been concerned about the direction that a large meeting of central bankers
would take and particularly about the possibility—indeed the probability—that,
as the representative of the only powerful creditor country, he would be placed
in the difficult position of having to disappoint the hopes for financial assistance
that might be entertained by the various European central bankers who would
attend the conference.26 As time went on, Strong found additional reasons to
resist Norman's repeated attempts to convene a formal meeting of central bankers.
As he explained them to Sir Arthur Salter in the spring of 1928, some of Strong's
reasons were strikingly similar to those expressed by Norman three years later
when he testified to the Macmillan Committee on the periodic meetings of the

23

An account of these difficulties and how they were resolved is given on pages 116-30.

24

Strong letter to John J. Mitchell, October 4, 1927, pages 1-2.

25

Strong letter to Owen D. Young, June 11, 1928, pages 6-7.

26

Strong letter to Norman, July 27, 1922.




39

BIS governors.27 According to a memorandum written by a member of the Federal Reserve staff who attended the meeting with Sir Arthur:28
Governor Strong stated that he had always been opposed to any sort of a formal
conference or meeting between the world's central banks, as contemplated at
Genoa, for several reasons. In his opinion, it was expecting entirely too much of
human nature to think that representatives of the central banks of a great many
nations having differences of language, customs, beliefs and financial and political
needs could sit down together and agree on anything at all. Moreover, in the
course of his travels he had had occasion to meet a great many of the central
bankers, as well as governmental and political representatives, and to learn considerable about the relations between the various banks and their governments, and
he was not at all convinced that at a general meeting or in any common organization of central banks the policies of certain banks would not be dictated by the
interests of their respective governments rather than by purely monetary considerations. Another point was that in any formal meeting of the central banks, the
Federal Reserve Bank would represent the only lending market, while the others
would all be borrowers, and he would never consent to go to such a conference
with the prospect of being outvoted on every issue of any importance which affected
the Federal Reserve System. He would have to be sure of having one more vote
than all the borrowers combined.

A PRAGMATIC APPROACH TO CENTRAL BANK COOPERATION

Strong's misgivings about the Genoa proposals did not prevent him from developing a distinctive approach to central bank cooperation that, largely because it was
consistent with his and his colleagues' responsibilities as national central bankers,
was the approach that was generally followed during the 1920's and early 1930's.
Briefly, he would encourage the Federal Reserve to cooperate with foreign central banks so long as the assistance contributed to the welfare of the United States
or at the least was consistent with United States interests.29 Apart from his 1922
flirtation with the idea of attending the proposed central bankers' meeting, he
consistently favored what John H. Williams later called the "key currency" approach to cooperation. Early in 1919, he wrote Norman that, if currency stabilization could be worked out between the British, French, and United States
central banks, he doubted that "anything further would be required or desirable

27

Supra, page 32.

28

O. Ernest Moore, memorandum of conversation between Governor Strong and Sir Arthur
Salter, May 25, 1928, pages 4-5.
29

Strong letter to Norman, March 21, 1921, cited in Chandler, op. cit., page 247.

40




for many years to come".30 As things turned out, Strong participated with other
central bankers in the work of monetary reform in numerous countries, but his
major efforts were indeed directed successively to the stabilization of the British
and French currencies. In all these efforts, his approach was selective and pragmatic. Discussing his monetary stabilization experiences before a Congressional
committee in the spring of 1926, he declared that the best chances for success
lay in dealing with the problems in detail, taking one situation at a time, rather
than to attempt to deal with all countries at once.31
It was within this framework that cooperation between central banks developed. The first steps involved the performance of services that are now considered routine but that had not ordinarily been performed prior to World War I:
the earmarking of gold, the investment of deposits, and the regular exchange
of information on current developments in the financial markets. As the friendship between Strong and Norman grew in the early 1920's, it was entirely
natural for them to exchange views on the economic outlook and on prospective
changes in monetary policy and to offer each other advice on how to deal with
the problems with which they were confronted. Such exchanges grew with the
widening of Strong's circle of acquaintances among central bankers although
none attained the intimacy of the exchanges with Norman. After Strong's death
in October 1928, George L. Harrison continued these exchanges of information
and opinions and was aided in this by the development of the transatlantic telephone which provided the central bankers with a means of daily and even hourly
communication.
Financial cooperation between central banks arose in the 1920's as one of the
ingredients normally required for currency stabilization. In these stabilization
efforts, Strong and Norman (the latter often working in cooperation with the
Economic and Financial Organization of the League of Nations) took the lead
in formulating the conditions that, in their views, would facilitate an early return
to gold payments and fixed exchange rates. Governments, which had dominated
the central bankers during the war and early postwar years, should resume the
role in which they were cast under the gold standard conception.32 Above all,

30

Strong letter to Norman,

February

5, 1919, page 3.

31

House Committee on Banking and Currency, Stabilization, Hearings on H. R. 7895, April 12, 1926,
page 504.

32

Supra, page 29.




41

they should put their finances in order. Budgets should be balanced; large-scale
borrowing from the banking system should stop; excessive amounts of floating debt should be funded so that the need to refinance large and frequent
short-term maturities would not impede the central banks' monetary control.
Beyond this, central banks should be legally independent of governments and
the continuity of central bank managements should be assured. On the external
side, each country's prices should be brought into alignment with those of its
major competitors at the exchange rate chosen for stabilization; its merchandise
trade should be in balance or at least approaching balance; and the central
bank should have accumulated international reserves adequate to convince
the market that it could defend its chosen exchange rate against the payments pressures to which its currency was likely to be exposed in the ordinary
course of events.33 When a country had met these conditions and had thus covered most if not all of the road toward its goal, the central bank of the stabilizing
country would obtain from one or more other central banks a credit calculated
to insure market confidence in the stabilization plan. Sometimes the central bank
credit would be supplemented by a commercial bank, or market loan, to the government of the stabilizing country. In either case, the arrangements normally
would be of no more than two years' maturity and would be considered most
successful if the funds made available under them were never drawn. Indeed, such
loans and credits were most easily obtained when they were least needed.
In the areas of routine services, information exchange, and stabilization, central bank cooperation by and large was quite successful in 1924-31, but the main
challenge lay elsewhere. It involved the question whether, once stabilization had
been completed, cooperation could be continued to the extent required to maintain stable exchange rates. In meeting this challenge, the central bankers were
subject to serious constraints. Monetary policy could be brought into play only
when the central bank's international aims happened to coincide, or at least not
conflict, with its domestic ones. In the efforts of the Federal Reserve to support
sterling, for example, these aims happened to coincide in 1924 and again in 1927
when the United States economy was in recession. But when sterling came under
severe pressure during the 1928-29 American boom, the conflict that arose in the

33
Chandler, op. cit., pages 261-62 and 281; Brown, op. cit, pages 343-44; and Strong letter to
Herbert Hoover, April 22, 1922.

42




internal and external aims of Federal Reserve policy was resolved in favor of
the central bank's domestic economic objectives.
The same constraints applied to inter-central-bank credits because, whatever
form they took, their use for support operations acted to expand the cash base
of the creditor country's banking system. In a period during which the Federal
Open Market portfolio was never much over $500 million, the addition of as
much as $50-100 million to member bank reserves was always a matter to be
taken seriously into account. Such an addition to the cash base was acceptable if
the Federal Reserve was in any case aiming to ease money market conditions.
It might even be acceptable if the Federal Reserve aimed at no change, provided that the holdings of the open market account were sufficiently large to
permit offsetting sales. But when, as happened in the early months of 1929, the
ammunition in the open market portfolio was virtually exhausted and when the
Federal Reserve was struggling to keep a tight rein on the cash base, the scope
for currency support operations was drastically restricted. Limitations such as
these were even more severe in the cases of the Reichsbank and the Bank of
France, neither of which had adequate authority to conduct meaningful open
market operations.34
In addition to these problems, there were other more fundamental limitations
on central bank cooperation that stemmed from the conceptual confusion about
the nature of the international financial system. The authorities perceived only
dimly, if at all, that in a system managed by national central banks the process of
international adjustment was likely to be far slower and more difficult than it
was supposed to be under the gold standard conception. Guided in their thinking
by this vision, the authorities tended to demand too much from the adjustment process and to allow too little in the way of financing to support weak
currencies while needed adjustments were taking place. Paradoxically, the British
authorities, who presumably favored the managed currency proposals of Genoa,
frequently took the sternest line—as did one economic adviser to the Bank of
England in December 1930 when he said that Federal Reserve operations in
support of sterling should not go above £ 1 0 million because of the danger that
additional support would encourage England to postpone the making of funda-

34
M. G. Myers, Paris as a Financial Centre (New York: Columbia University Press, 1936), pages
29-30; M. H. deKock, Central Banking (3rd ed.; London: Staples Press, 1954), pages 206-8.




43

mental adjustments. Norman was just as stern toward the end of August 1931
when he intimated that his government should adopt a retrenchment program
sufficiently drastic to eliminate entirely the need for Britain to seek credit from
abroad.35 Presumably, attitudes such as these explain, at least in part, why on
several occasions the British authorities declined financial assistance offers that
were made by the Federal Reserve.
In these circumstances, it is hardly surprising that inter-central-bank currency
support operations were on a relatively small scale and for relatively short
periods—although certainly considerably larger in amount and longer in maturity
than any that had been granted before 1914. Apart from the 1927-28 Bank of
France swaps—which in any event were undertaken only to facilitate the achievement of French domestic monetary aims—the largest market operations of which
we have any record were those conducted by the Federal Reserve in support of
sterling. These rose to a peak of $60 million equivalent during the British-French
difficulties of May-June 1927 but did not again exceed $35 million even in the
difficult days of 1929-31. Larger amounts were granted under the more formal
inter-central-bank credits, but even these credits tended to be too little and too
late and—despite all the emphasis on the need for adjustment—were often uncoordinated with any convincing program to correct the underlying difficulties.
The suggestion for a credit sufficiently large to stop a liquidity crisis and restore
market confidence was made only once—by Hans Luther—at the height of the
flight from the reichsmark in July 1931 but in such unpropitious circumstances
that it was brushed aside as completely impractical.36

35
See Chapters
Issues (London:
36

7 and 8; Report of the Committee o n the Currency and Bank of England N o t e
H. M. Stationery Office, February 1925), cited in Chapter 5.

Infra, page 194.

44




4. The Bankers' Role in the Stabilization
of the German Mark: 1924
INTRODUCTION

The great international effort to stabilize the German currency surmounted many
of the problems of the early 1920's reviewed in the preceding chapters. International political conflicts and financial rivalries were muted. The work that had
previously been done both by the central bankers and by the League of Nations
on the formulation of the conditions for currency stabilization bore fruit. Although
the stabilization was heavily influenced by orthodox gold standard conceptions,
the bankers' pragmatism also found expression in the adoption of management
techniques that afforded a degree of protection to the German currency against
external pressures.
Judged by its immediate outcome the stabilization was undoubtedly a great
success. It marked the turning point of Europe from the political and economic
upheavals of the early postwar years toward the short-lived normalcy of the later
1920's. It set the stage for the return to gold by Britain and several other countries
and for the great outflow of United States capital that was to play such a crucial
role in the developments of the next four years.
These achievements grew out of earlier disillusionments. By 1924 it was clear
that the recovery of Europe was inseparable from that of Germany1 and that a
reparation settlement was a prerequisite for the reestablishment of international
monetary order. Bitter experience had shown that attempts to collect reparations
—regardless of the impact on the mark—led to extreme instability in the foreign
exchange markets for the major Allied currencies.2 As the mark depreciated,
Germany's economic difficulties rubbed off on its major creditors and particularly on France, whose budgetary and reconstruction plans were based to a
significant extent on the collection of reparations. Eventually, economic breakdown in Germany led to default on reparations and to the occupation of the
Ruhr by French and Belgian troops. Germans retaliated with passive resistance,

1
See, for example, the address of Charles E. Hughes before the annual meeting of the American
Historical Association at New Haven, Connecticut, December 29, 1922. The relevant parts of the
address are reprinted in H. G. Moulton and Leo Pasvolsky, World War Debt Settlements (Washington, D. C: The Brookings Institution, 1929), pages 168-74.
2
W. A. Brown, Jr., England and the New Gold Standard, 1919-1926 (New Haven: Yale University
Press, 1929), page 119.




45

and the resulting stalemate led to further depreciation not only in the mark but
also in the French and Belgian currencies. With the collapse of the mark
in 1923 and the strangulation of the German economy, it became unmistakably
clear that a dead end had been reached.
The incentive for Germany to cooperate in a stabilization program was certainly as strong as that of her former enemies. Such a program raised hopes that
the Allied occupation could be terminated, that Germany's currency, precariously
stabilized since the introduction of the rentenmark in the autumn of 1923, could
be given a firmer foundation, that foreign exchange would become available to
finance a revival in German foreign trade, and that this revival would in turn
facilitate a recovery of the domestic economy.
The potential for recovery was at hand. To protect their financial assets against
inflation, German industrialists had used their resources to modernize and expand
plant and equipment.3 The labor force was highly skilled. Germans held large
foreign assets that could be mobilized if confidence in Germany's currency and
in its economic and political future could be restored.4 On the other hand, the
inflationary process had destroyed the domestic financial assets of many families
and had reduced the Reichsbank's holdings of gold to a mere $111 million,
of which all but $48 million was pledged against credits from abroad.5
Thus, a catalyst was required that would release the existing potential of the
German economy and attract foreign financial support. Conflicts over reparations
had to be settled or at least suspended; Germany's territorial integrity had to be
restored; and reparations could be extracted only to an extent compatible both
with German economic recovery and with the stability of the German currency
on the foreign exchange markets.
In the attainment of these conditions, central and commercial bankers from
both sides of the Atlantic played a vital role about which little is to be found
in the literature on the 1920's. History has of course recorded that prominent
bankers served on the two Experts' Committees that were appointed in Novem-

3

Committees of Experts, Report to the Reparation Commission, reprinted in Federal Reserve Bulletin, May 1924, page 356 (hereafter referred to as Experts' Report); Carl Bergmann, The History
of Reparations (Boston: Houghton Mifflin Company, 1927), page 255.
4

Experts' Report, pages 371-72; Bergmann, op. cit., page 255.

5

Hjalmar Schacht, testimony before a subcommittee of the Experts* Committees, January 19, 1924,
page 8 (Young Papers).

46




ber 1923 by the Reparation Commission. Charles G. Dawes, the American
financier, who was soon to become Vice President of the United States, served
as chairman of one of these committees while Reginald McKenna, formerly
British Chancellor of the Exchequer and head of one of London's large clearing
banks, chaired the other. Among other notable members were Owen D. Young
and Sir Robert Kindersley, directors respectively of the Federal Reserve Bank
of New York and of the Bank of England. Much has also been written about
the conference of July 16-August 16, 1924, at which the major European
governments, using the Experts' Report as a basis, negotiated the London
protocol which laid the foundation for the stabilization.6 But while this much
is well known to historians, little has been published on the private discussions
that lay behind the Dawes Plan, as the report of the Experts' Committees came
to be known, or on the influential part that the bankers played during the
London Conference and the frantic negotiations that preceded the flotation of
the Dawes Loan.
Throughout this effort, central and commercial bankers cooperated closely.
Although neither Schacht, who had been appointed President of the Reichsbank
in December 1923, nor Norman were members of the Experts' Committees,
both were at the center of the stabilization effort. American commercial bankers
came on the stage largely because the Federal Reserve, like the Bank of France,
played no major role. The French authorities of course were desperately preoccupied with their own financial difficulties and in any event were unsympathetic
with Germany. The Federal Reserve's attitude reflected the ambivalent position
of the United States Government. The doctrine that there was no connection
between Germany's reparation obligation to the European Allies and the Allied
war debt to the United States kept the Washington administration officially aloof,
although—as we shall see— Secretary of State Hughes maintained a strong, informal interest in the stabilization effort.7
The vacuum left by the United States authorities was filled by J. P. Morgan
& Co. This firm of investment bankers had vigorous and informed leadership,

6

Bergmann, op. cit., pages 260-65; J. W. Angell, The Recovery of Germany (New Haven: Yale
University Press, 1929), page 62.

7
Address by the Secretary of State before the annual meeting of the American Historical Association
at New Haven, Connecticut, December 29, 1922, cited above. See also Bergmann, op. cit., page 191,
and F. L. Benns, Europe Since 1914 (5th ed.; New York: F. S. Crofts & Co., 1944), page 244.




47

was the fiscal agent in the United States of the British and French governments,
and had close working relations with the banking houses of Morgan Grenfell in
London and Morgan, Harjes in Paris. J. P. Morgan was on almost as close terms
with Norman as Strong himself. For several years before the Dawes Loan was
floated, Morgan had, together with Norman, been considering the possibility of
an international loan to deal with the reparation problem.8 When the question
of securing American capital to support German stabilization arose in 1924,
it was therefore natural for Morgan's to play a major role. In these negotiations
the firm was represented by one of its partners, Thomas W. Lamont, who was in
Europe almost continuously from January 1924, when the work of the Experts
began, until the Dawes Loan was floated in mid-October. Indeed, Lamont, together with Owen D. Young, played the role that Strong might have played had
his government's policy permitted him actively to participate.

COOPERATIVE ASPECTS

The provisions of the Dawes Plan in which the bankers were particularly interested were those relating to the annual amounts of reparation payments, the
transfer of these payments across the exchanges, and of course the international
loan.
The thorny question of Germany's total reparation obligation, on which earlier
negotiations between Germany and the European Allies had broken down, was
sidestepped. The Experts' Committees recommended an interim arrangement under which the annual payment would be geared to the changing strength of the
German economy. Accordingly, the Committees established a rising scale of
payments for the first five years of the plan and provided that payments thereafter should be determined by a "prosperity index" whose details they specified.
Under this system, as the Experts' Report stated:9
Germany will retain her incentive to develop, as she retains the major part
of the advantage of any increase in prosperity, while the Allies obtain a reasonable
share in this increase and avoid the risk of losing through a premature estimate of
future capacity.

8
9

Bergmann, op. cit., pages 125, 135, and 137.
Experts' Report, page 359.

48




The Experts' second crucial recommendation was that reparations should be
made in German currency to the Agent General for Reparation Payments who
was to represent the Reparation Commission in Germany. Once such payments
had been made, Germany's obligation to the Allies was fulfilled. Here the Experts
introduced a significant degree of currency management into their proposals.
For it was the responsibility of the Agent General, acting under the direction of
a Transfer Committee, to determine how these funds were to be employed: to
pay for reparation deliveries in kind, to meet Allied expenditures in Germany, or
to be transferred across the exchanges. But in all cases this power was to be
exercised only to the extent permitted by the exchange markets "without threatening the stability of the German currency".10 Commenting on this provision,
the Experts' Report said:l1
We are convinced that some kind of coordinated policy, with continuous expert
administration in regard to the exchange, lies at the root of the reparation problem
and is essential to any practicable scheme in obtaining the maximum sums from
Germany for the benefit of the Allies.

Having safeguarded the stability of the German currency against excess pressure from reparations, the Experts also proposed that the currency be supported
through the flotation by the German government of a foreign loan of 800 million
gold marks ($191 million), equivalent to 8 per cent of Germany's imports in
1924. In the Experts' view, the loan would serve a variety of needs. It would
facilitate the stabilization of the currency by financing the bulk of reparation
deliveries during the preliminary stages of Germany's recovery and thus help to
balance the government's budget and reduce government borrowing from the
banking system. It was also regarded as essential to the successful establishment
of the reformed Reichsbank. In particular, the loan would reconstitute the central
bank's depleted gold reserve and would provide the basis on which the bank
could at least partially satisfy the German economy's tremendous credit needs.
It would thus serve as a catalyst that would facilitate the revival of the German
economy and in so doing would help Germany, in the longer run, to meet its
reparation obligation out of its own expanding resources.12

10

Ibid., page 404.
Ibid., page 367. Italics added.
12
Ibid., page 368.
11




49

POLITICAL AND ECONOMIC CONDITIONS FOR THE DAWES LOAN. The bankers played

a prominent part in formulating the political and economic conditions for the
Dawes Loan, both while the Experts' plan was being drawn up and later during
and following the London Conference.
Our information on the early stages of the negotiations comes primarily from
the papers of Owen D. Young whose voluminous cable correspondence with
members of the New York banking community during March and early April
1924 foreshadowed virtually all the principal features of the loan that was floated
the following October. There was agreement that the loan should be a long-term
one and that service of the loan should have a claim on Germany's resources
prior both to reparations and to the service of any subsequent foreign loans to
Germany. It was expected that half of the loan would be taken in New York,
half in Europe, and of the latter part three quarters would be underwritten in
London. Since seven eighths of the whole operation was to be done between
New York and London, special emphasis was placed on the need for close AngloAmerican cooperation. On the broader political and economic issues, Young was
advised that American support for the proposed loan could be obtained only if
the Experts' recommendations had the "hearty" approval of Germany, France,
and England. The New York bankers did not object to the continued occupation
of the Ruhr per se, but pointed out that, if the occupation led to German passive
resistance or to a lukewarm British attitude toward the Experts' plan, the sale
of the bonds in the United States "would be most seriously affected . . . if not
made entirely impossible".13 Perhaps most significant was an unsigned cable
Young received, quite possibly from Morgan's, which emphasized that the flotation of the loan depended heavily on the adoption of a plan that would insure
the stability of the German currency and which employed arguments and
language especially designed to influence the French:14
it goes almost without saying that the ability of the American market to
absorb say $100,000,000 of the loan to the German Government would depend in
part upon the bankers' ability to convince the American market that reparations
[sic] payments in cash or in kind were not to be exacted from Germany in excess

18
Gerard Swope to Young, February 29, 1924. See also message transmitted to Young by Swope
on March 29, 1924; Swope to Young, March 22, 1924; and unsigned messages March 25 and 26,
1924 (Young Papers).
14

Unsigned message of April 2, 1924 to Owen D. Young (Young Papers).

50




of her ability to meet them. (Stop) It would of course be a mistake for any members of the Experts Committee to infer that since the proposed currency loan was
to constitute a first lien on Germany's assets and revenues (having priority to all
reparations payments) our markets would not therefore be concerned with this
question of the time when and the amount in which reparations payments were to
be required. (Stop) That is not the case. (Stop) Our markets will need to be
assured not only that the loan is a first lien on Germany's assets and revenues, but
also that it is the obligation of a solvent Government and a solvent country. (Stop)
The Experts who have studied cause and effect as they have operated for the last
five years undoubtedly realize even better than we that if reparations payments in
cash or in kind are to be exacted sooner or in amounts greater than Germany's
international balance of trade and payments permits, then renewed depreciation of
Germany's currency is bound to take place and the market position of bonds issued
to provide Germany with a sound currency is bound to become impaired . . . .
There is no one here who would not like to see the German Reparations payments
made just as large and just as soon as possible and no one who does not realize the
great importance to France that this should be so. (Stop) Our sympathies are all
for making Germany pay to the last drop. (Stop) Even friendly judgment however
is rather against killing the goose which should someday lay the golden egg for
France. (Stop) The immediate point however is that no operation of the kind now
contemplated appears to us to be feasible at all unless France has come to the
conclusion that the time has come when her interest will be best served by aiding
Germany establish a sound currency, something which seems to us to be impossible
until France is willing to defer her claim for reparations, not only in lien, but also
in time and amount so far as may be necessary to insure preservation of a sound
currency system in Germany.

The spirit of this message harmonized almost completely with that of the
Experts' Report, and from the date of the report's publication on April 9 until
the Dawes Loan was floated six months later, Morgan's worked in close collaboration with Young and Norman to persuade the Allied governments, particularly that of France, to bring their policies toward Germany into conformity.
Hope for success rose and fell like the tide. Within a few days of its publication
the Experts' Report had been approved by the Reparation Commission and by
the governments of Belgium, Britain, and Italy. In France, however, the government of Raymond Poincare reserved its position. Then toward the middle of May,
Poincare was defeated in the Chamber of Deputies on another issue and was
replaced by Edouard Herriot, whose government announced its approval of the
report early in June and joined with the British government to call a conference
to meet in London on July 16. But before the conference could convene a minor
crisis blew up that made it clear that even the conciliatory Herriot government
would be most reluctant to modify France's right to take independent action in
case Germany was found to be in default on reparations, or to give any body
other than the Reparation Commission the authority to decide whether such a




51

default had in fact taken place.15
Opposing the French position on both points, the British government sought
the views of the bankers. On July 12, Thomas W. Lamont wrote Prime Minister
MacDonald a letter that was expected to strengthen Britain's hand in the forthcoming negotiations and whose purport was made known to the French delegation
before the conference assembled.16 Alluding to the first point at issue, Lamont
stressed that the only basis on which the proposed loan could be successfully
floated in the United States would be that it have a first claim on all Germany's
resources. If this "fundamental condition" were to be qualified by an agreement
authorizing the Allied governments to take any action that would weaken this
prior claim or render service of the loan doubtful, then, he wrote, "I am confident
that American investors would reject the Loan". On the second point at issue
Lamont was equally categorical:17
I may say that in our opinion no Loan issued under the Dawes plan, one of the
conditions of which shall be that the Reparations [sic] Commission shall have
power to declare default on the part of Germany will be acceptable to the American
markets. It is not necessary to consider whether the attitude of the American public
towards the operations of the Reparations Commission is or is not justified: the
fact remains that, if the power to declare default is given over to the Commission,
America will not in our belief subscribe to the Loan.

A preferred procedure, Lamont suggested, would be to vest the sole power to
declare a default in the Transfer Committee, whose advantages from Lamont's
viewpoint were that it was expected to be independent of the Reparation
Commission and to be composed of bankers and foreign exchange experts in
whose appointment Morgan's expected to have a say.
After the conference began, the bankers raised additional questions. During
one long meeting that Lamont and Norman had with MacDonald,18 the Prime
Minister was told that New York not only was unsatisfied on the questions of
default and independent action but wished, in addition, to have information re-

15
Sir Henry Clay, Lord Norman (London: Macmillan & Co., Ltd., 1957), page 212; The Economist,
July 12, 1924, pages 43-44.
16

Lamont cable to MacDonald, July 25, 1924 (Lamont Papers).

17

Lamont letter to MacDonald, July 12, 1924, pages 1-2 (Lamont Papers).

18
Memorandum of August 5, 1924 (Lamont Papers). This memorandum is unsigned but was clearly
written by Norman.

52




garding (a) the withdrawal of French and Belgian railwaymen from the German
railways, (b) the military evacuation of the Ruhr, (c) the future means of interpreting the Versailles Treaty, (d) the date for the eventual evacuation of the
Rhineland, and (e) the role, in the period before evacuation, of the Rhineland
Commission which had been established under the Versailles Treaty to represent
the authority of the Allies in the occupied territory. MacDonald for his part
replied frankly that he saw little prospect of an early and satisfactory understanding with the French on any of these points. On the following day Norman gave a
detailed report of this discussion to the Chancellor of the Exchequer, Philip
Snowden, who indicated (according to Norman's report) that he was entirely in
agreement with the bankers' attitude. Norman also noted that he was to inform
Schacht that, in the circumstances, Morgan's was unwilling to open negotiations
for the loan with any representative of Germany.19
Gradually the conference hammered out compromises on these difficult issues.
The deciding of a default on reparations was left with the Reparation Commission, but a United States citizen would participate, and if any such decision were
not unanimous, it could be taken to arbitration by any dissenting member of
the Commission. Furthermore, the Allied governments agreed that, in case
sanctions were imposed on Germany for a default, the servicing of the loan
would be accorded priority over all other claims on the Germany economy.20
But still no date was agreed upon for the military evacuation of Cologne and the
Ruhr. On August 16 Norman wrote to Lamont that he had a long talk with
Snowden, who "was a good deal distressed at the way things had shaped themselves". Snowden, Norman wrote,21 had asked whether
under the conclusions now contemplated, the money required by the Dawes
Report would be forthcoming, but I could give him no definite answer whatever,
nor did I pretend to do so. He said, too, that so much pressure had been brought
to bear on the Germans in order to obtain their agreement that they were almost
acting under compulsion and, indeed, they might represent that they had done
so. . . .
I also had a talk with the Prime Minister [MacDonald] and found that he had
no definite ideas about the Loan and not much interest in it. He wanted a political

19

Memorandum of August 5, 1924 (Lamont Papers).
Final protocol of the London Conference and Annex IV, Inter-Allied
1924.
21
Clay, op. cit, page 215.
20




Agreement

of August 30,

53

agreement without delay, however it might have to be reached. Too much, he said,
had been heard of the Bankers' needs which continually antagonised the French
and some of his own friends (the Socialists). He was sure that no one in his
position could reach a better settlement than was being reached and even if not
perfect or so good as had been hoped, it would produce a new spirit in Europe.

Thereafter Norman left London on a much needed holiday,22 and discussions
on the loan seem to have abated until mid-September. Then, on September 17,
J. P. Morgan arrived in London. Preparing for his arrival, Lamont reviewed
the position with Norman. The latter reported that Schacht, whom the German
government had designated to conduct the loan negotiations, had already been
in touch with the Continental bankers about their tranches of the loan, and
had obtained indications that he might raise $10 million equivalent each in
Holland and Switzerland and an additional $2 million equivalent in Italy.
Another hopeful development was that the French Finance Minister, Clementel,
had indicated that his government would inform the bankers of its unequivocal
intention to complete the evacuation of the Ruhr within a year unless Germany
willfully defaulted on reparations.23 Nevertheless the bankers still hesitated.
When J. P. Morgan and Lamont saw Norman on September 17, they told him
that they had not yet decided on the principle of issuing the loan, let alone the
details, and Norman himself told the Bank of England's Committee of Treasury
that he (in conjunction with Morgan's) intended to conclude the negotiations
with Germany only after a satisfactory statement had been made regarding the
Ruhr and Cologne.24
Despite these hesitations, it seems clear that by this time Norman at least was
convinced that the loan must be issued. Indeed, Norman is depicted as a strong
advocate of the loan in a message that Lamont and Morgan sent to their New
York office after the September 17 meeting. The New York bankers voiced

22

Loc. cit.

23

T. W. Lamont cable to J. P. Morgan & Co., September 16, 1924 (Lamont
Allied military evacuation of the Ruhr was completed in 1925.
24

Papers).

In fact, the

Clay, o p . cit., pages 215-16. The Committee of Treasury is a committee of the Bank of England's
Court of Directors. In the 1920's it consisted of the Governor, Deputy Governor, and seven
other members of the Court and acted as a special body with which the Governors could consult
on all the more important business of the bank.

54




doubts about political pressures in Germany but they reported that25
Norman is quite certain (and he has seen Schacht quite a little) that apart
from the Communists on one side and the Nationalists on the other, in Germany
the great masses of the people want peace and are ready to make the necessary
sacrifices to g e t . . . .
Montagu Norman advances many other arguments; most of them may be
summed up in the statement that he has honest belief that the Germans intend to
meet this loan and the conditions of the Dawes Report honourably. Further that
the Germans will want a great deal more money than the contemplated loan all
of which will be subsequent [i.e., subordinate in lien] to that loan and to the total
reparations payments. But our loan will be a first charge on everything the country
has and the country is subjected to foreign control to an extent that has never yet
been accomplished in dealing with any nation; in fact he believes there is no foreign
loan at present in existence which offers as good security as this one.
Montagu Norman was perfectly clear that in his opinion unless the loan is made
Europe will break. If on the contrary it is made he believes that the results will be
as favourable as those of similar operations for Austria and Hungary have turned
out but on an even larger scale.

This message was included by Dwight Morrow, then a partner of J. P. Morgan
& Co., in a letter of September 18 to Secretary of State Hughes. After telling the
Secretary that the "German Loan proposed in the Dawes Plan now seems
imminent", Morrow made a prescient comment:26
What really impresses us favourably in Governor Norman's opinion is not the
extent of the foreign control upon Germany but the disposition of the German
people at the present time. We have some fear however that that disposition may
not continue. However desirous Germany is of getting the loan at the moment in
order to free the hold which France has upon the industries of the Ruhr it is almost
inevitable that this loan will be unpopular in Germany after a few years. The
people of Germany in our opinion are almost certain after sufficient time has
elapsed to think not of the release of the Ruhr but of the extent to which what was
once afirstclass power has been subject to foreign control.
The opinion of Governor Norman that unless the loan is made Europe will break
is also of great importance to us. Our main reason for going on with the business
would be the heavy responsibility that would rest upon us if our failure to proceed
caused a breakdown.

25

J.P. Morgan, New York, to Morgan Grenfell, London, September 22, 1924 (Lamont Papers),
pages 2-3. The original of the message sent after the September 17 meeting has not been found.
However, the message was repeated in the September 22 message from New York.
26

Ibid., pages 4-5.




55

Replying on September 19 to Morrow's request for his views, the Secretary
of State went about as far as he could to support the loan, while at the same
time he explicitly avoided any obligation to the bankers on the part of the
Government. He saw no reason to question the correctness of Norman's views;
indeed he supported them. The failure of the plan would, the Secretary felt, lead
not only to "chaotic conditions abroad" but to "a feeling of deep despair". He
emphasized that the United States Government could give no pledge or guarantee,
either legal or moral, but he nevertheless expressed the hope that the "American
financiers would see their way clear to undertake the participation [in the loan]
which the world expects and which is believed to be essential to the success
of the Plan".27 In forwarding Hughes' reply to Morgan and Lamont in London,
Morrow commented on September 22: "We think on the whole that this is as
good as we can expect."
The substance of this exchange was apparently made known to Norman, who
in the meantime had had further discussions with MacDonald. These developments Norman reported on September 24 to the Committee of Treasury, which
decided that their remaining doubts and uncertainties should be suspended, that
the Bank of England should proceed with the loan negotiations and, if necessary,
commit itself to a considerable subscription on its own account.28
THE TRANSFER COMMITTEE AND THE AGENT GENERAL.

Although

much

of

the

bankers' energy was directed toward establishing the general economic and political conditions that they felt were prerequisite to the success of the loan, they
were also anxious that the administrative machinery established under the Experts' plan should be staffed with individuals in whom they had confidence.
Particularly was this true of the Transfer Committee, which, as already noted,
was to receive reparation payments from Germany and to transfer them to the
Allies to the extent that the exchanges allowed. The committee was to consist
of the Agent General for Reparation Payments, who was to be its chairman,
and five additional persons, qualified to deal with foreign exchange problems—
one each from Belgium, Britain, France, Italy, and the United States.29 Appoint-

27

Ibid., pages

28

Clay, o p . cit., page 216.

29

Experts' Report, Annex No. 6, page 404.

56




8-9.

ments to the committee were to be made formally by the Reparation Commission and, at one remove, by the governments represented on the commission.
As it turned out, the bankers were to play—and insisted on playing—a major
role in choosing the committee members and particularly the Agent General.
Even before the Experts' Report had reached the United States, Morgan's took
the position that the members "should meet the commendation of the investment
community and should be such as to insure autonomous and sound decisions and
action in the interest of the preservation of the integrity of the new currency and
such as to resist all political pressure from Germany or abroad".30
At an early stage, London bankers apparently had hopes that an Englishman
would be appointed Agent General, but this hope faded as it became clear that
the largest tranche of the loan would be raised in New York. Early in June,
Strong wrote to James A. Logan, the United States unofficial representative on
the Reparation Commission and an aspirant to the post of Agent General:31
I learn that the English have probably accepted the notion that an American will
be necessary in that position. On the other hand, I have heard the view expressed
that it is the key to the success of the plan as well as the insurance of the security
for the loan, and on that account there may be a desire among the bankers to have
some internationally known and outstanding figure, especially some one very well
known in this country, appointed to the job.

After difficult negotiations, the choice of candidates narrowed to two.
MacDonald, Herriot, and Coolidge favored Young while Norman and Lamont
put forward the name of S. Parker Gilbert, who had been Under Secretary of
the United States Treasury in 1921-23, having been appointed to that post when
he was twenty-eight years old, and who had recently become a member of a
New York law firm. As it turned out, Young was reluctant to accept the post for
personal and business reasons and was persuaded to become Agent General only
ad interim. He served for the first two crucial months when the Dawes Plan was
coming into operation, that is, until October 31, 1924, when Gilbert took over
as Agent General on a permanent basis.32

30
J. P. Morgan & Co., New York,
(Young
Papers).
31

to Morgan,

Harjes

and Co., Paris, April

18, 1924, page 4

Strong letter to Logan, June 10, 1924, pages 1-2.

32

Reparation Commission, T h e Execution of the Experts' Plan: Reports of the Agent General for
Reparation Payments (Berlin: The Office for Reparation Payments, September 1, 1924 to August
31, 1925), page 9 (hereafter referred to as Reports of the Agent General).




57

COMPETITIVE ASPECTS: DOLLARS OR STERLING?

THE GOLD DISCOUNT BANK. The negotiations to implement the Experts' Report
were conducted to the accompaniment of a competitive byplay that reflected
the fundamentally changed international economic positions of Britain and the
United States, and more particularly New York's challenge to London's traditional role as the world's major financial center. It is not difficult to understand
how this conflict arose. In the winter of 1923-24, the German currency had been
stabilized only precariously and the outcome of the Experts' work was still most
uncertain. Schacht was in urgent need of external help. For this it was natural
that he should turn to London, the traditional financial center, from which many
of the proposals for the reconstruction of Germany's finances had emanated since
the end of the war. Schacht arrived in London on New Year's Eve in 1923, with
proposals that appealed to Norman. His aim was to employ sterling to stimulate
a revival of German foreign trade and of the German economy generally. He
would do this through the establishment of a Gold Discount Bank whose capital
would be subscribed in sterling by the Reichsbank and by German private interests, and which would be managed by the Reichsbank and would extend shortterm sterling credits to finance German foreign trade.
Before he left London on January 3, 1924, Schacht had obtained Norman's
full support. The Bank of England would extend a £ 5 million credit to the
Reichsbank for two or three years. The proceeds of this loan, together with an
additional £ 5 million in sterling that would be raised in the German market,
would provide the bank's capital. Additional arrangements were made under
which bills of exchange, endorsed by the Gold Discount Bank, could be discounted in the London market up to a total of £ 10 million, it being understood
that such bills would be eligible for rediscount at the Bank of England. Beyond
this, Schacht obtained Norman's support on another important front, i.e., in
opposing a Franco-Belgian scheme to establish an independent central bank
in the Rhineland.33
In their contemporary comments on the Gold Discount Bank proposal, both
Norman and Schacht emphasized that Germany's economic difficulties required
prompt attention and could not wait upon the outcome of the Experts' deliberations. Writing to Strong a few days after Schacht had left London, Norman

33

Norman to Strong, January 7, 1924, and attached memorandum; Hjalmar Schacht, My First
Seventy-Six Years (London: Allan Wingate, 1955), pages 195-202; and Clay, op. cit., page 212.

58




commented that the Gold Discount Bank proposal (about which he enclosed a
detailed memorandum) probably represented the "last chance" of preventing a
complete collapse and that, if the proposal bore fruit, "it would be a grand thing
from every point of view if you could co-operate".34 Strong was probably embarrassed by this invitation because he was unwilling to bypass the Experts. When
he did reply, almost a month later, he avoided any commitment to cooperate.35
Nevertheless, Norman was determined to go ahead with the proposal for the Note
Bank as he then called it. He wrote gloomily about the Experts:36
I have been twice in Paris within the last couple of weeks and from my talks with
Kindersley and others it is clear that there are as many different angles of vision as
there are members on that Committee. As I wrote to you on the 7th January, our
wish here is to enable Germany at once to stabilise her position; that is, to provide
her at once with some sort of a currency which will take care of her foreign transactions and provide foreign valuta for her trade, thus taking the strain off the
Reichsmark and the Rentenmark. If something of this kind is not done pretty soon,
I am afraid the Rentenmark will start to depreciate. This we believe can be avoided
by the setting up of a Note Bank and we are prepared to enable the Reichsbank to
start such a Note Bank; so, in principle, is Vissering and one or two other Central
Banks I expect. The Committee, on the other hand, have so far refused their
blessing or moral support to a scheme of this nature which was put before them by
the new President of the Reichsbank. Your people on the Committee I fancy
wanted to keep the whole German position fluid until they have dealt with it
comprehensively, and if it worsens between now and then, that perhaps won't
prevent their dealing with it. The Belgians have a plan of their own by which they
want to combine the functions of a Note Bank and of a Caisse de la Dette, and
nothing but that will please them. The French are entirely unwilling that any Note
Bank should be started which would tend to mobilize the remaining liquid assets of
Germany outside of Germany and therefore beyond the reach of the Reparation
Commission and yet would make them available for German Trade. So there is a
very confused "kettle of fish" and when the Committee come to leave Berlin—
about the time this letter should reach you—I think we may have to go ahead with
the Reichsbank in spite of the Committee, if we cannot obtain their goodwill, to
the immediate establishment of a Note Bank.

Indeed it was only with some difficulty that Schacht persuaded the Experts
not to block his proposal. Understandably, the Experts were concerned lest
Schacht's scheme prejudice their own as yet unformulated proposals. At the
same time, however, they wished to avoid responsibility for putting a new strain

34

Norman letter to Strong, January 7, 1924, page 2 .
Strong cable to Norman, No. 2, February 5, 1924.
36
Norman letter to Strong, January 30, 1924, pages 3-5.
35




59

on the German currency by killing the Gold Discount Bank plan.37 Hence, they
resolved their dilemma temporarily by issuing on January 23 a qualified public
endorsement of Schacht's proposal, but in private they continued to haggle.
Finally, Schacht publicized the difficulties in a speech in Koenigsberg on February
8. The Experts, he said, had generally approved his scheme, but they had
admonished him that38
it would not suffice to create a gold-bank which would answer to present
circumstances, but that a definite gold-currency should be created in Germany.
I replied to these gentlemen as follows: in principle I agree with you in recognising
that we should create this gold-currency in Germany. I should gladly hear that you
were willing to help us to achieve this result. I am also willing to believe that in
the course of the next few weeks, you will come to a unanimous agreement on
this point in the report you will submit to the Reparation Commission. But it
remains to be seen what results will be obtained from your report. According to
our experience up to the present with regard to nearly all the reparation negotiations, I am authorised to believe that not only weeks, but months will elapse before
your report will result in a unanimous decision being adopted by the Allied and
Associated Governments. We cannot afford to wait so long.**** I hope they
(the Experts) will be reasonable enough not to thwart our efforts to create a
gold-note bank, which will provide us with an instrument making it possible for
us to recover and to work.

Thereafter, a compromise was reached under which the Gold Discount Bank
was established subject to the condition that, if the Experts' proposals were implemented, the new institution would be absorbed by the Reichsbank.39 In fact,
the Gold Discount Bank did become a fully owned subsidiary of the German
central bank in 1925 and, as such, continued throughout the interwar years to
carry on a variety of functions in connection with German foreign exchange
operations.
With the settlement of the problem
of the Gold Discount Bank, the basic question of how the reformed Reichsbank should be required to hold its international reserves came to the fore. Here
THE REICHSBANK'S INTERNATIONAL RESERVES.

37
Hjalmar Schacht,
translated by Ralph

38
39

T h e Stabilization of the Mark (London:
Butler, page 142.

George

Allen & Unwin, Ltd.,

1927),

Leon Fraser's memorandum on the origin of the Gold Discount Bank, April 22, 1927, pages 3-4.
Ibid., page 4. See also letter

of July 7, 1924, to the Organization

Committee

for the

Bank

(Kindersley and Schacht), in which the Reichsbank-Direktorium (Schacht and Bocke) stated their
intention not only to acquire all the shares of the Gold Discount Bank by exchanging them against
new Reichsbank shares but also "to liquidate all transactions of the Gold-Discount Bank".

60




strict logic—if not the economic realities—favored the Americans. Virtually
everyone in authority in both Europe and the United States upheld the theoretical merits of the gold standard and agreed at least in principle that its
rules should be applied to the German central bank. The implications were
clear. The reformed Reichsbank's international reserves should be held either
in gold or in balances that were freely convertible into gold at a fixed price.
Since the United States was at the time the only major country where balances
were thus convertible, Germany would be required to keep in New York any
foreign exchange reserves that it might choose to hold. This logic implied in
particular that, insofar as the Dawes Loan was raised outside the United States,
its proceeds would have to be converted into either gold or dollars.
The logic of gold standard principles combined sometimes with self-interest
to give various influential Americans a rather doctrinaire attitude about how
the Reichsbank's international reserves should be composed. The Americans
on the Experts' Committees, and those like Professor Edwin Kemmerer who
were attached to the Committees' staff, took the view that the establishment
of the reformed Reichsbank on a strict gold basis was essential for the maintenance of Germany's monetary stability and would exert powerful leverage on
other European countries to return to gold.40 Others looked at the question from
a more frankly business viewpoint. Equating dollars with gold, they wanted Germany to hold at least part of the foreign exchange component of its international
reserves in New York. Thus Paul M. Warburg, President of the International
Acceptance Bank, sent Young a cable that applied both to the Gold Discount
Bank (of which Warburg seemed primarily to have been thinking) but also to
the Reichsbank:41
I understand Dr. Schacht is in Paris. Is it not advisable to point out to him
importance of resting his credit bank on stable gold instead of fluctuating sterling?
Incidentally, it would be invaluable advantage for American discount market if as
a result of America's entering the field now substantial portion future German gold
reserve were invested in dollar acceptances. England realizes that and makes
sacrifice to preserve predominance sterling market. Could not America be brought

40

E. W. Kemmerer et al., memorandum, "On the Necessity of a Gold Basis for Germany's Currency", March 19, 1924 (Young Papers). See also R. C. Dawes, The Dawes Plan in the Making
(Indianapolis: Bobbs Merrill, 1925), page 262.
41

Warburg cable to Young, March 14, 1924 (Young Papers).




61

into the first line right now pari-passu with British banks. Pardon this cable but I
am frankly alarmed at the thought that we may miss this unique opportunity for
putting America's discount market on the map and complete our position as world
bankers.

Among the Europeans, whose currencies were still a long way from the desired
return to gold, there was vigorous opposition to views such as those of Kemmerer
and Warburg. To some extent the opposition stemmed from considerations of
national prestige: the French authorities in particular were far from enthusiastic
about any proposal that the United States and the Allies should help Germany
return to gold while the franc continued to fluctuate on a depreciated basis.42
No doubt such prestige considerations were influential also in the British
opposition but, in London, international banking needs and aspirations played
a preponderant role. Despite its difficulties, sterling remained, as Norman
reminded Strong, "very much the exchange of Europe".43 Of the capital that
had fled Germany during the inflation, the bulk was probably held in sterling.
London played a major role in the Austrian and Hungarian stabilizations and
in establishing the Gold Discount Bank, and it was expected that sterling would
play almost as large a role in the Dawes Loan as the dollar. In the City of London,
and particularly in the Bank of England, any proposal that would oblige Germany
to transfer a significant part of its London balances to New York was quite
unacceptable. Such transfers would not only reduce the City's banking business;
far more important, they would push the sterling-dollar rate still farther away
from the $4.867 parity. Britain's international financial aims thus required—as a
minimum—that nothing should be contained in the statutes of the reformed
Reichsbank that would prevent Germany from holding part of its international
reserves in London, even in the period prior to sterling's return to gold. A more
ambitious aim would have been to draw the bank's statutes so that the new
German currency could be based entirely on sterling; there is strong evidence
that this was the aim that Norman actually sought to achieve.44

42

Hamlin Diary, account of talk with Professor Kemmerer, June 9, 1924; S. M. Crocker,
interview
with Kemmerer, May 9, 1925, pages 4 and 12 (Young Papers); and Dawes, op. cit., page 262.

43

Norman

44

letter to Strong, January 30, 1924.

In his letter to Strong of June 16, 1924, Norman expressed his feeling, with reference to the
German stabilization,
that the European currencies were "better stabilised on Sterling than on
Gold . . . ." See also the excerpt from this letter quoted below, pages 65-66, and Hamlin
Diary,
June 9, 1924.

62




Schacht's views on this issue are unclear. It is true that Schacht, like his
central banker colleagues, professed devotion to gold standard principles, but
he also held that Germany could not be expected to apply these principles while
convalescing from inflation and while other major European countries still did
not apply them.45 As to how he believed the Reichsbank should hold its reserves
in the interim, the evidence is contradictory and he may indeed have been quite
opportunistic, taking full advantage of the divergence of opinion between the
European and American experts. Young believed in mid-March 1924 that
Schacht wanted a dollar basis.46 In contrast, Kemmerer told Governor Hamlin
early in June that Schacht strongly favored sterling.47 The truth seems to have
been someplace between these two extremes, and this view is supported by
Schacht's statement at the end of May in a letter to Paul Warburg that "it is our
desire not to base ourselves entirely on the Pound Sterling".48
The Experts' Report reflected these conflicting pressures. In its main body,
it clung to the principle that the Reichsbank note issue should be convertible
into gold but averred that conditions were not yet such that this principle could
be put into practice.49 Hence it recommended only that the Reichsbank's reserve,
which was to be equal to at least one third of its note liabilities and to 12 per cent
of its deposit liabilities, "be held largely in the form of deposits in foreign
banks".50 However, the report's annex on the organization of the bank reflected
the more orthodox views of the American members of the committees. It recommended that the reserves51
be kept in gold bars or gold coin at any office of the bank, and or in the form
of demand deposits made payable in gold or its equivalent at the rates at which the
deposits were made in banks of high standing located in foreign financial centers.

Plainly the Experts had left the issue unresolved. In effect, they passed the

45
46

Hjalmar Schacht, The Stabilization of the Mark, pages
Young cable to Warburg,

March 14, 1924 (Young

47

Hamlin Diary, June 9, 1924.

48

175-76.

Papers).

Schacht to Warburg, May 31, 1924, page 3.

49

Experts' Report, page 357.

50

Loc. cit.

51

Ibid., page 386.




63

question along to Schacht and Kindersley, to whom had been assigned the task
of drafting the central bank's statutes. And so the debate continued. Five weeks
after the Experts' Report was published, the Federal Advisory Council (FAC)
of the Federal Reserve System, of which Paul Warburg was chairman at the time,
issued a statement that could not fail to raise hackles in the City of London.
After discussing various other aspects of the Experts' Report, the FAC launched,
somewhat sententiously, into the question of whether the German currency
would be best stabilized on sterling or dollars:52
It is the question of whether the world is more likely to regain the blessings of
economic stability under the sway of several fluctuating standards of exchange or
by a general return, as speedy as circumstances may permit, to definite relations of
exchanges to gold as the ultimate measure and regulator.
The Dawes Report leads the world to the crossroads in this regard. It provides
for a German note-issuing bank on a gold basis, but leaves the door open to place
it on a sterling basis, and it cannot be denied that there is no small probability of
the latter basis being chosen. In the opinion of the council the sooner Germany
can be placed on a gold or gold exchange basis, the sooner can England, and other
countries also, return to an unrestricted gold standard, while if Germany were
placed on a sterling basis, England—in returning to an unrestricted gold basis—
would have to pull not only her own weight, but that of Germany also. It is
obvious, therefore, that, if the new German bank is placed on the sterlingexchange basis, the world must prepare itself to remain on a basis of exchange
instability for a prolonged period, the end of which can not be foreseen, while the
adoption of the gold (that is, the dollar) basis would accelerate the return to
world-wide stability.

The FAC's pronouncement gave rise to an exchange in which Strong and
Norman set forth their contrasting positions on this issue. Strong seems to have
been somewhat embarrassed; he wrote Norman on June 3 that, on returning
from his European trip,53
One o f the first things I encountered was the report of the Federal Advisory Council
on the D a w e s plan, in which were interjected certain remarks in regard to dollars
and sterling! It sometimes strikes m e that w e are unduly burdened with people in
this world w h o believe that human ingenuity and efforts of imagination can
perform miracles.

52
Federal Reserve Board, Eleventh Annual Report, 1924 (Washington,
Office, 1925), page 281.
53
Strong letter to Norman, June 3, 1924.

64




D . C: Government

Printing

On the substance of the question Strong took a balanced view. He recognized
that, even though sterling was still detached from gold, London had sound claims
as an international financial center because of the "better organization" of its
money market and also because of its "better knowledge of German conditions".
It was therefore natural that Germany should turn to London to satisfy part of
its credit requirements; certainly "no discrimination should be directed against
the London market" in the plans to raise the Dawes Loan. On the other hand,
Strong felt that, if Germany was to have a gold-valued currency, it was to
Germany's advantage "to obtain all possible credits" in New York because
"it is the gold market". Heavy German reliance on New York would also be
to Britain's advantage54
because the credit burden on the London market is one of the present difficulties in the way of a return of sterling to par. Our interests in this subject are mutual
and interdependent and should not be the subject of any controversy or dispute.

Strong made a third point that was only partly valid. He argued that it was
to Germany's advantage "to borrow in the market where the currency is not
at a discount with gold so as to escape the loss which might arise through the
enhancement of the value of sterling vis-a-vis a German currency at par with
gold".55 However, Strong's point would have been valid only if the Reichsbank
expected to spend the London tranche of the Dawes Loan (e.g., to acquire gold
or dollars) at a time when sterling was still at a discount from its gold parity
and was then obliged to service the London tranche in terms of sterling that
was at par with gold.
Norman saw another possibility. Germany could hold the sterling portion of
the loan in the expectation that the British currency would soon rise to par.
He illustrated his point by referring to the Hungarian stabilization which was
then being carried out largely on the basis of borrowing in London:56
Is it necessary under those conditions that Hungary should start off with a Gold
currency? If so she must transfer the proceeds of her foreign loan to New York
in order to avoid the risk of exchange. Is she not justified in basing her currency

54

Strong letter to Norman,

July 9, 1924, page 12.

55

Ibid., page 1.

56

Norman letter to Strong, June 16, 1924, pages 2-3.




65

on Sterling and hoping, as I confidently do, that over a few months or years Sterling
will attain parity with Gold and her currency at the same time will become a Gold
valued currency?

But even while Strong and Norman were debating, the question of the composition of the Reichsbank's reserves was being settled by Kindersley and Schacht.
Their draft of the bank's statutes made significant departures from, and tightened
up, the rather loose recommendations of the Experts' Report and, despite
Norman's efforts to the contrary, did so in a manner more favorable to the
American view than to the British. The proposal to require 12 per cent cover
for the Reichsbank's deposit liabilities was dropped because such liabilities
(which would include deposits held for the account of the Agent General for
Reparation Payments) were expected to show large fluctuations.57 Instead, the
required cover against the bank's note liabilities was raised to 40 per cent from
the recommended 331/3per cent. In addition, no more than one quarter of this
cover could be held in foreign exchange while no less than three fourths was
required to be held in gold.58
These provisions, which were incorporated unchanged into the bank's statute
as it was adopted on August 30, 1924, were clearly disappointing from Norman's
point of view, but there is no record that he continued to struggle for a sterling
basis after mid-June. A variety of considerations may have led him into acquiescence. He may have been concerned lest insistence on his view jeopardize the
success of the Dawes Loan in the United States, where sentiment was so strongly
in favor of a gold basis. Beyond this, he had two sound reasons for expecting
that the Reichsbank would hold its outstanding and accruing sterling balances
at least in the near term. To do so was clearly in Germany's interest, provided
Schacht could assume that Britain was determined to bring the sterling-dollar
rate—which averaged $4.32 in June 1924—back up to the $4.867 parity as soon
as possible. In fact, it is clear that Germany's imminent stabilization had put
Norman under great pressure to commit himself to an early "return", and it is
not inconceivable that he had indicated as much to Schacht. Secondly, it was
equally clear that Schacht would feel no immediate need to employ his sterling

57
Kemmerer
told Crocker on May 9, 1925 that the recommended
he had wanted it set at 40 per cent (Young
Papers).
58

Kindersley and Schacht, memorandum
Bank Law of July 28, 1924, page 11.

66




to the Reparation

12 per cent ratio was too low;

Commission,

July 11, 1924, and Draft

balances to acquire gold. With half of the Dawes Loan scheduled to be floated
in New York, the Reichsbank, initially at least, would be in a position to buy any
gold that it might require for note cover in the United States where authorities
would welcome a reduction in what they considered to be their excessive holdings
of the metal.

FINAL NEGOTIATIONS FOR THE DAWES LOAN

During the week beginning Monday, September 22, 1924, as has already been
seen, Norman and J. P. Morgan reached the decision to begin negotiations for
the German loan. They talked first with Schacht, who came to London, and
then with commercial bankers from other European countries in which tranches
of the loan were to be placed, and also with Dr. Luther, the German Finance
Minister.59 By all accounts these negotiations, in which Young also played a
major part, proceeded smoothly. According to Lamont, they were "neither
prolonged nor difficult"; both Schacht and Luther were "quite evidently prepared
to accept any fair terms which the investment markets deem necessary".60 Of his
part in the negotiations, Schacht wrote in 1927 that he had made "no attempt
to haggle", that he had emphasized the desirability of making the loan "a big
success", and that he considered terms offered to be in line with the prevailing
market conditions.61
While negotiations with Schacht and Luther gave little trouble, the raising of
the European tranches was difficult. A relatively minor problem was that most
of the Continental bankers preferred to have their tranches issued in sterling;
to this the British authorities were agreeable, subject to the understanding that
for two years such sterling bonds would be offered on the British market only
with the consent of the Bank of England.62 The main problem, of course, was
that there was much public opposition to lending money to Germany. In one of
his letters Lamont noted the steady campaign against the loan in several British
newspapers, and commented that British coal miners had protested to the Prime

59

Jay letter to Crissinger,

60

Lamont draft of letter to Clementel, October 6, 1924, pages 1-2 (Lamont Tapers).

61

Hjalmar Schacht, The Stabilization of the Mark, page 185.

October

17, 1924, page 4.

62

Lamont letter to Herman Harjes, September 24, 1924 (Lamont Papers); The Economist, October
18, 1924, page 600.




67

Minister against the entire Dawes Plan. On the Continent, Lamont reported,
only in Sweden and Switzerland was the arranging of the loan going more or
less smoothly. In Paris the bankers were especially balky. To overcome this
resistance, Morgan's told the French Minister of Finance in firm but diplomatic
terms that a $100 million long-term loan, which the French government had
asked Morgan's to arrange in New York, would not be floated unless the French
bankers cooperated.63 Elsewhere, too, the flotation of the loan required strong
leadership. In Britain, Norman summoned together representatives of the banks
and issuing houses and allocated to each the amount necessary in order to underwrite the total London tranche. Similar procedures were apparently followed by
the Belgian, Dutch, and Swiss authorities.64
The arranging of the loan became more hectic as time went on. J. P. Morgan
and Lamont sent a telegram to their firm in New York that is illustrative:65
For the last two weeks aside from the constant work of drafting, etc. our heaviest
efforts have been directed towards getting the Continental shares into line (stop)
If we had taken towards the Banking representatives or the Governments themselves
a take it or leave it attitude we believe that we should have gotten nowhere and
the whole operation would have broken down an outcome far worse to be sure
for Europe than for us yet nevertheless at this stage not to be contemplated (stop)
Each Continental share has had its own particular difficulties due to public sentiment
against Germany as in the case of France and Belgium and exceeding difficult
currency situations as applied to those two countries and Italy (stop) Holland has
had its own financial trouble and heavy difficulties of personnel as well (stop)
Switzerland has had a particular egg of her own to try to hatch out (stop) Sweden
is literally the only country that we have not had to bleed and die for (stop) In
France Italy and Belgium it has not as you suggest been a question of how much
the Governments wanted them to take but rather how much the Governments could
force the reluctant and short-sighted Bankers to take (stop) It has taken a long
time to get the Governments in the right frame of mind (stop) At first Theunis
[Prime Minister of Belgium] and then Herriot at last called the Bankers together
and read the Riot Act to them (stop) The Governments having thus come up to
the scratch and having forced their Bankers to do likewise it became of vital
importance to seize this psychological moment and not to let it escape probably
never to return (stop)

After three weeks of negotiations in London, Young returned to Paris with
an agreement on the loan and with a set of stipulations that he was to present

63

Lamont

draft of letter to Clementel,

October

6, 1924, pages 4-5 (Lamont

Papers).

64

Ibid., page 3.

65

Morgan and Lamont cable to J. P. Morgan & Co., New York, October 8, 1924 (Lamont Papers).

68




to the Reparation Commission for adoption; these made it absolutely clear that
the loan would be a first charge on Germany's resources, ahead of reparations
and all other payments. Pierre Jay related66 how on October 10 Young presented
these stipulations:
The Reparation Commission which, as I wrote you before, had not heretofore been
very prompt, to say the least, in making its decisions has during the past month been
called upon by Mr. Young to make promptly a large number of decisions on very
important matters. The one they were called upon to make on the 10th without
much previous opportunity for discussion, declaring for the absolute priority of
the loan and dealing with a lot of other details, with the bankers at the other end
of the telephone wire waiting to send cables that evening to the members of the
underwriting syndicate in America was an especially large order . . . . Mr. Young's
handling of the situation was masterful and the evening wound up with a complete
decision on the part of the Commission, a telephone message to the London
bankers, and their dispatches of the necessary cables to the underwriting syndicate.

Then one last hurdle had to be surmounted. The bankers required a resolution
by the Transfer Committee to the effect that service of the loan would be given
priority and absolute right of remittance irrespective of the effect on the German
exchange rate. On October 13, Young contacted as many of the members of
the committee as he could reach and obtained their informal approval67 In due
course the committee, at its first meeting on October 31, formally adopted a
resolution to this effect.68
With this final question settled, the loan was issued in New York on October
14 and in London and other European centers the following day (see table). It
was underwritten at 87 in New York and871/2in Europe. The issue price was 92,
its maturity was twenty-five years, final redemption being scheduled for October
15, 1949 at par in European markets and at 105 in the United States. It carried
a 7 per cent coupon, thus yielding approximately73/4per cent to maturity.69 The
loan was rapidly oversubscribed in New York, London, and most of the Continental markets and quickly rose to a premium in early trading.70
Much has already been written about the economic recovery of Germany after

66

Jay letter to Crissinger,

67

Young cable to Crocker for Sterrett, October 13, 1924 (Young Papers).

68

Reports of the Agent General, September 1, 1924 to August 31, 1925, page 10.

69

Bergmann, op. cit., page 280; an "Official Statement" dated October 10, 1924 (Lamont Papers).

70

The Economist, October 18, 1924, page 600, and October 25, 1924, page 638.




October

17, 1924, page 4.

69

THE DAWES LOAN

Market

Currency
of tranche

Nominal amount
of tranche*
(in millions of
specified currencies)

Net proceeds
In millions of
In millions of
tranche currency* gold marks

United States

Dollars

110.0

95.6

400.6

Britain

Sterling

12.0

10.3

192.9

Belgium

Sterling

1.5

1.3

24.4

Holland

Sterling

2.5

2.2

40.6
48.2

France

Sterling

Italy

Lire

Sweden

Kronor

Switzerland

Sterling
Swiss francs

Germany

Sterling

3.0

2.6

100.0

86.3

15.7

25.2

21.8

24.3

2.1

38.9

11.9

9.5

2.36
15.0
0.36

0.3

Total

5.9
801.0

* The difference between the nominal amount of each tranche and the net proceeds reflects commission charges and the discount at which the bonds were issued.
The exchange rate at which the net proceeds of the New York tranche were converted into reichsmarks was 4.1896 marks to the dollar, the noon buying rate for cable transfers in New York on
October 14. Parity for the gold mark was 4.198 marks to the dollar. The sterling tranches were
converted at 18.8138 marks to the pound. This rate reflects the 8 per cent discount from sterling's
gold parity at which the British currency was being quoted in the London market at the time the
loan was floated.
Source: Reports of the Agent General, September 1, 1924 to August 31, 1925, page 11.

the flotation of the Dawes Loan,71 and part of that story enters into the account
that follows of central bank cooperation during the later 1920's. Here it suffices
to add that the immediate outcome of the Dawes Plan realized, and perhaps even
surpassed, the highest hopes of the bankers who negotiated the loan. Confidence
in the economic future of Europe revived and political tensions eased. The
despair of 1923 gave way in 1925 to hopes that a new era had begun.

71
See, for example, J. W. Angell, op. cit., Chapters III-IX; Hjalmar Schacht, The Stabilization of
the Mark (London: George Allen & Unwin, Ltd., 1927), Chapters VII-IX; Horst Mendershausen,
Two Postwar Recoveries of the German Economy (Amsterdam: North-Holland Publishing Company, 1955); and C. T. Schmidt, German Business Cycles 1924-1933 (New York: National Bureau
of Economic Research, Inc., 1934).

70




5. Britain's Return to Gold
INTRODUCTION

The stabilization of the German currency was followed in the spring of 1925 by
the long awaited return of sterling to its prewar gold parity. The conditions were
unique and compelling. The temporary reparation settlement had brought
Europe a surge of renewed optimism. This, combined with the mild 1924 recession in the United States, led to a greatly enlarged outflow of American capital.
The United States balance of payments, which had been in surplus during
the three previous years, swung into deficit during the latter part of 1924, when
the capital outflow was heaviest, and this helped to buoy the European exchanges.
In these circumstances sterling might have climbed back to its prewar parity
under its own power, but the climb was accelerated by the known desire of the
authorities in Britain, the United States, and elsewhere for an early return to
gold. Speculative balances that had fled Europe in the troubled years before the
German stabilization now left New York and, encouraged by the prospect for
sterling appreciation and by relatively high British interest rates, settled in
London.
Strong and Norman showed a high order of statesmanship in grasping this
favorable opportunity to achieve their international financial objectives. Their
subsequent difficulties arose from a failure fully to understand that the return
was only the beginning of their tasks and that the maintenance of sterling's link
with gold would involve problems far more complicated than those encountered
in bringing about the return. The complex reasons for this failure are discussed
in other chapters1 and need only be summarized here. Among them were ( 1 )
the illusion that once the major countries had stabilized their currencies in terms
of gold, the functioning of the international economy would be regulated automatically, as was thought to have been the case under the nineteenth century
gold standard; (2) the related faith that national economies were best regulated
by the free play of market forces, a belief that discouraged the search for policy
instruments to hasten the international adjustment process and to foster steady
economic growth; ( 3 ) the priority accorded by the monetary authorities to
national policy objectives, a priority that restricted the extent to which the

1

See especially Chapters 3 and 9.




71

authorities put into practice the gold standard-free market principles they professed and that resulted in the shifting of most of the burden of international
adjustment on to the economies of the deficit countries; (4) the primitive state
of balance-of-payments statistics and analysis, which left the authorities and the
public badly informed about major imbalances in the international economy,
about the slow pace of adjustment, and about the extent to which such imbalances were being financed by short-term borrowing; and (5) the failure to
develop international cooperative techniques to cope with severe financial strains
between major countries; central bank cooperation remained merely a fair
weather instrument. But these were lessons that became clear only after the
tragic events of the twenties and thirties had run their course.

THE OBJECTIVES OF ANGLO-AMERICAN COOPERATION

In working together to restore sterling to its 1914 gold parity, Strong and Norman
were fostering national objectives about which they were quite clear. In Strong's
view, the stabilization of sterling was the key to the elimination of fluctuating
exchange rates which had been a "withering influence" on international trade
ever since 1914.2 Expanding world trade, he felt, would bring with it a growing
demand for United States products, particularly agricultural output, and thus
help to alleviate the chronic difficulties of the American farmer. He also saw in
sterling's return to gold an opportunity for New York to gain in stature as an
international financial center since, under the cooperative arrangements to facilitate the return, New York would for a time maintain its interest rates below
London's and so become the more attractive center in which foreigners could
borrow.3 Finally he hoped that the return would reduce the threat of inflation in
the United States by reversing the inflow of gold that had proved so troublesome
to the Federal Reserve during the early twenties.4
For his part, Norman saw the return to gold as the culmination of Britain's

2

United States Congress, House Committee on Banking and Currency, Stabilization, Hearings, 69th
Congress, 1st Session, on H.R. 7895 (Washington, D. C, 1927), Strong testimony, April 12, 1926,
page 507.
3

Hamlin Diary, May 22, 1924.

4

Strong letter to Mellon, May 27, 1924, page 6; Norman cable to Bank of England, No. 16,
paragraph 6, January 6, 1925.

72




postwar efforts to restore its position as a major center of international trade and
finance. In this view, he was supported by the influential report published in 1918
by the Cunliffe Committee (so called because it was headed by the then Governor
of the Bank of England, Lord Cunliffe), by the overwhelming weight of informed
British opinion, and by the country's three major political parties. There was little
dissent from the view that the long-term economic advantages to Britain would
greatly offset the temporary adjustment difficulties that might arise. And this view
was echoed both by the Committee on the Currency and Bank of England Note
Issues,5 whose report was presented to the government some weeks before the
decision to return was announced, and by the press after the announcement.6
In their private discussions, the British authorities were more specific about the
aims they hoped to achieve by the return to gold and about the costs. Sir James
Grigg, who was Winston Churchill's private secretary at the time, recalls a dinner
at which some of the major proponents and opponents of the return presented
their arguments to the Chancellor.7 Keynes and McKenna, who represented the
opponents, argued that sterling would be seriously overvalued in terms of the
dollar, that the achievement of external equilibrium would consequently entail
a deflation of British wages and prices, and that such a deflation would bring
prolonged strikes and a permanent contraction in some of the country's heavy
industries. To avoid these difficulties Keynes, at least, favored the maintenance
of domestic wages and prices at the existing level and the continuation of a fluctuating exchange rate for sterling.
The proponents' case was argued by Lord Bradbury, chairman of the Committee on the Currency and Bank of England Note Issues. He held, according to
Grigg's account, that the return to gold would save Britain from living in a fool's
paradise of false prosperity and would force the export industries to become competitive. Bradbury rejected the suggestion that the return could be at a lower
rate than the prewar parity, contending that this would cause a shock to confidence and thus endanger Britain's international reputation merely for an ease-

5
This committee, which was appointed by the government in June 1924 to explore the problems
surrounding Britain's return to gold, included Austen Chamberlain, a former Chancellor of the
Exchequer, Sir (later Lord) John Bradbury, then a member of the Reparation Commission, Gaspard
Farrer of Barings, Sir Otto Niemeyer of the Treasury, and Professor A. C. Pigou of Cambridge
University.
6

R. S. Sayers, "The Return to Gold, 1925", Studies in the Industrial Revolution (London: Athlone
Press, 1960), page 315.

7

Sir James Grigg, Prejudice and Judgment (London: Jonathan Cape, Ltd., 1948), pages 182-84.




73

merit that would be small and ephemeral. The old export industries, he held,
would probably need to contract in any event, and the country's best course was
to develop its newer and more sophisticated manufacturing industries and its
international banking, insurance, and shipping services.8
Bradbury's view faithfully reflected the weight of opinion in the City of London
and particularly at the Bank of England in the early months of 1925, and to a
considerable extent this view continued to prevail during the difficult years ahead.
Norman testified in this same spirit before the Macmillan Committee in March
1930. With reference to maintaining the gold value of sterling, he declared that9
the disadvantages to the internal position are relatively small compared with
advantages to the external position.... the whole international position has preserved for us in this country the wonderful position which we have inherited, which
was for a while thought perhaps to be in jeopardy, which to a large extent, though
not to the full extent, has been re-established. We are still to a large extent international bankers. We have great international trade and commerce out of which
I believe considerable profit accrues to the country; we do maintain huge international markets, a free gold market, a free exchange market—perhaps the freest
almost in the world—and all of those things, and the confidence and credit which
go with them are in the long run greatly to the interest of industry as well as to
the interest of finance and commerce.

The central bankers were motivated, not only by expectations of the advantages that would be derived from the return to gold, but also by their fears about
what the course of developments might be if sterling were not stabilized at the
old parity. Bradbury's belief that any other course would jeopardize London's
future as an international financial center has already been noted. Norman and
Strong were even more apprehensive. They felt, according to a memorandum
that Strong wrote early in January 1925, that the consequences of a failure of
Britain to resume gold payments would be "too serious really to contemplate".
Such a failure, Strong continued,10
would mean violent fluctuations in the exchanges, with probably progressive
deterioration of the values of foreign currencies vis-a-vis the dollar; it would prove
an incentive to all of those who were advancing novel ideas for nostrums and
expedients other than the gold standard to sell their wares; an incentive to Govern-

8

Ibid., page 183.

9

Macmillan Evidence, Vol. 1, questions 3332-33, pages 212-13.

10

Strong memorandum of January 11, 1925, page 4.

74




merits at times to undertake various types of paper money expedients and inflation;
it might, indeed, result in the United States draining the world of gold with the
effect that, after some attempt at some other mechanism for the regulation of
credit and prices, some kind of a monetary crisis would finally result in ultimate
restoration of gold to its former position, but only after a period of hardship and
suffering, and possibly some social and political disorder.

THE STRATEGY OF COOPERATION

As it became clear that the circumstances of 1924-25 would be propitious,
Strong and Norman facilitated sterling's return to gold by adopting a strategy
of cooperation that dealt with relative prices and interest rates and with stabilization credits. Regarding the first, the central bankers understood the limited
extent to which monetary policy could influence prices. Nevertheless, they
stressed the need to realign British and American prices from the distortions induced by the war and postwar inflations. Strong himself estimated that the prices
of internationally traded commodities in Britain in the spring of 1924 were on
the average still about 10 per cent higher than those in the United States, calculated from a common 1913 base, and he held that this gap would need to be
more or less eliminated before Britain could return to gold.11
That spring, after he had spent some weeks with Norman in London, Strong
wavered from the position that he had previously adhered to so firmly,12 and was
to reassert thereafter, on the question whether the United States authorities might
condone or even facilitate a rise in American prices in order to help Britain
return to gold. He wrote to Mellon at the end of May 1924 that, since Britain
was suffering from poor trade and heavy unemployment, it would "be difficult
politically and socially for the British government and the Bank of England to
force a price liquidation in England beyond what they have already experienced".
Hence, he argued, the burden of the readjustment of prices "must fall more
largely upon us than upon them". More specifically, there would need to be "some
small advance in prices here and possibly some small decline in their prices".18
As events developed, Strong does not seem to have persuaded his Federal
Reserve colleagues that the United States should accept the larger part of the

11

Strong letter to Mellon, May 27, 1924, pages 11-12.

12

Supra, page 31.

13

Strong letter to Mellon, May 27, 1924, pages 11-12.




75

adjustment burden, and he soon reverted to his earlier views, perhaps partly because of talk among his colleagues suggesting that he was too much under
Norman's influence.14 In any case, the record of the Strong-Norman discussions
in New York early in January 1925 contains no reference to the desirability of
higher American prices. Strong was extremely cautious. He told Norman:15
it was my belief, and I thought it was shared by all others in the Federal Reserve
System, that our whole policy in the future, as in the past, would be directed towards
stability of prices so far as it was possible for us to influence prices, but that it was
a matter on which no engagements could be made, as we must be free at all times
to reduce or increase discount rates and to establish our open market policy so as
to adequately meet domestic developments.

For his part, Norman was content to stress that the maintenance of the gold
standard in Britain would depend on "a certain degree of stability" in the United
States and that, in returning to gold, the British authorities were working on the
assumption that the United States intended neither "to embark upon a deliberate
policy of deflation" nor to "permit the development of any considerable inflation
of prices".16
The second major element in the strategy involved cooperation between the
Federal Reserve and the Bank of England to maintain lower interest rates in the
United States than in Britain.17 Both Norman and Strong hoped that a significant
interest rate differential between London and New York would help to close the
gap between British and American prices but, recognizing the multitude of other
influences that act on prices, they placed their main reliance on the hoped-for
effects of interest rate differentials upon international capital flows. Strong felt,
to be sure, that so long as sterling was not stabilized, exchange risks would minimize the movement of United States short-term capital to London, but he hoped
that foreign long-term borrowing might be diverted from London to New York.
Even here, however, Strong was not overly optimistic because he questioned how

14
On May 5, 1924, Charles Hamlin noted in his diary that Governor Harding of the Boston
Reserve Bank had commented to him that he was "certain that the movement for lower rates at
N.Y. was inspired by Gov. Strong, now sick in Gov. Norman's house in London; that Norman
wanted inflation in United States to put us more nearly on a parity w. Gt. Britain".
15

Strong memorandum of January 11, 1925, page 5.

16

Loc. cit.

17
Strong letter to Jay, April 23, 1924, page 7; Hamlin Diary, May 7, 1924; and Strong letter to
Mellon, May 27, 1924, pages 11-12.

76




far Americans would place their capital abroad when there were safe and profitable investment opportunities in the United States. He also questioned how far
the City of London would be willing to forego long-term lending business that
traditionally had been theirs.18 On the latter point Strong was close to the truth.
Despite a change in interest rate relationships that made New York the cheaper
center in which to borrow, the British authorities found it necessary in November
1924 to impose an informal control over the flotation of new foreign issues in
the London capital market.19
The third part of the strategy of cooperation would come into play—as Strong
wrote in his May 27, 1924 letter to Mellon—when the basic conditions for the
return to gold had been achieved or were at least within reach: the United States
would need to provide credits to Britain. At this early stage, Strong had two
types of credit in mind. At first, he wrote:20
large private credits must be opened in this country for use by the Bank of
England or by the British Government or British banks in order to steady the rate
of exchange and gradually work it to a higher level corresponding to the international price parity, and hold it there, until the time arrives to actually announce
a plan of resumption based upon such adequate credits in this country as will
insure the final recovery of sterling to par.

Although Strong was not explicit, it seems clear that he expected the Federal
Reserve to participate only in the credits that would be granted upon the
announcement of the plan to return to gold.
While Norman certainly agreed with Strong about the need for American
credits to support the return to gold, others in Britain were opposed or at least
doubtful. When Norman was in New York in January 1925, he received cables
indicating that several Bank of England directors felt that, if Britain was to
return to gold, it should do so solely on the basis of its own resources. They
felt that the acceptance of foreign credits would cast doubt upon the determination and capacity of the country to establish sterling at its prewar parity and to

18

Strong letter to Norman, July 9, 1924, pages 3-4.

19

W. A. Brown, Jr., England and the New Gold Standard, 1919-1926 (New Haven: Yale University
Press, 1929), page 224; Royal Institute of International
Affairs, T h e Problems of International
Investment (London, 1937), page 134.
20

Strong letter to Mellon,




May 27, 1924, page 12.

77

maintain it there "at all costs".21 The Report of the Committee on the Currency
and Bank of England Note Issues also expressed concern lest the proposed
credits be employed to soften the economic disciplines that Britain was expected
to accept when it returned to gold. The report agreed that the credits would help
to discourage speculation against sterling and would contribute to the creation of
confidence, but it warned that, if the authorities felt it was necessary to arrange
for credit in the United States,22
we feel strongly that recourse should not be made to it unless and until substantial
gold exports have taken place and are already producing their normal effects on
the monetary situation at home, and in the event of the credit being actually drawn
upon, the amount drawn should, until it has been repaid, be treated from the point
of view of the Bank of England's monetary policy as equivalent to a corresponding
loss from its own reserves.
Unless these precautions are taken, borrowing abroad will, as has again and
again happened when it has been resorted to as a remedy for exchange difficulties,
merely aggravate the mischief which it has been applied to cure.

TIMING

The timing of sterling's return to gold was inherent neither in the objective itself
nor in the strategy that was adopted. Rather it was set by the same sweep of
events that made it possible to discuss the return as a practical objective. Indeed,
given the favorable circumstances of 1924-25 and the intensity with which the
return was desired by the British authorities—to say nothing of Strong and central bankers elsewhere—the task of resisting the pressures that were pushing
sterling back to gold would have been difficult indeed.
It is true that Norman himself took a cautious line in his discussions with
foreign central bankers, but within his own councils he advocated a prompt
return. As early as May 1924, he expressed his view23 that the wartime power
under which the government had embargoed gold should not be renewed when
it expired at the end of 1925—which was tantamount to recommending that the

21

Lubbock

cable to Norman,

No. 54, January 10, 1925.

22

Report of the Committee on the Currency and Bank of England Note Issues (London:
Stationery Office, February 1925), page 7, paragraphs 30 and 31.
23

Sir Henry Clay, Lord N o r m a n (London:

78




Macmillan

& Co., Ltd., 1957), page 149.

H.

M.

return should be during that year. On the other hand, his 1924 correspondence
with Strong reads as if he were in no hurry at all.24 As late as October 16, 1924,
Norman wrote that attempts to stimulate market support for sterling through
official proclamations of confidence in a return to gold "would have been difficult,
and perhaps dangerous". It was his feeling that "however wearisome the pace
has been, we have been wise so far to hurry slowly" and that, in view of the unexpected change in government from Labor to Conservative, "we must 'wait and
see' " what course would now be taken. Purely as a guess, he suggested that the
return might come "either at the end of 1925 or at the end of a somewhat later
year, say 1927".
Norman's display of caution seems in part to have been merely a reaction to
pressure from Strong, who did not conceal his wish that Britain should return
to gold as quickly as possible. Indeed it was natural that Norman should take
such a line with Strong, because he could not commit himself until his government had made the needful decisions and because his bargaining position was
improved if he let Strong be the one to urge him to do what he wanted to do
anyway.
However, his caution probably also reflected the attitude of some of his colleagues at the Bank of England. When he was in New York in January 1925,
Norman received a long cable from Lubbock, his Deputy Governor, who urged
that even after sterling had reached parity the authorities should delay any final
commitment to the gold standard until they had had an opportunity to test
whether sterling could be maintained at that level "by the natural play of the
market force of supply and demand without resort to any artificial aids", by which
he presumably meant the restrictions on the flotation of new foreign issues in London and the use of the proposed credits from the Federal Reserve and J. P.
Morgan. He warned that it would be a mistake to fix a date for the return "before
conditions warrant it", for the risk "would be too great and the consequences of
failure too grave for us to recommend it".25 In subsequent cables,26 Lubbock
indicated that three influential members of the Bank of England's Court of
Directors agreed with the substance of this cable and that Sir Otto Niemeyer

24

See especially Norman

25

Lubbock cable to Norman, No. 54, January 10, 1925.

26

Lubbock cables to Norman, Nos. 55, 57, and 59, January 10, 12, and 13, 1925.




letter to Strong, June 16, 1924.

79

of the Treasury had similar reservations.
But such counsels of caution were overwhelmed by the pace of events. Sterling
was pushed back to parity in the spring of 1925, partly by the movement of
speculative balances in markets dominated by the conviction that the return was
imminent and partly by fear among Britain's financial leaders lest the opportunity
to restore sterling to a preeminent place in international finance might never
return. Prestige plainly played a large role. Sweden, whose central bank governor
had urged Norman to lead the way back to gold, had grown tired of waiting and
in March 1924 returned alone.27 When the acceptance of the Dawes Plan in the
spring of 1924 made it seem likely that the German currency would be stabilized
in terms of gold before sterling, the British authorities were struck with consternation.28 Strong was quite aware of British sensitivity on this score and, when
the Hungarian currency was stabilized during the summer of 1924, he twitted
Norman by writing that "the question uppermost in my mind is whether sterling
is not now rather far behind in the procession".29 Subsequently, in early December, Strong warned Norman of the rapidly changing monetary conditions in the
United States and of upward pressures on our interest rates. Norman did not
need to read between the lines to know that conditions in the United States favoring sterling's return could not be expected to last indefinitely.
As time went on, the pressure grew. In January, while Norman was still negotiating in New York, the South African government announced that its currency
would return to gold effective July 1, 1925.30 Similar action was urged by the
monetary authorities in Australia whose currency had moved to a premium in
terms of sterling during the winter of 1924-25.31 Then during the spring word
came that a return to gold was also being planned by Holland, Switzerland, and
perhaps other Continental countries.32 Noting this when he announced Britain's

27

Clay, o p . cit., page 142; W. A . Brown, Jr., T h e International Gold Standard Reinterpreted,
1914-1934 (New York: National Bureau of Economic Research, Inc., 1940), Vol. I, page 323.

28

Strong

letters

29

to Jay, April

Strong

letter to Norman,

30

Lubbock

cable to Norman,

23-28,

1924.

July 9, 1924, page 4.
No. 59, paragraph

7, January 13, 1925.

31

L . F. Giblin, T h e Growth of a Central Bank: T h e Development of the Commonwealth Bank of
Australia (Melbourne: Melbourne University Press, 1951), pages 11, 24-26.

32

Report of the Committee o n the Currency and Bank of England N o t e Issues, page 6, paragraph 21.

80




return at the end of April, Churchill commented that "we could not have afforded
to have remained stationary while so many others moved".33 Subsequently, Churchill was more explicit about the reasons behind the timing of the return. After
mentioning the seasonal strength of sterling during the spring and the expiration
at the year-end of the authority to embargo gold, he noted that34
if we had not taken this action the whole of the rest of the British Empire would
have taken it without us, and it would have come to a gold standard, not on the
basis of the pound sterling, but a gold standard of the dollar.

IMPLEMENTING THE STRATEGY

The favorable circumstances of 1924-25 and the compatibility of the objectives
of Strong and Norman combined to simplify the task of cooperation between the
Federal Reserve and the Bank of England. In this respect, the discussions on
sterling's stabilization contrasted sharply with the difficult German negotiations
of 1924.35 Before the Dawes Loan could be floated, the recuperative strength
and economic potential of Germany had to be assessed, distrust among the participating countries alleviated, and conflicting objectives reconciled. In the end,
it was agreed that the stabilization of the German currency called for long-term
and large-scale financing and that means must be provided whereby future pressures on the exchange rate could be relieved if necessary. Thus recognition of the
difficulty of the German problem brought international agreement on means that
were appropriate for its solution. But the very success of this international effort
helped to make sterling's problem appear simpler than it really was; it encouraged
the British authorities to adopt policies from which they could not turn back
without precipitating a major international crisis, and to adopt them with only
limited and rather short-term assistance from the United States.
THE FEDERAL RESERVE AND MORGAN CREDITS. The lines of credit from the Federal Reserve and a syndicate headed by J. P. Morgan & Co. which supported the stabilization of sterling were negotiated with very little difficulty. The

33

Parliamentary Debates (Great Britain, House of Commons),

34

The Economist, August 8, 1925, page 222.

35

Vol. 183, April 28, 1925, column 57.

Cf. Chapter 4.




81

principal negotiators were Strong and Norman themselves, their major tasks being
to ensure that the use of the credits should conform to their need to maintain
control over their respective money markets and that official press releases on
the credits should stress that neither party had made any commitment that would
limit the conduct of his monetary policy.36
The course and outcome of these negotiations have been detailed elsewhere37
and need not be repeated here. Suffice it to say that the improvement in the international economic balance during the second half of 1924 made it unnecessary
to consider such private credits to Britain as Strong had originally envisaged.38
The purpose of the credits that were in fact negotiated was not to help push sterling up to parity but rather to strengthen confidence that sterling would be maintained at parity once that level had been attained and Britain had again committed itself to the free export of gold. For its part, the Federal Reserve Bank
of New York arranged to grant the Bank of England a $200 million line of credit
that could be drawn upon at any time during the two years beginning May 14,
1925, either in gold or in dollars. Interest was to be charged only on amounts
actually drawn, the rate being set at 1 per cent above the New York Reserve
Bank's rate of discount on ninety-day bills, with a minimum of 4 per cent and a
maximum of 6 per cent, except that a rate equivalent to the discount rate would
be charged if the latter should be raised above 6 per cent. The terms on the
Morgan line of credit to the British government were identical except that Morgan's charged an additional commitment fee of 11/4per cent on the full amount
of the credit during the first year and, if no drawing were made, half that rate
in the second year.39 Reluctant to bear this fee, the British Treasury cut down
the amount of the line from the $300 million originally desired by Norman and
Strong to only $100 million.40
The central bankers' concern to ensure that the use of the credits should not
impair their monetary control is fully understandable. Each wanted to be free

36

Strong letter to Norman,

April 15, 1925, page 2.

37

L. V. Chandler, Benjamin Strong, Central Banker (Washington,
1958), pages 308-21.

38

Strong

Institution,

Supra, page 77.

39

D. C : The Brookings

cable to Norman,

No. 60, April 4, 1925; Parliamentary Debates (Great Britain, House of

Commons), May 4, 1925, column 623, and May 5, 1925, column 815.
40

Chandler,

82




o p . cit., pages

316-17.

to adjust monetary policy in accordance with his assessment of his money
market's changing needs, and both were concerned because the credits were
large relative to the central banks' holdings of marketable securities. In the
United States the total of Federal Reserve "bills bought" and government securities averaged a little less than $650 million in April 1925, while the holdings
of the Open Market Investment Account averaged only $248 million, compared
with maximum possible drawings on the credits of $300 million, or $500 million
under the original plans. Clearly, a Federal Reserve policy that aimed to keep
a tight rein on member banks' reserves could have been seriously complicated,
or even vitiated, if the Bank of England could draw on its Federal Reserve credit
to make payments to the Street; or, to take the opposite extreme, a policy that
aimed to pump reserves into the banking system could be checked if the Bank
of England were to use the Morgan credit to buy gold at the Federal Reserve.
By the same token, the credits could clearly be so employed as to reinforce
Federal Reserve policy or have a neutral effect on the money market. From
the beginning of the negotiations for the credits, it was therefore understood
that the Bank of England would consult with the Federal Reserve both about
the amounts to be drawn and about which of the credits was to be used.41 By
the time the negotiations were complete, Strong had obtained a large measure
of control, it being agreed by the Bank of England and the British Treasury that42
In the event of recourse being desired to either of the Credits established in
America, the Treasury or the Bank of England shall use the particular Credit
which the Governor of the Federal Reserve Bank shall from time to time select.

For the Bank of England the problem and its solution were similar, at least
so far as the Federal Reserve credit was concerned.43 It was agreed that, upon
making a drawing, the Bank of England would credit the account of the New
York Reserve Bank with the sterling equivalent, the amount so credited to be
adjusted daily to reflect fluctuations in the cable rate of exchange.44 However,

41

Strong memorandum of January 11, 1925, page 7.

42

Norman letter to Strong, April 22, 1925.

43

No relevant

information

44

is available

regarding

the Morgan

credit.

Norman letter to Strong, April 8, 1925; Minutes of the Board of Directors of the Federal
Bank of New York, April 23, 1925; Strong and Moreau conversation, August 23, 1926.




Reserve

83

in the event that the full amount were drawn, the total so deposited would come
to some £ 4 1 million, equivalent to two fifths of the average government and
other securities holdings of the Bank of England's Banking Department in April
1925 and to almost three fifths of average bankers' balances at the bank.45 In
order to reconcile the management of such Federal Reserve balances with Bank
of England monetary policy, it was agreed that these balances could be invested
by the Federal Reserve only in specified types of sterling commercial bills, and
that the timing and magnitude of any bill purchases would be arranged in a
mutually satisfactory manner between the two central banks. It was also agreed
that earnings from such commercial bills would be credited at the current rate
of exchange against any interest payable in dollars by the Bank of England on
its drawings.46
Such was the harmony between Strong and Norman at this time that these
were the only significant conditions on the credits. Other conditions were minor.
Drawings could be in gold or dollars and were similarly repayable; service was
guaranteed by the British Treasury.47 Strong asked for and received assurance
that Norman would remain governor at least for the next year and probably for
the next two years.48 With a view to strengthening monetary control, he also
asked assurance that no increase in the British Treasury currency note issue be
allowed without the approval of the Bank of England, but Norman wrote that
formal compliance with this request would provoke a serious public controversy,
and Strong was content to let the matter drop on receiving Norman's assurance
that he would attempt to persuade the Committee on the Currency and Bank of
England Note Issues to include in its report a recommendation that would
meet the request.49 Actually no such recommendation was included in the committee's report which was not made available to Parliament and the public until
April 28, the day Churchill announced the return to gold. At this point, when
it would have been futile to press the matter, Strong agreed with as good grace

45

Macmillan Report, page 302.

46

Norman letter to Strong, April 8, 1925, pages 2-3; Strong letter to Norman, April 20, 1925; and
resolution passed at the Board of Directors' meeting, Federal Reserve Bank of New York, April
23, 1925.
47

United Kingdom

48

Strong letter to Norman,

49

Norman

84




Treasury Minute, May 14, 1925, signed by Sir Otto
January 15, 1925; Norman

letter to Strong, April 20, 1925.

Niemeyer.

letter to Strong, February

10, 1925.

as possible to postpone discussion of control of the Treasury note issue until a
more opportune time.50
As events developed, neither of the credits was drawn upon. How far they
contributed to the confidence in sterling that developed immediately after the
return is a moot point, but there is firm evidence that foreign balances that had
moved to London before the end of April 1925 remained for several months
thereafter. On the other hand, Strong's hope that the credits might be employed
to reduce the Federal Reserve's "excessive" gold holdings was never realized,
despite at least one hint to Norman that the purchase of gold in New York would
be welcome.51 In the end, no efforts were made to renew the credits and they
expired on schedule, May 14, 1927, a few weeks before another major cooperative effort of a different character was launched.52
The monetary policies of the United States
and Britain went through several phases between May 1924, when the cooperative effort for sterling's return to gold began in earnest, and the latter part of
1926, when the return appeared to have been firmly consolidated.
May-November 1924. The first phase extended over the six months ended
November 1924, a period during which the United States economy reached the
trough of, and started to recover from, a mild recession and when United States
monetary policy gave more support to sterling than it was able to give at any
time thereafter. During the winter of 1923-24, the Federal Reserve had begun to
ease monetary conditions but, as late as April, New York money rates were still
well above the corresponding London rates, as they had been during the two
preceding years. That spring the Federal Reserve embarked on a further major
easing of monetary conditions. The New York Reserve Bank's discount rate was
reduced in three 1/2 percentage point steps—on May 1, June 12, and August 8,
1924—from 41/2per cent to 3 per cent, where it was held until February 1925
(Chart 2 ) . At the same time Federal Reserve credit was pumped into the banking
system. The Open Market Investment Committee (OMIC), at its meetings between April 22 and July 16, authorized Government securities purchases totaling
COORDINATION OF MONETARY POLICY.

50

Strong letter to Norman, April 30, 1925.

51

Strong letter to Norman, May 21, 1925.

52

Infra, page 119 f.




85

$300 million, and the Federal Reserve average holdings of such securities actually
increased $264 million in the six months ended November 1924. Over the same
period, the Federal Reserve bought some $188 million in bills and $84 million
in gold. These purchases, along with those of Government securities, swamped
the effect on member bank reserves of a moderate outflow of currency into
circulation, and permitted further reductions in member bank borrowings at
the Federal Reserve, a significant expansion in member bank reserves, and a 14.5
per cent increase in net demand deposits of weekly reporting member banks—
the most rapid deposit expansion to occur over any six-month period in the
years 1924-31. 53
While United States monetary conditions were becoming easier, the Bank of
England maintained firmness in the London money market. The total of London
clearing banks' cash and balances at the Bank of England was permitted to expand only slowly in the six months ended November 1924, and total clearing
bank deposits remained virtually unchanged. However, with unemployment in
Britain continuing high, Norman was unable, as he wrote Strong in mid-June,
to find a satisfactory excuse to raise the discount rate above the 4 per cent at
which it had been held since mid-1923,54 but he was able to help widen the
spread of London money market rates above New York through Bank of England
operations that were reflected in a 1/2 percentage point rise during July 1924
in the London rate for three-month bankers' bills.55 The rise in the London
market rates continued in August and subsequent months, three-month bankers'
bills fluctuating in the 3.72-3.79 per cent range for the remainder of the year,
about 0.70 point above their April-June level.
These changes brought a major shift in the interest rate relationships between
New York and London. Market discounts on bankers' bills in New York, which
had been about 1 percentage point above London in April 1924, fell to 1 1/2
points below London in July and were still more than 1 point below in November.
For other money market rates the relationship changed correspondingly, although

53
For the sake of complete accuracy, it may be noted that the percentage increases in net demand
deposits in the two overlapping six-month periods ended in October and November 1924 were virtually the same.
54
55

Norman letter to Strong, June 16, 1924.
Norman cable to Strong, No. 95, July 19, 1924.

86




CHART 2

SHORT-TERM INTEREST RATES
NEW YORK AND LONDON, 1924-26

Note: Bankers' acceptances are monthly averages of rates on three-month
bankers' acceptances.
Sources: Board of Governors of the Federal Reserve System, Banking and
Monetary Statistics (Washington, D. C, November 1943); R.G.Hawtrey,
A Century of Bank Rate (London, 1932).

the change in call loan rates involved only a narrowing of New York's spread
above London.86 In the capital markets, yields on United States long-term
Government bonds, which had been only 0.17 point below corresponding yields
in Britain in April 1924, declined to 0.46 point below them by July and, contrary to the movements in the two money markets, this spread did not narrow
significantly during the rest of the year. Even more important, long-term funds
became increasingly available in New York, whereas in London, as already

56
Paul Einzig, The Theory of Forward Exchange (London: Macmillan & Co., Ltd., 1937), chart
opposite page 276.




87

noted, informal controls were adopted to curtail the outflow of long-term capital.
November 1924-May 1925. In the second major phase, which extended from
November 1924 to May 1925, United States monetary policy changed from
ease to moderate restraint and, in order to maintain the interest rate relationship
with New York that had been established during the previous half year, Bank
of England policy was further tightened. The initial change in Federal Reserve
policy was made at the October 24, 1924 meeting of the OMIC, which moved
away from the buying of Government securities toward a more neutral posture
under which either purchases or sales of Governments were authorized, depending on market conditions. In the event, the recovery from the 1924 recession
proceeded apace, and the Federal Reserve's holdings of Government securities
were reduced $227 million in the six months ended May 1925. Changes in other
factors—primarily money in circulation and the gold stock—were largely offsetting during the six months. Thus, the overall effect of Federal Reserve operations was a withdrawal of $212 million from member bank reserves, all but
$32 million of which was recouped, however, through increased member bank
borrowings from the central bank. At the same time, United States interest rates
were again moving upward. The upturn in money market rates had begun as
early as August 1924, but not until February 27, 1925 was the discount rate
of the Federal Reserve Bank of New York increased—by 1/2 point to 31/2per
cent, at which level it was to remain for almost a year.
Strong gave Norman full advance warning of these changes in United States
monetary policy.57 Early in December 1924 he cabled Norman about the possibility of a 1/2 point increase in the New York Reserve Bank's discount rate,
and inquired whether, if such an increase became necessary, Norman would
prefer to advance his discount rate first.58 Norman's reply was that the Bank
of England would increase its rate by 1 point if New York went up 1/2 point
and that he would prefer to follow New York "and so appear to have our hands
forced by you".59 Later, after Strong had cabled that the increase in the New
York rate was imminent, Norman confirmed his intention and added that he
expected that a further increase in the Bank of England discount rate to 6 per

57

Strong letter to Norman,

58

Strong cable to Norman, No. 4, December 5, 1924.

59

Norman

88




cable to Strong,

November

4, 1924; Strong letter to Norman,

No. 39, December

8, 1924.

December

2, 1924.

cent would be required if free gold payments were resumed in April.60 As events
developed, the Bank of England did raise its discount rate to 5 per cent on
March 5, just a week after the New York Reserve Bank's increase, but the
further increase to 6 per cent was made unnecessary by the unexpected strength
of sterling in the spring and summer following the return.
The tightening of Bank of England policy was also reflected in its market
operations. In the first phase, the cash base had been allowed to expand, albeit
at a much slower rate than in the United States, but now that the expansion of
the United States cash base had been stopped, the British cash base had to be
contracted. Bank of England holdings of government and other securities declined 10 per cent in January-May 1925 and, along with a modest contraction
in discounts and advances, far more than offset various expansionary factors,
leading to a 7 per cent decline in the central bank's deposit liabilities.
Monetary developments in the two countries reflected these shifts in policy.
In the United States, the sharp expansion of May-November 1924 was
halted. Weekly reporting member banks' reserves declined slightly in the six
months ended May 1925, and their total deposits dropped 1/2 per cent. In Britain
the contraction was sharper, London clearing banks' cash and balances at the
Bank of England and total deposits each declining 2 per cent in the six-month
period. The greater degree of stringency in Britain kept market rates of interest
there well above those in the United States. On both three-month bankers' bills
and long-term government bonds, the spread of the London rates above the corresponding New York rates was wider in May 1925 than the previous November.
Even in the call loan market, in which London rates were normally relatively low,
the sharp rise in London rates reduced the margin in favor of New York to only
0.28 point in May 1925, compared with 1.71 points in November 1924.
The achievements of the year ended May 1925. The changes in United States
and British monetary conditions in the year ended May 1925 were accompanied
by the achievement of the international financial goals sought by Norman and
Strong. The sterling-dollar rate, which had fluctuated in the $4.30's during May
1924, moved in late July into the $4.40-$4.50 range, where it continued through
October, after which it pushed on, sometimes falteringly, until at the end of
April 1925 it had reached the lower part of the $4.84-$4.90 range within which

60

Norman cable to Strong, No. 71, February 24, 1925.




89

the rate could fluctuate under the old gold parity. Throughout the month following the return to gold, the rate continued to strengthen and averaged just
over $4.85. The appreciation of sterling appears the more remarkable when
it is remembered that the British authorities were building up dollar balances
with which to meet war debt payments to the United States, some $92 million
having been accumulated during the fall and early winter of 1924 to meet the
payment due in mid-December61 and an additional $166 million in the early
months of the following year to meet the payments due in June and December
1925. 62
Gold flows also shifted significantly. The United States continued to be a net
importer of gold through most of 1924, as it had been during most of the postwar
period, but the net inflow slackened markedly after midyear. Reflecting the
change, the average monthly increase in the United States gold stock fell from
$42 million in January-May 1924 to only $12 million in June-November.
Beginning in December the gold stock itself dropped sharply for the first time
in four years, the decrease averaging $28 million monthly in the six months ended
May 1925.63
No corresponding movements are reflected in the Bank of England gold
holdings, because until the end of April 1925 its statutory buying and selling
prices for gold remained out of touch with the market. Its holdings of gold coin
and bullion did increase by £ 2 7 . 6 million in the year ended April 29, 1925,
but £.27 million of this was attributable to the transfer to the bank of gold held
by the British Treasury as backing for its currency note issue.64 In the months
that followed, the bank did begin to gain a considerable volume of gold which,
as we shall see, caused Norman no little difficulty and embarrassment.
In the price field, the central bankers' goals were also achieved, although the
process was a slow one. The spread of British wholesale prices above those in
the United States which had been 11.5 per cent in April 1924, remained

61

Brown,

62

Parliamentary Debates (Great Britain, House of Commons),

England and the N e w G o l d Standard, 1919-1926, page 220.

63

Board of Governors of the Federal
D. C., 1943), pages 536-37.
64

Reserve

System,

April 28, 1925, column

56.

Banking and Monetary Statistics

Parliamentary Debates (Great Britain, House of Commons), April 28, 1925, column 56.

90




(Washington,

obstinately wide throughout the remainder of the year but thereafter narrowed
to only21/2per cent by May 1925 (see Chart 3 ). The realignment of export prices
in the two countries was also elusive until March 1925, the problem during 1924
being that British export prices tended to move in sympathy with those of the
United States. Indeed, the spread of British export prices above those of the
United States (Federal Reserve indexes, 1913 = 100) was only 2 per cent in
April 1924, but widened to 5 per cent by the autumn of that year. Not until
January 1925 did this spread begin to narrow, but then the change came rapidly
because of a sharp but temporary upturn in United States export prices and
a sharp downturn in those of Britain. By March 1925 British export prices
were 3 per cent below those of the United States and, despite the subsequent
sharp downturn in United States export prices, the margin in favor of Britain

CHART 3

WHOLESALE PRICE INDEXES
UNITED STATES AND UNITED KINGDOM, 1924-26
1913=100




Note: Data on wholesale price indexes for the United States were compiled by the
Bureau of Labor Statistics, and those for the United Kingdom by the Board of Trade.
Source: Federal Reserve Bulletin.

91

had widened slightly by May.65
Analysis of the achievement. It is already clear that the attainment of sterling
parity, the reversal of the gold flow to the United States, and the realignment of
United States and British prices resulted from a host of forces. Strong and
Norman were aware of the complexity of the conditions in which they were
operating, and at least in 1924, when the outcome of their efforts was still
uncertain, they were little inclined to exaggerate the power of central bank policy
to achieve their objectives. As already noted, their greatest hopes lay in the
employment of monetary policy to shift long-term foreign borrowing from London
to New York. In this they were highly successful, and there can be little doubt
that this shift contributed significantly both to the strengthening of sterling and,
to a lesser degree, to the reversal of the gold flows to the United States. New
capital issues in London on behalf of foreign countries (excluding British
dominions and colonies) dropped to a monthly average of only £2.8 million
in the seven months ended May 1925, about half the rate in the first ten months
of 1924.66 At the same time the outflow of capital from the United States was
greatly enlarged. Although the United States current account surplus was considerably larger in both 1924 and 1925 than it had been in 1923, the change
was far more than offset by a rise in long-term capital outflows to $700 million
in 1924 and $570 million in 1925 from only $45 million in 1923.67 The nominal
value of new capital issues in the United States on behalf of foreign countries
was four times larger in the year ended June 1925 than in the year before.68
The available balance-of-payments data also support Strong's expectation that
little United States short-term capital would move to London, but since no
figures are available to show changes in short-term capital movements within
the halves of the year, the evidence is not conclusive. In any case, the enlargement of capital outflows was sufficient to change the United States overall
balance of payments from a $266 million surplus in 1923 to a surplus of

65
Various issues of the Federal Reserve Bulletin. The data for the period up to March 1925 are
collected in E. V. Morgan, Studies in British Financial Policy, 1914-25 (London: Macmillan & Co.,
Ltd., 1952), page 363.
66

Midland Bank, N e w Capital Issues, 1926.

67

United States Department of Commerce, United States in the World Economy
D. C.: United States Government Printing Office, 1943), Appendix Table 1.
68

Midland

Bank, Monthly Review, August-September

92




1927.

(Washington,

only $28 million in 1924 and to a deficit of $40 million in 1925.
This change in the outflow of United States long-term capital and in the
United States balance of payments stemmed from shifts: (1) in the relative
attractiveness of Europe and the United States as investment outlets, (2) in
the availablity[av ilabilty]of funds in the London and New York capital markets, and
(3) in relative interest rates. The first element was clearly of overriding importance. The huge rise in long-term capital outflow from the United States
could hardly have developed without the simultaneous brightening of European
investment prospects and the dampening of those in the United States. And
the very recession that was associated with this dampening contributed to a
slackening in the demand for funds in the United States and—along with Federal
Reserve policy—to the greater availability and reduced cost of credit here.
Apart from the shift in long-term borrowing from London to New York,
other factors also supported the strengthening of sterling. Speculation on the
appreciation of sterling was almost certainly a major influence; it was widely
noted by contemporary observers69 and seems to have involved primarily foreign
balances that had fled to New York in the uncertain months and years before
the spring of 1924 and thereafter moved back to Europe, and especially to
London, as the prospects on the other side of the Atlantic improved. The speculative movement seems to have developed in both the spot and forward markets.
At a time when forward sterling should theoretically have been at a discount,
the speculation was reflected in the disappearance of the discount on three-month
forward sterling during the late autumn and the appearance thereafter of a
premium; this premium was maintained almost until sterling's return to gold70
and, combined with the differential in short-term money rates, it gave a substantial covered incentive to move funds from New York to London (Chart 4).
A partly fortuitous influence that helped support sterling during the autumn
and winter of 1924-25 was the exceptionally heavy demand from continental
European countries for sterling commodities. Much of the Continent's demand,
which originated in the need to make up the deficiencies of a bad harvest in

69
See especially Report of the Committee o n the Currency and Bank of England Note Issues,
page 7, paragraph 28; Reginald McKenna in the Monthly Review of the Midland Bank, FebruaryMarch 1925, page 3; and J. A . Schumpeter, Business Cycles (New York: McGraw-Hill, Inc., 1939),
Vol. 11, pages 725-26.
70

Einzig, op. cit., page 256 and chart opposite page 276.




93

CHART 4

ARBITRAGE OPPORTUNITIES BETWEEN
NEW YORK AND LONDON, 1924-26
Three-month bankers' acceptance rates; monthly averages

Sources: Board of Governors of the Federal Reserve System, Banking and
Monetary Statistics (Washington, P. C, November 1943). Data on forward
exchange calculated from figures in Paul Einzig, The Theory of Forward
Exchange(London, 1937), Appendix 1.

1924 and to rebuild inventories depleted during the war and early postwar years,
was of course directed to the United States. However, a large share also went
to such British dominions as Australia, New Zealand, and South Africa. The
magnitude of the expansion in demand is indicated by the 275 per cent increase
in the physical volume of wheat imports from Australia by five major Continental
countries71 in the year ended June 1925. The volume of Continental imports
of wool, jute, and rubber from sterling countries also expanded significantly.72

71
Italy, France, Belgium, Germany, and the Netherlands.
of Australia (Melbourne,
1927), page 648.
72

Official Yearbook of the Commonwealth

Brown, England and the N e w G o l d Standard, 1919-1926, page 194.

94




At their peak in the early months of 1925, prices of these commodities had about
doubled from a year earlier.73
The reversal of the flow of gold to the United States was also the result both
of international financial cooperation (and the accompanying large outflow of
long-term capital from the United States) and of fortuitous developments. Most
clearly the result of the former was the large export of gold from the United
States to Europe in the six months ended May 1925, particularly $87 million
sold to Germany during that period. Some $50 million of this was acquired by
the Reichsbank, using part of the proceeds of the New York tranche of the
Dawes Loan and United States currency that had been circulating in Germany
and was turned over to the bank after the stabilization of the mark began to
take hold during 1924.74
However, the larger part of the reversal of the gold movement to the United
States is attributable to factors that were connected only remotely, if at all, with
the efforts of Strong and Norman. The export of $27 million in gold by the
United States to Australia in January-April 1925 occurred because booming sales
abroad of Australian wheat and wool had pushed the Australian pound to a
substantial premium over sterling. The constellation of exchange rates made it
advantageous, as a pure arbitrage transaction, to acquire gold in the United
States, ship it to Australia, and sell the Australian pound proceeds in London.75
Similarly, the export of some $67 million of gold from the United States to India
in the ten months ended May 1925 is attributable to a favorable monsoon that
brought bumper crops in 1924, a strengthening of India's balance of payments,
and speculation that the rupee might appreciate in terms of gold.76
Increased Indian demand for gold led, not only to imports directly from the
United States, but to a cessation of British gold exports to the United States.
Indian demand was of course normally a major element in the London gold
market, but its expansion in 1924 and early 1925—combined with some rise
in the off-take of continental Europe—was so great as to absorb fully the residual
supply that had previously been sold to the United States.

78

"Commercial History and Review", The Economist, 1924 and 1925.
G. L. Harrison letter to G. B. Winston, December 4, 1924; Crissinger letter to Strong,
December
26, 1924.
75
Brown, England and the N e w Gold Standard, 1919-1926, page 194, and The International Gold
Standard, Reinterpreted, 1914-1934, Vol. 1, pages 369-70; Giblin, o p . cit., pages 24-26.
76
Brown, England and the N e w Gold Standard, 1919-1926, pages 186-90.
74




95

The movements of United States and British export prices during the year
ended May 1925 were the product of complex forces in which monetary policy,
as both Strong and Norman recognized, played only a relatively small part.77
In the fluctuations of United States export prices, the influence of Federal Reserve
policy was swamped during the summer of 1924 by expectations that agricultural
produce (which constituted about half of United States exports) would be
moving into rather weak markets. It was swamped again by unexpectedly heavy
European demand, which pushed up agricultural prices during the autumn and
winter and which slumped thereafter, leading to an almost equally sharp price
decline. It is true that, in facilitating the large expansion of United States longterm loans, Federal Reserve policy made it easier for Europe to buy American
exports and thus acted to support United States export prices. However, even
in the absence of increased American lending, Europe might have been forced
to curtail spending in other countries, or to draw down its international reserves,
in order to make good some part of its agricultural deficiency by importing
from the United States. At least part of the rise in United States export prices
that developed during the winter of 1924-25 might, therefore, have occurred
even if there had been no easing of Federal Reserve policy.
Fluctuations in British and American export prices were closely associated.
They declined together during the summer of 1924 and, as the central bankers
noted with some dismay, they rose together during the following autumn and
winter. Thus, the upsurge in international commodity prices more than offset the
downward pressure on sterling prices that was being exerted both by monetary
restraint in Britain and by the appreciation in the sterling exchange rate. Only
when international commodity prices turned sharply downward in March 1925
did the deflationary pressures gain the upper hand to bring the export prices of
Britain and the United States into alignment.

CONSOLIDATING STERLING'S POSITION

The consolidation of sterling's position on the basis of the $4.867 parity was a
task at least as difficult as the return to gold. For the latter, the central bankers

77
Strong letters to Norman, July 9, 1924, page 2, and November 4, 1924, page 4; Norman letter to
Strong, October 16, 1924, pages 2-3; and Strong letter to Lubbock, September 10, 1924.

96




had chosen an occasion when economic conditions were highly favorable. Thereafter, the need was so to strengthen Britain's fundamental international economic
position that sterling's link to gold could be sustained when the economic climate
turned from fair to unsettled and even to foul.
As events developed, international economic conditions were to remain "fair"
for some years after the return. In this, the American prosperity of the midtwenties played a major part. The recovery from the 1924 recession was rapid
and continued, after the initial surge, at a moderate pace and with only minor
interruptions until the onset of another mild recession late in 1926. At the same
time, United States imports soared, far outpacing the modest rise in exports.
The outflow of long-term capital from the United States—attracted by high
interest rates and continued favorable prospects abroad—remained large, and
the overall balance of payments moved definitely into deficit which amounted
to $362 million in 1926. However, foreigners chose to increase their short-term
dollar assests[as ets]by no less than $455 million in 1926. Consequently, the gold
outflow that had developed during the winter of 1924-25 was again reversed,
and by the end of 1926 the United States gold stock was only slightly lower than
it had been before the outflow began in December 1924.
The combination of rising prosperity in the United States and the renewed
inflow of gold confronted the Federal Reserve with a problem that was handled
quite easily in 1925 and 1926 but became progressively more difficult toward
the end of the decade. In mid-1925, as in later years, the authorities hoped to
stretch out the country's prosperity, to prevent the development of undue speculation and other inflationary pressures that could lead to boom and bust. Lacking
any fiscal policy instruments, they could only curb such domestic pressures by
tightening monetary policy which would, however, tend to strengthen the dollar
against foreign currencies, increase the gold inflow, and so complicate the
Federal Reserve's control of member bank reserves. Strong anticipated this
problem toward the end of the summer of 1925 during which a similar unwanted
inflow of gold to Britain had led to pressure on the Bank of England to ease
its monetary policy. Writing to J. H. Case, Deputy Governor of the Federal
Reserve Bank of New York, Strong stated:78

78

Strong letter to Case, August 25, 1925.




97

that up to a certain point we should make every effort to accommodate our policy
to theirs [the Bank of England]. If they feel it necessary in meeting domestic
conditions to reduce their rate, then, of course, they must do so, but should not
necessarily expect us to go down with them if our domestic situation would render
it perilous to do so. And, in fact, that is our situation at home. We cannot now afford
a rate reduction, and the only question is whether we can afford to stay at our
present level as the best we can do, or whether, notwithstanding a reduction in
London say to 4%, it may be necessary for us to go up to 4%.
Of course you understand that the consequence of an increase in our rate, in
the face of rate reductions among the gold standard countries of Europe, will be
some transfers of funds to the United States by bankers generally, and by the banks
of issue specifically; and that we cannot then escape further shipments of gold.
Our earning assets are now nearly $300,000,000 above what they were a year
ago, and of our earning assets over $200,000,000 are Government holdings under
the control of the [Open Market Investment] committee and about $100,000,000
are Governments not under committee control. If, therefore, we are now obliged
to pursue a policy of higher rates and it does result in gold imports, the Federal
Reserve System must definitely face a period of liquidation of earning assets at
least equal to gold imports, and it may be that the sale of securities must be extended beyond the committee's holdings and take in some of the holdings of the
individual banks.
I am hoping that there will be no important change in the situation prior to the
date of the Governors Conference. At that time I shall be glad to explain to our
colleagues everything that I know about the situation abroad and endeavor to
induce them to a complacent state of mind about earnings during this temporary
period of readjustment. If we are not successful in doing that, we are going to
have a struggle with an inflation and a boom sooner or later, unless, indeed, they
have an up-set in Europe which would act as a deterrent.

As events turned out, the inflationary pressures with which the Federal Reserve System had to deal in the one and one-half years beginning May 1925
were quite moderate and, despite the shortage of open market ammunition, the
authorities were able to cope with them without undoing the achievements of
1924-25. Seasonal influences apart, the System kept member bank reserves
under pressure. The expansionary effects of the gold inflows, of net purchases
of commercial bills, and decreased currency circulation were more than offset
by a net decline in Government securities holdings. About $100 million of
reserves was thus withdrawn from the banking system in the eighteen months
ended in November 1926 which marked the downturn into the 1926-27 recession. However, in meeting the credit demands of the expanding economy,
member banks moderated the pressures on their reserves by borrowing more
than $200 million from the Federal Reserve. Even so, the money supply grew
98




less rapidly than the national product and interest rates tended to move upward.79
Those Federal Reserve Banks whose discount rates were not already at 4 per
cent moved to that level in November 1925 except for the New York Bank
which made the1/2per cent increase in January 1926. All the Reserve Banks remained at the 4 per cent level throughout 1926, except New York whose rate
was set at 31/2per cent for about four months ended mid-August when it was
again increased to 4 per cent. By the end of 1926, rates in New York on ninetyday bankers' acceptances were close to 4 per cent and call money was above
5 per cent, these rates being 0.6-1.3 percentage points higher than they had
averaged in May 1925.
The gradual tightening of monetary conditions in the United States was not
Norman's major problem in the months immediately following the return. His
problem was to enforce a monetary policy that would foster the economic adjustments that still had to be made by Britain if sterling was to be maintained on
gold in the long run. This objective, as Norman indicated toward the end of
May 1925, required a gradual contraction of credit and the continuation of
interest rates in Britain at higher levels than in major centers abroad.80 However,
the relatively attractive London interest rates combined with optimism about the
British outlook encouraged foreigners not only to maintain the sterling investments they had made during the winter of 1924-25 but to increase them. For
this and other special reasons, the Bank of England gained between May and
August 1925 almost £ 9 million in gold which acted to expand commercial
bank reserves and, more importantly, also stimulated public demands that
monetary policy be eased.81
Norman had a number of courses open to him in dealing with this perplexing
problem. He could attempt to offset the expansionary effect of the gold inflow
through open market sales, but here he faced the same dilemma as Strong. To
the extent that such open market sales kept the money market taut, they only

79
Milton Friedman and Anna J. Schwartz, A Monetary History of the United States, 1867-1960
(Princeton: Princeton University Press, 1963), page 711; Harold Barger, Outlay and Income
in the United States, 1921-1938 (New York: National Bureau of Economic Research, Inc., 1942),
pages 114 and 179.
80

Norman

letter to Strong, May 26, 1925.

81

Norman letter to Strong, May 26, 1925; Strong letter to Jay, August 1, 1925; Strong cable to
Jay, No. 6, August 1, 1925; and "Commercial History and Review of 1925", The Economist,
February 13, 1926, page 8.




99

encouraged the flow of foreign balances to London and of gold to the Bank
of England. The pressures for the relaxation of monetary policy would be increased. Other courses were suggested by Strong. One, which bore fruit in
later years, was that the bank should take advantage of the demand for sterling
to build up dollar balances that could be employed to meet the seasonal needs
that were expected to develop in the autumn.82 Another was that gold gains
should be taken, not by the Bank of England, but by the Treasury which could
employ them as backing for its note issue and so prepare for the day when
the notes issued by the Treasury and the central bank would be amalgamated.83
Both proposals had the advantage that they would prevent reserve gains from
being reflected in the bank's published gold holdings and so might help reduce
public pressure for easier money. However, neither proposal was taken up at the
time—perhaps because the Treasury itself seems to have been exerting heavy
pressure on the bank to relax its monetary policy.84
The upshot was that Norman took the more straightforward and—as it now
appears—more daring course. He followed what were then believed to be the
traditional rules of the game, and he did so with some vigor in order, in the end,
to return to the tight money policy that he felt was necessary. He was of course
aware that substantial foreign balances had been placed in London, and he
anticipated that a good part of them would be withdrawn if his money market
rates declined significantly relative to those offered in other centers, particularly
New York. If the reduction in London money rates coincided with the normal
autumn pressure against sterling, a substantial amount of gold would be lost,
the Treasury and the public would become concerned, and the gold standard
tradition would justify a sharp advance in the bank rate and the reimposition
of a strict monetary discipline. Norman had employed the seasonal strength in
the country's payments to foster the return to gold during the spring; he would
now employ its seasonal weakness to enforce the tight monetary policy that he
felt was necessary to support sterling in the long run.85

82

Strong cable to Norman, No. 91, May 23, 1925.

83

Strong letter to Harrison,

84

Grigg, o p . cit., page 193; Clay, o p . cit., page 293.

August

7, 1925.

85
Norman letters to Strong, May 26, August 21, and November 23, 1925; Strong letters to Case,
August 1 and 5, 1925, and to Harrison, August 7, 1925. See also Strong cable to Jay, No. 6, August
1, 1925.

100




The first phase of Norman's ease-squeeze tactics was executed during the
summer of 1925. The expansionary effects of the gold inflow and of a moderate
decline in note circulation was only partly offset by decreases in the bank's
holdings of securities and in discounts and advances, with the consequence that
bankers' and other deposits at the Bank of England rose 4.1 per cent in MayAugust. On August 6, the Bank of England's discount rate was reduced 1/2 per
cent to41/2per cent. The London rate for three-month bankers' bills averaged
3.94 per cent in August, 0.65 point below the May peak. With New York
money rates rising, the margin of London bill rates above the corresponding
rates in New York narrowed to an average of only 0.69 point in August from
1.40 points in May.
In September, the bank began to lose gold but the deflationary effect on the
cash base was more than offset by an increase in the bank's holdings of securities
and a modest increase in discounts and advances, so that bankers' and other
deposits rose another 2.8 per cent. Rates on bankers' bills in London continued
to decline, narrowing the margin above New York to only 0.18 point. On
October 1, the bank rate was again reduced by 1/2 per cent to 4 per cent and,
in the following two weeks, London rates on bankers' bills averaged 0.19 per cent
below New York.86 On a comparison of call money rates, London averaged
1.56 per cent below New York during October.87
Developments in the forward exchange market followed these changes in
interest rate relationships with considerable faithfulness. The discount on threemonth forward sterling which had been 0.82 per cent in June 1925, had
disappeared during September and had changed to a premium of 0.49 per cent
in early October. Although on a comparison of bankers' bills there remained a
modest incentive to move covered funds from New York to London, on a
comparison of call money rates there was a much stronger incentive in the
opposite direction, particularly during September and October when the movement of a significant volume of balances from London to New York was in fact
recorded.88
As these movements of funds gathered momentum, they added to the seasonal

86

The Economist, October

87

Einzig, op. cit, page 501.

88

Federal Reserve Bulletin, June 1926, pages




3 and 10, 1925; Federal Reserve Bulletin, October

1925.

377-78.

101

weakness of spot sterling. In the early summer the rate had hovered at or only
slightly below the $4.867 parity. It began to slip late in July. By mid-September
it was at $4.84%, approximately the gold export point to New York, and thereafter dipped still further to fluctuate at or a little above $4.84. In October some
$40 million in gold was shipped from London to New York, and for October
and November as a whole the Bank of England showed a £ 13 million decline
in its holdings of gold.
This very substantial gold loss, which pushed the Bank of England's holdings
below the £150 million minimum recommended by the Cunliffe Committee,
set the stage for the second phase of Norman's tactics, i.e., the restoration of
the 5 per cent bank rate. Particularly during October, the bank reduced its
securities holdings and so reinforced the squeeze that the gold losses were putting
on the market. Discounts and advances expanded, short-term interest rates rose
sharply, and the market began to expect an early bank rate increase. In these
circumstances, market tenders for Treasury bills declined and the Bank of
England found it necessary to come to the rescue to cover the Treasury's
short-term borrowing requirements. At the end of November, as the Deputy
Governor told the Macmillan Committee,89
the amount which the Bank had to provide in order to cover the required amount
of tenders was very substantial . . . . finally we were compelled, simply in order
to get the bills taken up, and to avoid our being driven into a very difficult position
by reason of the large additions of credit that we were having to create, to raise the
rate from four per cent to five per cent.

The bank's decision put a severe strain on its relations with the Treasury.
In the late evening of Wednesday, December 2, Norman called at the Treasury
to inform the Chancellor, as a matter of courtesy, of the impending bank rate
increase. About 11:20 a.m. the next day, Churchill telephoned to protest vigorously against the increase. He threatened to tell the House of Commons that he
had not been consulted and that the increase was against his wishes. Noting that
such action on the part of a Chancellor was without precedent, Norman stood
on his legal rights and for some time thereafter relations between the Governor
and Churchill were decidedly cold.90

89

Macmillan Evidence, Vol. 11, question

90

Grigg, o p . cit., page 193; Clay, op. cit., page 293.

102




7596, pages

179-80.

Norman's tactics were vindicated by a decided strengthening of sterling.
Immediately following the bank rate increase, the spot rate rose to $4.85 and
from January through August 1926 it was, with a few exceptions, at or above
$4.86. Even when the adverse seasonal pressures set in during the autumn, the
rate did not fall to the gold export point. Early in 1926 the Bank of England
ceased to lose gold, and during the entire second half of that year its holdings
remained somewhat above £ 150 million.
Behind these superficially satisfactory figures was the beginning of the adjustment process. It was painful although perhaps not more so than the authorities
had originally anticipated. The Bank of England enforced a strict monetary
discipline. Reductions in its holdings of securities and in discounts and advances
more than offset the expansionary effects of the rise in its gold holdings and
of a continued reduction in note circulation, with the result that bankers' and
other deposits at the Bank of England declined 6.2 per cent in the year ended
November 1926. Although the London clearing banks partially offset the squeeze
on their cash base by working to a lower cash ratio, the total of currency circulation plus clearing bank deposits was on the average virtually unchanged in 1926
from the year before and was slightly lower than in 1924.91 Pressure to reduce
wages and other costs was continued, leading to an especially long and embittered
strike in the coal mines and a nine-day general strike in May 1926. Unemployment continued high. The overall gross national product was only slightly larger
in 1926 than in 1924, but a beginning was made to changing its composition
away from the old staple trades such as coal, steel, and shipbuilding where output declined toward such technically advanced and expanding industries as those
producing rayon, motor vehicles, and electrical apparatus.92 Equally significant,
British export prices continued to decline in 1926 but—allowing for the difficulties
of international price comparisons—not enough to eliminate more than a little
of the competitive advantages held by Britain's major commercial rivals.93
While the adjustment process was moving slowly forward, the financing of
the deficit in Britain's current and long-term capital accounts remained pressing.
The problem was particularly serious in 1926. Although imports were somewhat

91
London and Cambridge Economic
(London, 1965), page 16.
92
Ibid., page 6; G. C. Allen, British
mans, Green and Co. Ltd., 1959), pages
93
League of Nations, Review o f World




Service,

The British Economy: K e y Statistics 1900-1964

Industries and Their Organization (4th ed.; London:
24 and 31.
Trade (1930), pages 63-65.

Long-

103

smaller in value than in the two previous years, exports dropped very much more
sharply, not only because of reduced sales of coal abroad, but also because of
smaller exports of textiles, iron, and steel.94 Beyond this, the lifting of the controls
in November 1925 had been followed by a spurt of new capital issues on foreign
account which rose by 27 per cent in 1926 from the year before. All told,
Britain's deficit on current and long-term capital account widened to almost
£ 100 million in 1926, more than double the 1925 deficit.95
Bank of England policy certainly facilitated the financing of this deficit by
maintaining attractive conditions in the London market. However, a renewed
flight from the French franc was probably the main factor in the large positive
errors and omissions item that balanced the accounts. While London rates for
three-month bankers' bills remained well above the corresponding New York
rate throughout 1926 and there was a significant incentive to move funds on a
covered basis both at the beginning of the year and at the end, the actual flow
of United States funds does not seem to have been significant. In relation to
most Continental centers, on the other hand, London money market rates were
relatively low, yet such was its attractiveness on other counts that London continued to be a preferred center in which to hold short-term balances. Overseas
borrowers invested in London a considerable proportion of the proceeds of
their loans, some of which were floated for the express purpose of acquiring
sterling reserves.96 Moreover, the proceeds of some of the loans raised in New
York by Australian, German, and other foreign borrowers were transferred to
London to be held there at least temporarily.97 Other ordinary commercial
transactions probably also contributed to the buildup of balances in London
during 1926, but the single most important factor was almost certainly the
renewed flight from the French franc. The flight began in October 1925 and
reached its climax in July 1926, by which time the foreign exchange value of
the franc had been reduced to half the level at which it had been stabilized

94

Board

of Trade,

H. M. Stationery

A c c o u n t s Relating to T r a d e and Navigation of the United K i n g d o m

(London:

Office, 1926), page 6.

95

T. C. Chang, Cyclical Movements in the Balance of Payments (Cambridge: Cambridge University
Press, 1951), table opposite page 144; A . E. Kahn, Great Britain in the World Economy (New York:
Columbia University Press, 1946), page 126.
96

Kahn, op. cit., page 166.

97

Clay, o p . cit., page 237; Strong letter to Bachmann,
Norman, October 21, 1927; and Nadler memorandum
2-3 (Federal Reserve Board
Papers).

104




November
18, 1927, page 1; Strong letter to
to Goldenweiser,
December 27, 1926, pages

during the summer of 1924. No doubt some of the huge movement of funds
out of France was reflected in the rise in foreign-owned dollar balances mentioned above, but the available—though admittedly skimpy—evidence suggests
that the bulk of it found its way to London.98
The magnitude of the flow of funds to London and the resulting strength of
sterling during most of the year following the December 3, 1925 bank rate
increase once again confronted the authorities with the problem of avoiding an
unwanted easing of monetary policy. Partly, the problem was to avoid gold gains
so large as to arouse renewed public pressure for easier money. With this objective
in view, the Bank of England followed one of the suggestions that had been made
by Strong in 1925." Even when it was still losing gold during October-December
1925, the Bank of England was increasing its dollar balances at the Federal
Reserve Bank of New York. The process was continued during 1926. Of the
$115 million increase in Bank of England holdings of dollars at the Federal
Reserve and of gold, only the $38 million rise in gold was revealed in its weekly
balance sheet. The $77 million gain in dollars was lumped with changes in the
bank's "other securities" and was thus hidden from public view. Moreover, the
bank fully offset the expansionary effect of these dollar gains on the domestic
cash base by selling sterling securities, thus foreshadowing the operations of the
Exchange Equalization Account that was to be established six years later.100
Britain's need to maintain a tight monetary policy in order to foster adjustment
was forcing Norman, like Strong, to ignore the gold standard tradition under
which monetary policy was supposed to be automatically linked to movements
in international reserves.

THE UNFINISHED TASK OF INTERNATIONAL ECONOMIC COOPERATION

By the end of 1926, the restoration of sterling to the prewar gold parity appeared
to have been successfully accomplished. The British currency was relatively strong
on the foreign exchange markets, and London was regaining its position as a
major international financial center. Beyond this, eighteen other countries had

98
Institut National de la Statistique et des Etudes Economiques,
Annuaire Statistique de la France,
Retrospectif
Edition, 1961, page 211; J. M. Keynes, "The British Balance of Trade", E c o n o m i c
Journal, Vol 37 (December 1927), page 556.
99
Supra, page 100.
100
Clay, op. cit., page 255.




105

by then stabilized their currencies in terms of gold, and France, the only major
country whose currency was not formally stabilized, was making rapid progress
in that direction.
Compared with the chaos of the earlier postwar years, the relatively stable
exchange conditions of the latter part of 1926 represented a major advance. Yet,
despite all appearances, these conditions rested on shaky foundations. An appreciable part of the large foreign balances in London consisted of French flight
capital that was likely to be repatriated as confidence in the franc strengthened.
The heavy outflow of capital from the United States was continuing, but there
could be no assurance of how long it would last. Nor was there any assurance
that, if this outflow declined, the consequent difficulties could be handled
through central bank cooperation. The United States economic conditions that
had made it possible for the Federal Reserve to cooperate so successfully with
the Bank of England in 1924-25 were bound to change, as the central bankers
themselves clearly recognized. As early as January 1925, well before the British
authorities had made the final decision to return to gold, Strong had warned
Norman that,101
in a new country such as ours, with an enthusiastic, energetic, and optimistic population, where enterprise at times was highly stimulated and the returns upon capital
much greater than in other countries, there would be times when speculative tendencies would make it necessary for the Federal Reserve Banks to exercise restraint
by increased discount rates, and possibly rather high money rates in the market.
Should such times arise, domestic
sympathies . . . .

considerations

would likely outweigh

foreign

Looking back, it seems clear that Norman and Strong were confronted with
urgent questions: Could the monetary contraction that was being pursued by the
London authorities bring about an adequate strengthening of Britain's fundamental external position before the disappearance of the international conditions
that favored sterling? Was it not necessary to develop more effective policy
techniques to accelerate the adjustment process, or failing this, could new forms
of international cooperation be arranged to support sterling in the interim?
Confronting their day-to-day problems, Strong and Norman almost certainly
sensed the long-term difficulties with which they were faced, but there is no
evidence that they thought in terms of questions such as these. Such questions
were foreign to the intellectual milieu in which they worked. They were influenced

101

Strong memorandum of January 11, 1925, page 3. Italics added.

106




by the illusion that the stabilization of most currencies in terms of gold had
brought about the realization of the gold standard conception to which they
aspired. Consequently, they tended to deal with some of the knotty problems of
the mid-1920's in terms of glib textbook formulas that were of doubtful validity
even in the nineteenth century heyday of the gold standard. It is certainly true
that, where their own pressing national problems were concerned, Strong, Norman,
and their colleagues were tough, penetrating, and gave short shrift to orthodox
preconceptions that ran counter to their views of the public interest. However,
where the problems of others were concerned, they sometimes based their judgments on the pieties of the gold standard rules of the game. These were venerable,
accepted, and were particularly convenient when they shifted to foreign shoulders
the burden of dealing with the adjustment problem. But there was far more than
an understandable orthodoxy behind the failure to develop and systematize
central bank cooperation. The priority that the monetary authorities were
expected to give to their respective domestic needs discouraged the discussion of
how the burdens of adjustment should be distributed among deficit and surplus
countries. In a period when financial probity was equated with budgetary balance
and the virtue of government noninterference in the economy was almost beyond
question, there was little disposition to ask whether monetary policy alone could
be relied upon to foster required international adjustments in the time that was
likely to be available, or whether policy-induced movements of capital might
actually obscure the need to adjust.
Looking at the records of the midtwenties, one is forced to the conclusion
that the authorities failed to develop any genuine understanding of the nature
and magnitude of the international adjustment problem and of how and on
what terms such adjustments should be financed. But one senses that, in his
persistent efforts to convene a central bankers' conference, Norman was groping
in his intuitive way toward such understanding and that a significant step forward was made when Strong agreed in November 1925 that "a quiet meeting
of some of the heads of the central banks" might be useful.102 Such a meeting
was eventually held but not until July 1927. It was only with painful slowness
that the central bankers reviewed and extended their cooperative techniques,
far too slowly to cope effectively with the overwhelming international economic
pressures that were soon to be upon them.

102

Strong letter to Norman, November 20, 1925, page 2.




107

6. The Defense of a Key Currency: 1926-28
INTRODUCTION

The strains to which the international financial system was subject in the later
1920's arose from the sharply contrasting economic conditions of the major
industrial countries, from divergences both in national aims and in official views
about how the international financial system should operate, from inadequacies
in monetary instruments and policies, and from peculiarities in the international
market for gold.
It has already been noted how the rapid recoveries of Germany and France
led to heavy movements of capital into those countries from the more slowly
expanding United States and British economies. The German and French central
banks consequently acquired substantial short-term claims on New York and
London but, as they did so, they pumped cash into their own economies to an
extent which caused official concern about the maintenance of monetary stability.
Having inadequate instruments with which to enforce monetary control in their
own markets, the German and French authorities attempted to shift the application of gold standard remedies abroad by converting foreign balances into
gold. They hoped that, by so doing, they would force the gold-losing countries
to raise interest rates and thereby to check the outflow of capital. In the United
States, where the Federal Reserve felt its stock of gold to be excessive, such
conversions presented no problem, but in Britain they soon gave rise to grave
concern. Despite the flow into the London market of the gold output of South
Africa and other British Empire countries, the addition of German and French
conversions to existing demand exhausted market supplies and at times forced
the Bank of England to meet the residual demand from its own holdings. Since
all but a slim margin of these holdings served as note cover, the loss of only
relatively modest amounts of Bank of England gold tended to force the authorities to intensify the deflationary pressures on the already depressed economy,
and confronted them with the questions whether the accentuation of such pressure would be politically tolerable in Britain and whether sterling's link with
gold could be maintained.
The Bank of England's problem was openly discussed by Norman, Strong,
Moreau, and Schacht, who understood that the breaking of the link between
sterling and gold would gravely threaten the newly restored international financial
system and who therefore worked for a time to alleviate the pressures on sterling.
But little was done either to correct the underlying difficulties or to develop more
108




permanent international machinery by which the major countries could support
each other's currencies. Indeed, the temporary rise in the Bank of England's
gold holdings that was the immediate outcome of the cooperative efforts diverted
attention from the fundamental economic difficulties, encouraged the authorities
to adopt measures that eventually exposed sterling to renewed pressures, and
complicated the problems of central bank cooperation.

PAYMENTS SURPLUSES COMPLICATE MONETARY POLICY

Cooperation among the major countries in 1926-28 was fundamentally different
from that of 1924-25. In the earlier phase, Norman and Strong had worked to
attain a mutually desired objective: the restoration of sterling to its 1914 parity.
In the second phase, cooperation involved the reconciliation of the major countries' conflicting needs in order to maintain that parity. Of several sources of
conflict, the one that is relevant here arose from the heavy flow of funds into
Germany and France in 1926-28 and from the problems of currency management
that grew out of these inflows. In both countries, these flows were considered by
the authorities to threaten not only domestic monetary control but also the
external stability of their currencies.
The trouble arose first in Germany, to which a particularly large
volume of foreign long- and short-term capital flowed in 1926. So large was the
inflow that, despite substantial outpayments on account of the country's merchandise trade deficit and reparations, the Reichsbank's international assets rose
by one fourth during the year. However, the Reichsbank was not well equipped
to deal with the domestic monetary effects of these reserve gains. It held no
government debt that could be shifted to the market, and its holdings of bills
and other marketable paper were small in relation to the value of the foreign
balances that were coming into Germany. For a few months in the early part
of 1926, the Reichsbank succeeded in offsetting its foreign exchange purchases
by reducing its holdings of marketable securities but, as Schacht subsequently
related to Norman, when this open market ammunition had been exhausted,
"I lost totally control of the market".1
GERMANY.

1

Schacht letter to Norman, May 21, 1927, page 1.




109

At the same time, Schacht was gravely concerned about the threat to the
external stability of the reichsmark. He never ceased to insist that heavy foreign
borrowing was placing Germany in a completely false position. It was false
because, under the Dawes Plan, reparation payments and the service of the
Dawes Loan were given absolute priority over all other German obligations:
Germany's private creditors abroad were being deluded about the security of
their loans. Moreover, Germany itself, Schacht insisted, was living in a fool's
paradise because the proceeds of a significant proportion of its foreign borrowings
were not being employed in a manner that would increase the economy's ability
to export and so meet its foreign obligations. As Schacht saw it, the time would
inevitably come when the inflow from abroad would cease, bringing the threat
of default not only on the private obligations incurred abroad but perhaps even
on the Dawes Loan and reparations. Unsound and excessive foreign borrowing
thus threatened to undermine the financial structure that had been so painfully
built only two years before.2
At about the same time, parallel problems were developing in France.
There they arose, not only from the huge influx of balances from abroad that
followed the successful July 1926 effort of the newly formed Poincare government to deal with France's chronic fiscal difficulties, but also from the inability
of the French authorities to agree upon the exchange rate at which the franc
should be legally stabilized. Indeed, this exchange rate debate continued with
varying degrees of intensity until June 1928 when the franc was finally stabilized
FRANCE.

de jure.

In the intervening period, the exchange markets were swept by ever-changing
rumors about the level at which the franc parity would be fixed. However, after
August 7, 1926 when the Bank of France was authorized to acquire gold, silver
coin, and foreign exchange at a premium over the rate at which it was empowered
to operate under its prewar statutes, the authorities were in a position to choose
whether to let these market pressures be reflected in movements of the exchange
rate or of official reserves. Actually they operated at first to moderate the appreciation of the franc, as balances were repatriated during the fall of 1926.

2

Hjalmar Schacht, The Stabilization of the Mark (London: George Allen & Unwin, Ltd., 1927),
Chapter 9, and The End of Reparations (New York: Jonathan Cape and Harrison Smith, 1931),
Chapter 3.

110




Subsequently, however, they pegged the currency at roughly 25 francs to the
dollar, at which level the exchange rate was stablized[stabilzed]de facto in December 1926.
During the first half of 1927, the French economy continued to attract balances
from abroad. Frenchmen, whose confidence in their currency had revived,
repatriated their funds. Yields on long-term government bonds were high. The
country's economic prospects seemed promising. Equity prices on the Paris
bourse seemed likely to rise. Above all, influential persons—including Poincare
and Doumergue, the President of the Republic—were working for a further
substantial appreciation of the franc.3 Indeed, certain directors of the Bank of
France favored a return to the prewar parity and said as much in public.4 When
the Bank of France had the satisfaction of announcing in April 1927 that it had
paid off a substantial loan from the Bank of England in a lump sum and well
in advance of maturity and had thus regained custody of gold which had been
pledged against the loan, speculation that the franc would be allowed to appreciate
further was greatly intensified. Not only did Frenchmen hasten to repatriate their
funds but the evidence suggests that other Europeans, and perhaps Americans
too, joined the rush to get into the franc. The Bank of France's foreign exchange
holdings, which totaled no more than $50 million at the end of November 1926,
mounted to $770 million by the end of May.5
The heavy inflows of foreign balances caused problems for Moreau in the
spring of 1927 similar to those that had confronted Schacht during 1926. On the
domestic side, the threat was perhaps not so great in France as in Germany
because close cooperation between the French central bank and Treasury enabled
the authorities to maintain a degree of monetary control. In effect, much of the
cash that was pumped into the market by the Bank of France's purchases of
foreign exchange was mopped up by Treasury funding issues, the proceeds of
which were employed to reduce the government's short-term debt to the central
bank. The authorities thus shifted government debt from the central bank to the
market and achieved the same results that would have been attained through
central bank open market sales.6 The drawback, of course, was that the central

3

Moreau Diary, April 29, 1927, January 9, 1928, and page 572.

4

Strong letter to Harrison, May 15, 1926. Also Moreau Diary, April 29-May 12, 1927.

5

Quesnay letter to Goldenweiser, November 29, 1928 (Federal Reserve Board Papers).

6

League of Nations, International Currency Experience (Princeton, 1944), page 76.




111

bank could maintain monetary control only so long as the cooperation with the
Treasury continued, and despite the success of this cooperation there is evidence
that Moreau believed the continuing inflow of foreign balances was leading to
an excessive increase in the bank's note issue.7
It was the external threat posed by France's balance-of-payments surplus that
caused the gravest concern to Moreau, because the inflow of funds was working
on the side of those influential persons who wished to have the franc stabilized
at, or at least nearer to, its 1914 gold parity. To Moreau and his closest colleagues
any such appreciation of the currency was unthinkable, not only because it
would bring severe losses on the Bank of France's existing exchange holdings,
but also because it would involve the deflation of French wages and prices, an
increase in the burden of debt on the domestic economy, and a serious weakening of France's international competitive position. They felt intensely that France
should avoid the mistake that they believed Britain had made in 1925.8

THE SHIFT TOWARD GOLD

While the inflows of foreign funds led to problems that were broadly similar in
Germany and France, the manner in which Schacht and Moreau attempted to
deal with these problems in 1926 and 1927 showed marked contrasts, stemming
in large part from the differences in the relations between their two countries and
Britain and between themselves and Norman.
GERMANY. Throughout the early postwar years Britain, against French opposition, had generally supported a lenient Allied policy toward the defeated Central
Powers and had worked actively for the political and economic rehabilitation of
Germany, Austria, and Hungary. In this work, Norman, as we have seen, had
taken a leading part. It was, therefore, to Norman that Schacht turned as soon
as he had become President of the Reichsbank at the end of 1923, and he
received a warm and helpful welcome during his visit to the Bank of England

7

Moreau Diary, April 28, 1927, February 28, 1928, March 22, 1928, and April 5, 1928; Strong's
memorandum on discussion with the Bank of France, May 27, 1928; Rist letter to Strong, June
9, 1928.
8

Strong memorandum, "Program of Stabilization of the Franc", August 19, 1926, pages 3-5; Moreau
Diary, April 25 and May 12, 1927 and page 573; Moreau and Norman conversation, May 27,
1927, page 1.

112




early in January 1924.9 Norman's subsequent work to obtain acceptance of the
Dawes Plan and to float the Dawes Loan further solidified his relations with
Schacht. For all these reasons and because, in addition, the British government
was the most likely one to lead a movement for the reduction of reparations
and war debts, Schacht was inclined to look with sympathy and understanding
on Britain's financial difficulties, to do what he could to relieve them, and thus
to practice with moderation and restraint the gold standard principles which
he professed.
Accordingly, Schacht's external policy in 1924-26 was designed to minimize
the pressure on the London gold market and particularly on the Bank of
England's gold reserve, and to transfer as much of this pressure as possible to
New York, which was willing and able to take it. His policy thus served as the
prototype for the broader cooperation among the central bankers that was to be
arranged in the summer of 1927. It was, of course, to the Reichsbank's advantage to hold the sterling proceeds of the Dawes Loan until the British currency
had returned to par and to obtain in New York the gold that it needed toward
the end of 1924.10 Nevertheless, there is evidence that for a year or more after
April 1925 Schacht did not let his sterling balances fall much, if any, below
the level to which they had been raised by the Dawes Loan and that he continued, though London was the cheapest market, to meet the bulk of his gold
requirements elsewhere, primarily in New York.11
However, as Schacht's currency management problems mounted during 1926,
his willingness to spare London declined. Measures to curb the heavy inflow of
foreign funds became essential. Accordingly, the authorities unpegged the reichsmark from its fixed relation to the dollar and, by allowing the exchange rate to
fluctuate within the gold points, introduced an element of exchange risks into
the foreign exchange market. They also suspended the preferential tax treatment
that had previously been accorded to income from German bonds floated in
foreign markets.12 More important from the viewpoint of this monograph, Schacht

9

Hjalmar Schacht, My First Seventy-Six Years (London: Allan Wingate, 1955), Chapters 24 and 25.

10

Supra, page 65. Shepard Morgan letter to Strong, January 28, 1925.

11

Schacht, The Stabilization of the Mark, page 208; W. A. Brown, Jr., The International Gold
Standard Reinterpreted, 1914-1934 (New York: National Bureau of Economic Research, Inc.,
1940), Vol. 1, page 636.
12

Brown, The International Gold Standard Reinterpreted, 1914-1934, Vol. I, pages 479-80.




113

sought to treat his currency management problems with a gold standard remedy:
he no longer operated quite so gently as before in the London gold market.
Toward the end of October, Norman was writing to Strong that "Schacht had
already sucked £ 61/2million in gold out of London".13 A little later, the Reichsbank encouraged private interests to acquire gold from abroad by announcing
that it would accept the metal in Bremen as well as in Berlin, thus saving private
arbitragers the cost of shipping the gold to the bank's head office.14 In the six
months ended February 1927, British gold exports to Germany totaled almost
$60 million, more than half the output of the South African mines during that
period.15
How far Schacht expected his gold purchases to mitigate his monetary control
problem is not clear. He probably expected them to have no effect on the Federal
Reserve, whose strength and policy he understood, but he may have hoped to
encourage a tightening of British monetary policy that would curb the flow of
funds from London. If so, he was disappointed when he received a letter toward
the end of December 1926 in which Norman said:16
Your German concerns borrow money in America (of which I do not think they
have need, but that is another question) and Dutch or German Bankers, seeing the
strength of the Mark-Pound Exchange, withdrew the proceeds of these Loans from
London in gold. There is nothing within reason which I could do to protect this
market: it is not a question of interest rates against which a Central Bank may
need to be defended but is rather a question of1/2% or thereabouts on the charges
for the shipment of gold from New York.
I am not complaining but merely thinking aloud of the difficulties which nowadays
arise owing to our international machine being so greatly out of gear .

Clearly Norman and Schacht were at cross-purposes. When he received a
sympathetic reply from Schacht, Norman went on to encourage the German
central banker to consider cooperative means by which to avoid gold movements
that, in Norman's words, "are nowadays often fortuitous and irrelevant to the

13

Norman letter to Strong, October 29, 1926.

14

W. A. Brown, Jr., England and the New Gold Standard, 1919-1926 (New Haven: Yale University
Press, 1929), pages 293 and 295.
15

The Economist, January 14, 1928, page 61.

16

Sir Henry Clay, Lord Norman (London: Macmillan & Co., Ltd., 1957), page 224.

114




monetary situation in one country or another".17 But before these discussions
could proceed very far, Germany's balance-of-payments position turned around.
By mid-1927, the Reichsbank's international assets had dropped $250 million.18
The pressure on London was no longer coming from Berlin but from Paris.
The atmosphere in which Norman and Moreau attempted to deal with
the flow of funds to Paris was hardly conducive to cooperation. They came to
the task without the fund of mutual understanding and confidence that existed
between Norman and Strong and Schacht. This lack stemmed in large part from
the long history of rivalry between the two countries and particularly from conflict
in their postwar policies relating to Germany.19 It also grew out of Norman's
unhelpful attitude toward France in the difficult years before July 1926. It is
impossible to say how far Norman's attitude was determined by the broader
political conflict and how far by the inability of successive French governments
to deal firmly with the country's deteriorating finances and thus create conditions
that would justify help from the Bank of England. The fact remains that there
is nothing in the record of the years up to July 1926 to indicate that Norman
went out of his way to help the French authorities, and several instances are
known in which he opposed the granting of any financial assistance or, if assistance was given, insisted on terms that the French considered onerous.20
The early conflicts and humiliations were keenly remembered by the management of the Bank of France after the franc recovered its strength in mid-1926.
Led by Emile Moreau, who became Governor in June 1926, the bank's management was intensely patriotic, as dedicated to the political objectives of the
Poincare government as to the strength of the franc, but isolated from the
monetary authorities elsewhere in Europe and unfamiliar with the problems of
the evolving international financial system. Indeed Strong, in the course of his
protracted discussions with Moreau and his colleagues in August-September 1926,
became convinced that the Bank of France management was suspicious and
mistrustful of most European central bankers, and especially of Norman and
FRANCE.

17

Ibid., page

18

Reports of the Agent General, September

225.
1, 1926 to August 31, 1927, Vol. I, page

19

Arnold Wolfers, Britain and France Between Two Wars (1st ed.; New
Co., 1940). See also Chapter 2.
20

Clay, op. cit., page 223; Moreau Diary, February
December 13, 1924.




26, 1927; Norman

240.

York: Harcourt
cable

Brace

to Strong,

No.

and
43,

115

Schacht, and would listen to advice only from the Federal Reserve authorities,
whom the French considered to be uninvolved with Europe's perennial quarrels.21
An instance of the mutual suspicion that permeated the relations between the
British and French central bankers arose in the early months of 1927 when
Moreau approached Norman about modification of the terms of a loan that the
Bank of England had made to the Bank of France in 1923, secured by the
deposit of French gold in London. Writing to Harrison, Strong described the
atmosphere in which these negotiations took place.22
The Bank of France did not believe that the [$] 90 millions of gold due from the
Bank of England was actually there, but feared that it had been used. The Bank of
England did not believe that the Bank of France had the sterling available with
which to repay the commercial debt, and feared the entry of the Bank of France
into the exchange market. I think this mistrust continued despite assurances by both
sides as to what the facts were. On the whole, it was a pitiful exhibition on both
sides, not as to the character and intentions of the individuals, but as to the general
atmosphere in which all of these matters are discussed and dealt with, and I admit
that I was greatly impressed by the fact that they both [i.e., Siepmann and Quesnay,
with whom Strong had been discussing the problem] accepted the point of view that
individually any and all of them were willing to do the utmost to improve matters,
but that collectively, because of the overshadowing influence of political considerations in Europe as a whole, they were almost helpless.

It was against this unpromising background that the difficulties arising from
the flow of funds to Paris were handled. Holding large short-term balances in
London, the Bank of France, after so many years of weakness, found itself
in a powerful bargaining position, which it fully appreciated and employed to
force the adoption abroad of measures to check the inflow. By the same token
the position of the Bank of England was extremely weak, forcing Norman to
fall back, after all else had failed, on the warning that, if pressed too far, Britain
might have no alternative but to abandon the gold standard and thus jeopardize
the reconstruction of the international financial system. In the end the impasse
was broken not only by a degree of accommodation on the part of both the
British and the French but also by the intervention of Strong, under whose
leadership the Federal Reserve adopted policies that were to shift much of the
burden of adjustment from the British economy to that of the United States.

21
Strong cable to Harrison, No. 46, July 31, 1926; Strong letter to Harrison, August 17, 1926,
pages 3-4 (Harrison Papers).
22

Strong letter to Harrison, July 27, 1928, page 6.

116




In the struggle to deal with the inflow to Paris, the Bank of France made the
first move. Its aim was to bring about a rise in the Bank of England discount
rate, a tightening in the London money market, and a curtailment of British
credit that the French authorities believed to have been extended to the major
Continental centers and, by this means, to force a tightening of monetary conditions in Amsterdam and Berlin as well as in London.23 To achieve these objectives
the Bank of France began, apparently without prior notification to the Bank of
England, to convert sterling balances both into dollars and into gold. On several
days in mid-May 1927, Moreau noted in his dairy that, although the Bank of
France was still taking in substantial amounts of sterling, it had sold even larger
amounts for dollars.24 At about the same time Moreau asked Strong to convert
$100 million of the Bank of France's dollar balances into gold, which was done
during the latter half of May. Similarly, he requested the Bank of England to
acquire gold in the London market to the value of £1.5 million on May 18,
another £ 1 million on May 21, and thereafter £ 3 million weekly until a total
of £ 2 0 million had been acquired.25 When Moreau learned of the alarm that
this program had created at the Bank of England, he agreed to postpone further
gold purchases in London, so that the British authorities might have an opportunity to suggest alternative means by which to curb the flow of funds to France.
It was to discuss such means that Norman visited the Bank of France on May 27.
The record of this meeting brings sharply into focus the opposing position of
the two central banks.26 In a brilliant exposition Norman described Britain's
domestic economic difficulties and argued that a tightening of Bank of England
monetary policy would only aggravate these difficulties while failing to offset the
overwhelming pull of the French economy on foreign balances. Norman concluded that the inflow could be effectively checked only if the French authorities
themselves were to adopt corrective measures including the de jure stabilization
of the franc or at least a firm declaration that speculative pressures would not
force the abandonment of the existing exchange rate. Unfortunately, the disagree-

23

Moreau Diary, May 12, 1927; Norman and Moreau conversation,
May 27, 1927, pages 1-3
and 13-17.
Moreau Diary, May 13 and 16, 1927.
25
Moreau Diary, May 18, 1927. See also Norman's cables to Strong, No. 38, May 24, 1927 and
No. 40, May 25, 1927, in which Norman refers to Moreau's ''capricious demands" and to his "fear"
that Moreau's gold buying program would "menace the gold
standard".
26
The record is published in full in the M o r e a u Diary, May 27, 1927, and in an abbreviated
English
version in Clay, op. cit., pages 228-32.

24




117

ment among the French authorities precluded any such declaration at the time—
let alone de jure stablization[stabilzation].Consequently, Moreau and his associates, Rist and
Quesnay, did not budge from their position that, since the flow was coming to
France primarily in the form of sterling, the responsibility for checking the flow
rested squarely on the Bank of England. They sympathized with Britain's domestic
problems; they did not wish to impair sterling's link with gold; but they insisted
that the Bank of England should tighten the London money market, if only to
attract new funds from abroad to replace the balances that the Bank of France
wished to convert into gold and dollars.
Although neither side gave any ground in these discussions, the May 27
meeting prepared the way for accommodation. The French authorities wanted
to shift the burden of adjustment abroad but, in exerting pressure to achieve this
aim, they did not wish to go as far as to push sterling off gold. Only a few days
before the meeting, Poincare had indicated to Moreau his opinion that Britain's
financial position had greatly deteriorated in recent months and that, if a Labor
government were to replace the Conservative one, the existing sterling parity
could not be long maintained; he recommended that the Governor follow a
cautious course in his dealing with the Bank of England.27 At the May 27
meeting, Moreau may have been left unmoved by Norman's warnings that French
pressure could undermine the gold standard, but subsequently he became convinced of Norman's desperation by a countermove that the British authorities
threatened to make if the Bank of France pressure became too severe. In effect,
the British would shift the dispute from one between central banks to one between
governments. As Moreau understood it, the British authorities would say:28
You have claims on us of £70 million. If you insist on repayment, if you wish to
liquidate and thereby ruin us, the [British] Treasury will collect its £600 million
claim on France [for war debts]. This debt is not yet consolidated. These notes,
which until now have always been renewed and which represent the indebtedness
of the French Government, we shall put on the market or present for collection.
And, in the ensuing chaos, we shall see who gets the better of it.

By mid-June, Moreau had had an opportunity to size up Britain's countermove
and had become convinced that his dominance over the Bank of England was
counterbalanced by the dominance of the British Treasury over that of France:

27
28

Moreau Diary, May 23, 1927.
Moreau Diary, May 30, 1927.

118




"When we exert pressure upon the [Bank of England], M. Churchill threatens
M. Poincare."29 The confrontation had ended in a draw.

BRITISH-FRENCH ACCOMMODATION

While this standoff was approaching, changes were occurring in the London
market that enabled Moreau to modify his position in good grace. The Bank
of England did not change its bank rate, but it did squeeze the market and thus
encouraged modest increases in the rate for acceptances toward the end of May
and again at the beginning of June. Simultaneously, the flow of funds to France
slackened, although Norman attributed the change less to the interest rate
increases than to the exhaustion of the speculative forces.30 Whatever the causes,
the pressures that were undermining Bank of France monetary policy were on
the wane and Moreau could shift his aim to other targets. After the end of May,
the Bank of France concentrated increasingly on the diversification and—as it
believed—the improvement of the quality of its international assets. Some of
its sterling balances were converted into gold, some into various other forms of
foreign exchange, primarily dollars. Beyond this, Moreau indicated that he would
continue to maintain balances in London equal to the amount of speculative
funds which he estimated would leave Paris when the prospective de jure
stabilization of the franc finally killed any remaining hopes for a further appreciation of the currency. However, he felt strongly that his London balances should
not rise above this minimum.31
A program to rearrange the composition of the Bank of France's international
assets was completed early in June 1927. Under the plan proposed to the
Bank of England by Quesnay, the Bank of France would increase its gold holdings by £30 million. Of this, Strong offered to supply £ 1 2 million32 against

29

M o r e a u Diary, June 13, 1927.

30

Norman and Moreau conversation, May 27, 1927, page 9; Moreau Diary, May 27, June 13, and
June 30, 1927; Moreau letter to Strong, June 2, 1927.
31

Moreau cable to Strong, No. 23, June 8, 1927.

32

The gold sold by the Federal Reserve to the Bank of France was originally part of the £18 million
held by the Bank of England against its 1923 loan to the Bank of France. When this loan was repaid
in April 1927, £6 million of the previously pledged gold was sold by the Bank of France to the
Irving Trust Company of New York and the remainder to the Federal Reserve, which kept its £12
million earmarked in London in order to facilitate the disposal of the gold in Europe by avoiding the
cost of shipping it to America and back again.




119

sterling. Another £ 15 million was to be acquired in the London gold market
within six months. The total was completed by the inclusion of £ 3 million that
had already been acquired from the Bank of England before the negotiations
were completed in the second week of June.33 On the foreign exchange side, the
Bank of France would attempt to discourage any renewed rise in its claims on
London by discriminating against sterling in its market operations, i.e., by offering a better price for dollars and for the various continental European currencies
than for sterling.34
On several counts these arrangements provided welcome relief to the Bank
of England. Strong's willingness to sell his London gold holdings reduced the
amount that Moreau planned to take from London to £ 1 8 million from the
£ 2 0 million he had set as his target in mid-May.35 At least for the time being,
no more gold was likely to be taken from the Bank of England's reserve, and as
events worked out only a relatively small amount seems to have been taken from
the London market. By late July, when the balance of payments was turning
temporarily against France and the Bank of France suspended its gold purchases
in London, the Bank of England held no more than £18.4 million of gold earmarked for France,36 of which the bulk was of course accounted for by the earlier
purchases from the Federal Reserve and the Bank of England.37 Thereafter, there
are no indications that the Bank of France was an important factor in the London
gold market until the end of 1928.38
Moreau's foreign exchange policy had certain obvious drawbacks for the Bank
of England, but it had the advantage that it tended to shift sterling balances into
more willing hands and, when France was gaining reserves, led to the buildup of
Bank of France claims on markets other than London. As events worked out,
France's claims accumulated largely in New York, where they were to complicate
the problems of the United States monetary authorities in later years.39

33

M o r e a u Diary, June 7, 1927.

34

Moreau

35

Supra, page 117.

cable to Strong, No. 23, June 8, 1927; M o r e a u Diary, June 9 and 14, 1927.

36

Norman cable to Strong, No. 112, July 28, 1927; Rist cable to Strong, No. 51, August 4, 1927.
These earmarks were shipped to France in February-March
1928 in connection with the preparation for the de jure stabilization of the franc.

37

38

Clay, o p . cit., pages

39

Moreau Diary, June 14, 1927.

120




243-44.

The June 1927 arrangements were not wholly satisfactory from the French
viewpoint. A major aim of the French authorities at this stage was to curb the
amount of cash pumped into the economy and, to do this, they wished to limit
the Bank of France reserve gains not merely in sterling but in any form. Consequently, when France's balance-of-payments surplus reappeared about the middle
of August 1927, the Bank of France changed its tactics and dealt with the
surplus of the next ten months mainly through swap operations that made it
advantageous for French private interests very substantially to increase their
balances abroad.
The circumstances were particularly favorable for the use of the swap technique under which the Bank of France sold foreign exchange spot and repurchased it for future delivery. Although long-term interest rates in France were
high relative to those in major financial centers abroad, the rate on three-month
bankers' acceptances in Paris had fallen sharply during the early months of 1927,
because the funding of government debt had led to an extreme scarcity of shortterm paper in the Paris market; in the year beginning May 1927, it fluctuated
between 1.82 per cent and 2.95 per cent.40 On the other hand, the corresponding
rate in New York moved in the 3.1-3.7 per cent range while in London, as already
noted, acceptance rates had been pushed up to 4.3 per cent at the end of May
and fluctuated within the 4.0-4.3 per cent range for most of the following year.
With this pattern of interest rates there was a particularly strong incentive for
commercial banks and others in Paris to hold balances in London, if forward
cover could be obtained on favorable terms, and this the Bank of France proceeded to provide on an increasing scale beginning about the middle of August.
Whereas the premium on forward francs had eliminated the incentive to move
covered funds to London from Paris during the spring of 1927, the premium
narrowed markedly thereafter, disappeared entirely in November, and fluctuated
between 0.2 per cent and 0.7 per cent during January-June 1928. On a comparison of acceptance rates there was, under these conditions, a covered incentive to
hold balances in London rather than in Paris of between 0.7 per cent and 1.9
per cent during the ten months ended June 1928 (see Chart 5). 4 1

40
R. G. Hawtrey, The Art of Central Banking (London: Longmans, Green and Co. Ltd., 1932),
pages 12-21.
41
Paul Einzig. The Theory of Forward Exchange (London: Macmillan & Co., Ltd., 1937), pages
352-53 and chart opposite page 286; Board of Governors of the Federal Reserve System, Banking
and Monetary Statistics (Washington, D. C, November 1943), pages 656-57.




121

CHART 5

ARBITRAGE OPPORTUNITIES BETWEEN
PARIS AND LONDON, APRIL 1927-JUNE 1928
Three-month bankers' acceptance rates; monthly averages

Sources: Board of Governors of the Federal Reserve System, Banking and
Monetary Statistics (Washington, D.C., November 1943). Data on forward
exchange calculated from figures in Paul Einzig, The Theory of Forward
Exchange(London, 1937), Appendix 1.

The Paris market was not slow to take advantage of this opportunity.
Moreau's diary indicates that, by September 20, 1927, the Bank of France
had entered into forward commitments to buy foreign exchange totaling almost
$100 million equivalent. By the beginning of June 1928, when these commitments had reached their peak, the total was almost $600 million equivalent and
was largely in sterling.42 In effect, the French authorities had induced private

42

Strong memoranda on "Bank of England-Bank of France Relations", May 24, 1928, page 7, and
on "Discussions with the Bank of France", May 27, 1928, pages 6-7. See also Einzig, op. cit., page
353; and Bank of France, Annual Report, 1928, published in Federal Reserve Bulletin (March
1929), page 203.

122




banks and other firms to increase their foreign balances by an amount that was
roughly equivalent to the surplus in France's balance of payments during
August 1927-May 1928. They thus minimized the increase in official foreign
exchange holdings and in Bank of France note circulation and facilitated the
attainment of the central bank's monetary aims. They also greatly eased the
pressure on sterling, although this result seems to have been merely incidental.
There is nothing in the record to indicate either that the French authorities were
aiming to help the Bank of England or that Norman and his colleagues realized
that the Bank of France swap operations were a major factor contributing to the
strength of sterling during the period.

THE CENTRAL BANKERS' MEETING OF JULY 1927

The respite enjoyed by sterling in the year beginning August 1927 stemmed from
policy changes not only by the Bank of France but also by the Federal Reserve.
Strong had, of course, been keenly aware of sterling's mounting difficulties.
During the autumn of 1926 he had learned from Norman of the Reichsbank's
purchases of gold in London. Later, frequent communications from both Norman
and Moreau told him of the growing tensions between the British and French
central banks. By the time he had read the record of the May 27 discussions,
which Moreau had sent to him, Strong felt deeply concerned for the stability
of sterling43 and was convinced that the question of how to reduce the pressure
on London should be topmost on the agenda of the informal meeting of central
bankers that was scheduled to begin early in July.
For years Norman had been working for just such a meeting, and Strong,
overcoming his earlier doubts, had issued invitations before the tensions between
the British and French central banks became acute.44 The meetings were held at
the home of Ogden Mills on Long Island from July 1-6 and were attended by
Norman, Schacht, Strong, and Rist, Moreau having been unable to come.45 An
attempt was made to keep the proceedings secret. No record of the discussions

43

Strong letter to Moreau,

44

Strong letter to Moreau, May 19, 1927, and cable No. 24, June 7, 1927.

45

June 20, 1927, page 3.

L. V. Chandler, Benjamin Strong, Central Banker (Washington,
1958), page 375; Hamlin Diary, July 7,1927.




D. C: The Brookings

Institution,

123

was kept, but enough was subsequently written by the participants to determine
that they explored the questions: ( 1 ) Whether the pressures on the European
authorities, especially in Britain and Germany, to tighten monetary policy could
be relieved by an easing of United States monetary policy; (2) whether there
was any connection between the monetary policies that were being followed by
the gold standard countries and the declining tendency of commodity prices generally; and ( 3 ) how to deal with the Bank of France's huge short-term claims
on London and New York and in particular (a) whether measures could be
found to enable the Bank of France to shift its sterling into dollars without inconveniencing the Bank of England and (b) whether the gold requirements of the
Bank of France, and indeed other central banks, could be met in New York
rather than in London, which was normally the cheapest market.46 The principal
outcome of these discussions was Strong's second great effort in support of sterling.

THE EASING OF FEDERAL RESERVE POLICY

In this effort the basic instrument, as in 1924, was an easing of monetary policy
which, in the light of the boom of the next two years and of the October 1929
crash, was to become one of the most controversial actions in the history of the
Federal Reserve System. Actually the mid-1927 policy change was far less
marked than that of 1924. In the earlier episode the discount rates of the Federal
Reserve Banks had been reduced in successive steps from41/2per cent to 3
per cent, whereas in 1927 they were cut by only1/2per cent to31/2per cent
over a period extending from July 29 to September 13. Similarly, the volume
of reserves that was pumped into the banking system was far smaller in MayNovember 1927 than it was in the 1924 episode, which as it happens covered
the same months of the year. System purchases of bills and Government securities
totaled $391 million, $61 million less than in the 1924 period. On the other
hand, the drain on member bank reserves from the outflow of currency into
circulation and gold losses was $217 million larger. By the same token, both
the decline in member bank discounts at the Federal Reserve and the rise in
their reserve balances were significantly smaller in May-November 1927 than
in the corresponding 1924 period. The rise in net demand deposits of weekly
reporting member banks was less than half the 14.5 per cent increase attained

46

Moreau Diary, July 16, 1927. See also Strong letter to Jay, July 21, 1927.

124




in May-November 1924.47
The controversy that was to develop over the 1927 easing of monetary policy
turned largely on the question whether the Federal Reserve, in seeking its international objectives, had exposed the domestic economy to undue inflationary
risks. The records of the meetings of the OMIC in early May and late July
indicate that, although both sides of the question were discussed, external
considerations probably weighed more heavily than domestic ones. Even so,
there is nothing to indicate that Strong told what he knew of the seriousness
of the Bank of England's position. The discussion was on a general plane;
the note of urgency is absent. There was some feeling that the maintenance
of the existing interest rate structure in the United States might attract additional gold to New York from Europe, force yet higher interest rates abroad,
and thus lead to a weakening of European demand for United States exports,
particularly of agricultural produce. If, on the other hand, monetary policy
were eased, upward pressure on European interest rates would be reduced, European demand for United States exports would be better maintained, the financing
of trade would tend to shift to New York from London, and America's pull on
Europe's gold reserves would be lessened. An important technical advantage of
a shift toward ease was that the System's open market portfolio could be replenished; open market sales to offset gold inflows during the early months of 1927
had reduced it to only $136 million on May 11.48
Apart from the stock market, the position and prospects of the United States
economy were considered by the Federal Reserve authorities to justify a modest
easing of monetary policy in mid-1927. Several of the participants at the joint
meetings of the OMIC and the Federal Reserve Board on May 9, May 12, and
July 27 felt that a recession was imminent or had already begun, slackness being
particularly noted in the shoe, textile, and oil industries. Although the Governors
of several of the interior Reserve Banks believed that conditions in their Districts
did not call for a policy change, most of them favored a systemwide move toward
ease because they felt that this would serve the national interest—by which they
apparently referred to the external aims mentioned above and the desirability of

47
Board of Governors of the Federal
140, and 370. Supra, page 86.
48

Reserve

System,

Banking and Monetary Statistics, pages 137,

Minutes of OMIC, May 9 and 12, 1927 and July 27, 1927; memoranda
meetings; Hamlin Diary, July 27, 1927.




prepared

for

those

125

acting promptly to check the recession. Only Governor MacDougal of the Chicago
Bank refused to go along with a systemwide move, but his position was based
primarily on what he took to be the special conditions of his District.49
Subsequent scholarly analysis confirms that the Federal Reserve authorities
correctly diagnosed the conditions of the economy in the spring and early summer
of 1927. If anything, they were late in recognizing the recession. The upper turning point, as subsequently determined by the National Bureau of Economic
Research, came in October 1926.50 Consumption turned downward in the fourth
quarter of that year; investment in producers' durables declined moderately but
steadily in each of the first three quarters of 1927 from a peak in the final quarter
of 1926; the substantial 1926 accumulation of inventories changed to decumulation in January-June 1927, changing again to what looks like involuntary accumulation in the third quarter and again to decumulation in October-December,
the quarter in which industrial production hit its cyclical low. Over all, however,
the recession was considerably milder than that of 1924: from peak to trough
the decline in industrial production had been 18 per cent in 1923-24, but it was
only 6 per cent in 1926-27.51
If the mildness of the recession called for only a modest move toward easier
money, so too did the continuation of speculative tendencies in the economy.
The recession, even attended as it was by the collapse of several real estate and
other speculative ventures, failed to dampen the exuberance of the New York
stock market. It was widely believed in the market that the business slowdown
was largely an outgrowth of the suspension of production at the Ford Motor
Company, which was retooling for a major model change, and that the longer
outlook was good. In this optimistic atmosphere, Standard and Poor's index of
New York stock market prices rose by almost one third between January and
December 1927.52

49

Minutes of OMIC, May 9 and 12, 1927, and July 27, 1927; memoranda prepared for these
meetings; Hamlin Diary, July 27, 1927; and Harrison, notes taken at the May 9, May 12, and
July 27 meetings.
50

United States Department
of Commerce, Bureau of the Census, Business Cycle Developments
(November 1964), page 61 (data compiled by National Bureau of Economic Research, Inc.).

51
Harold Barger, Outlay and Income in the United States, 1921-1938 (New York: National
of Economic Research, Inc., 1942), pages 93, 114, and 326.
52
Standard
page 7.

and Poor's

126




Corporation,

Daily Stock Price Indexes, 1926-1957 (New York,

Bureau
1957),

The question how the stock market would react to an easing of monetary policy
was a major worry to the Federal Reserve authorities at the time. Fear that a
speculative rampage would be encouraged led some to oppose any easing, and
Strong himself felt that, if the heavy speculation in the stock market were all that
had to be taken into account, some increase in discount rates might even be justified.53 But, although Strong and most of his colleagues understood that a move
toward easier money might invite further speculation, which could be dangerous,54
they felt intensely that the behavior of the securities markets should not divert
the System from the provision of an adequate supply of credit for what they considered to be the legitimate business of the economy.55 Edmund Platt of the Federal Reserve Board put the OMIC majority view bluntly: "Lower [the discount
rate] in New York first and to hell with the stock market."56

THE PAYMENTS BALANCE SHIFTS IN FAVOR OF EUROPE

While the United States economy was slipping into recession and Federal Reserve
policy was being eased, in Europe economic activity was quickening and interest
rates were on the rise. Investable funds were pulled eastward across the Atlantic,
and, as in 1924-25, the United States payments position weakened and sterling
appreciated on the exchange markets.
In Europe the volume of output expanded during 1927 everywhere except in
France, whose postwar recovery had been interrupted by a short-lived "stabilization crisis". Economic performance was particularly impressive in Germany and
Britain, where investment booms stimulated unusually rapid expansions in total
output. The acceleration of economic activity was accompanied by a tightening
of monetary conditions. In Germany, where sharply rising imports, together with
a slackening capital inflow, had led to substantial reserve losses in the early
months of the year, the Reichsbank discount rate was raised to 6 per cent early
in June and to 7 per cent in early October, from the 5 per cent to which it had
been reduced in January 1927. Throughout the second half of the year, short-

53

Strong letter to Moreau,

June 20, 1927, page 5.

54

Strong letter to Sprague,

April 12, 1927, page 2.

55

Minutes

56

Harrison, notes on OMIC meeting with the Federal Reserve Board on July 27, 1927.

of OMIC, July 27, 1927.




127

and long-term interest rates in Germany were 3 to 4 percentage points higher
than in New York. In France, too, long-term interest rates remained at levels
well above New York, despite the sharp drop in short-term interest rates, and
the stock market boom continued.57
In Britain the tightening of monetary conditions was rather gentle. Although
the British economy was recovering rapidly from the 1926 coal strike, unemployment in 1927 still averaged51/2per cent of the labor force,58 and the bank rate
remained unchanged. Nevertheless, the Bank of England so conducted its market
operations that bankers' deposits at the central bank dropped 2.6 per cent in the
six months ended November 1927 to the lowest level up to that time in the
1920's.59 Under this squeeze the acceptance rate held firmly at the 4.32-4.34
per cent level to which it had been pushed at the end of May, and long-term rates
hardened slightly. Since rates in the United States were declining, London's
normal margin above New York widened. More important, the incentive to move
covered funds from New York to London, which, on the comparison of acceptance rates, had narrowed during the spring, again widened andfluctuatedbetween
0.49 per cent and 0.73 per cent during July-November (see Chart 6).
These various economic changes had a major impact on the United States
balance of payments. Foreign long-term investment in the United States fell, and
United States investments abroad (both short and long term) increased. Hence,
despite a 6 per cent drop in imports, the deficit in the United States balance of
payments rose to $1,047 million, almost $700 million larger than in 1926 and
the highest for any year during the decade of the twenties.60 All indications are
that most of the increase in the deficit occurred during the second half of 1927
and that the net outflow of dollars continued at a very high rate in January-June
1928, when the volume of foreign capital issues publicly offered in the United
States reached its peak for the 1920's.61

57

Angus Maddison,
"Growth and Fluctuation
in the World Economy,
1870-1960",
Quarterly
Review (Banca Nazionale del Lavoro, June 1962), and Economic Growth in the West (New York:
Twentieth Century Fund, 1964), pages 201-40; League of Nations, Statistical Year-book, 1930-31.

58

Maddison,

59

Brown, The International Gold Standard Reinterpreted, 1914-1934, Vol. II, Appendix

60

Economic Growth in the West, page

220.

United States Department
of Commerce, Historical Statistics of the United States
D. C: United States Government Printing Office, 1960), page 563.

Table

3A.

(Washington,

61
United States Department of Commerce, Handbook on American Underwriting of Foreign Securities (Washington, D. C.: United States Government Printing Office, 1930), pages 140-41.

128




CHART 6
ARBITRAGE OPPORTUNITIES BETWEEN NEW YORK
AND LONDON, JANUARY 1927-JUNE 1928
Three-month bankers' acceptance rates; monthly averages

Sources: Board of Governors of the Federal Reserve System, Banking and
Monetary Statistics (Washington, D.C., November 1943). Data on forward
exchange calculated from figures in Paul Einzig, The Theory of Forward
Exchange (London, 1937), Appendix 1.

It was through this enlargement of the United States payments deficit and the
simultaneous strengthening of the major European exchange rates that the major
objectives of the Long Island central bankers' conference were accomplished. The
benefits to sterling of this payments shift were particularly great. Capital issues
that would normally have been floated in London were raised in New York, the
proceeds being sold for sterling and held in London.62 And, as Strong had hoped,
the financing of international trade was to some extent shifted to New York from
London; the value of dollar acceptances outstanding at the end of December 1927

62

Strong letter to Deputy Governor, Commonwealth Bank of Australia, September 7, 1927.




129

was almost $1.1 billion, 43 per cent higher than a year earlier.63 Especially in
the late summer and early autumn of 1927, when the United States economic
outlook was particularly uncertain and the money market was particularly liquid,
United States private short-term balances seem to have moved to London in considerable volume, taking advantage of both the covered incentive to move there
and of certain speculative opportunities.64 On one day in early October—a period
when sterling normally would still have been weak for seasonal reasons—Norman
cabled that he had bought $6 million in the market at rising rates.65 In December
and throughout the spring of 1928, sterling was only a little below the gold import
point from New York. For the year ended June 1928 as a whole, the sterlingdollar rate averaged the highest for any of the years 1924-31.

THE LIQUIDATION OF THE FEDERAL RESERVE'S STERLING BALANCE
AND THE SHIFTING OF THE DEMAND FOR GOLD TO NEW YORK

Sterling's exceptional strength facilitated both the speedy liquidation of the Federal Reserve's sterling balance and the rebuilding of the Bank of England's
holdings of gold. Arrangements to liquidate the £ 1 2 million66 that the Federal
Reserve had acquired from the Bank of France in exchange for gold were worked
out by Strong and Norman: the balance would be liquidated gradually at agreed
rates, partly in London by the Bank of England and partly in New York by the
Federal Reserve.67 Operations began in mid-August when the sterling-dollar rate
had climbed to parity and proceeded much more rapidly than Strong at least had
anticipated.68 The balance had been reduced to less than £ 5 million by the end
of August, to £ 1 . 5 million a month later, and by October 5 it was down to the

63

Board of Governors

64

Norman letter to Strong, October 11, 1927.

65

Norman

of the Federal Reserve System,

Banking a n d M o n e t a r y Statistics, page 465.

cable to Strong, No. 193, October 10, 1927.

66

The bulk of these balances was invested in a special "money employed" account managed by the
Bank of England and yielding the Federal Reserve 37/8per cent; the remainder was in sterling bills,
yielding 41/4per cent. Strong letter to Harding, September 6, 1927. Daily Statements of Foreign
Accounts, June 16, 1927-October 5, 1927.
67
Norman cable to Strong, No. 122, August 9, 1927; Strong
1927; and numerous other cables about this time.
68

cable to Norman,

No. 47, August

Strong letter to Norman, August 25, 1927; Daily Statements of Foreign Accounts, August
September 30, and October 5, 1927.

130




18,
31,

£ 150,000 at which the Federal Reserve's London balance had stood in early
June, before the gold was sold to the Bank of France. 69
The rapidity and ease with which the Federal Reserve's sterling balance was
liquidated did not relieve Norman and Strong from anxiety about longer term
problems with which they were faced. Both realized that the strengthening of
sterling was attributable primarily to temporary factors. Norman, fully aware
that sterling might need support again in the future, was glad that the Federal
Reserve's London balance could be reduced because, as he wrote Strong, "if we
get near to Queer Street I may ask [you] to increase it".70 For his part, Strong
had been clear from the beginning that, while changes in interest rate relationships between Europe and America might provide temporary relief for sterling
and other European currencies, such a change could hardly deal with fundamental
difficulties.71 He was also clear that the counterpart of the strengthening of
sterling was a buildup of American private claims on London, that these new
private claims were far larger than the original Federal Reserve claim which was
being so rapidly reduced,72 and that, if United States monetary conditions were
to tighten relative to Europe, these private claims would doubtless be repatriated.
Since he foresaw no fundamental improvements in Britain's position, he was
driven to agree with Norman on the need to prepare the ground for further
official support of sterling, although he made no specific commitments to
Norman.73 Therefore, once the sterling balance was back to normal, Strong asked
for, and on November 2, 1927 the Federal Reserve Board approved, authority
under which the System could act to prevent renewed physical movements of gold
to the United States by acquiring up to $100 million in foreign balances or gold
held under earmark abroad.74 This authority was not employed during the

69

Strong letter to Governor Young, October 5, 1927; Strong letter to Harding, October 5, 1927.

70

Norman letter to Strong, August 11, 1927.

71

Strong letter to Jay, July 21, 1927.

72

Strong letter to Norman, September 24, 1927.

73

Strong cable to Norman,

No. 95, September

24, 1927.

74

Minutes of OMIC, November 1-2, 1927; memoranda prepared for the meetings; Young letter to
Strong, November 2, 1927, in records of OMIC. The Board was asked to approve the proposed
gold and foreign exchange operations apparently because Strong expected that they would be
undertaken as a part of the System open market operations. On other occasions when it undertook similar, though smaller, operations, the New York Reserve Bank acted under the authority
conferred on the Reserve Banks by Section 14 of the Federal Reserve Act, merely consulting or
informing the Board of the action
contemplated.




131

remainder of Strong's lifetime, but it indicates the extent to which, despite his
hankering after the gold standard, Strong and indeed the System generally were
prepared to introduce flexibility and management into the operation of the international financial system.
Economic changes in the year ended mid-1928 also led to the shifting of a
considerable proportion of the world's rising demand for gold from London to
New York, an objective that clearly was sought by all the central bankers who
participated in the Long Island meetings. They had hoped that new gold production could be channeled to those monetary authorities, especially in Europe
and Latin America, who wished to increase their holdings of the metal in connection with the stabilization of their currencies, and that supplies from new
production could be supplemented by sales from the Federal Reserve on terms
at least as advantageous as in London. 75 Moreover, they had agreed to do their
best to avoid gold sales to the United States. If such sales were unavoidable, the
Federal Reserve would endeavor to take delivery in London, or elsewhere abroad,
where the gold could be resold more cheaply because the cost of moving it across
the Atlantic would not be included in its price.76
The key to the achievement of most of these objectives lay in the strengthening
of sterling. As sterling rose during the second half of 1927, the cost of buying
gold in London tended to rise in relation to the cost in New York. For certain
monetary authorities, the cost in New York—which previously had been the
highest of any major center—actually fell at times below the cost in London.77
By the same token the rise in the sterling rate also made it increasingly advantageous to sell gold in London rather than in New York.
The change in the pattern of gold flows was fostered not only by these exchange
market developments but also by administrative measures. In order to discourage
gold arbitrage between London and Berlin, the Reichsbank in October 1927 reduced its buying price for the metal to its legal minimum.78 A little later, when
the French authorities resumed the buying of gold to prepare the way for the

75

Preliminary
memorandum
for the OMIC, October
1927, page 372; Strong letter to Schacht, June 2, 1927.

18, 1927, page 11; M o r e a u Diary, July 16,

76
Norman cables to Lubbock, No. 11, July 13, 1927, and No. 14, July 19, 1927; Lubbock
Norman, No. 105, July 18, 1927.
77

Preliminary memorandum for the OMIC, October 18, 1927.

78

Reports of the Agent General, September 1, 1926 to August 31, 1927, Vol. I, pages 240-41.

132




cable to

de jure stabilization of the franc, the Bank of France met its needs in New York,
although at the time gold could apparently have been obtained somewhat more
cheaply in London.79 Moreover, with a view to bringing more of the supplies
that were coming onto the London market into the Bank of England's reserve,
the Committee of Treasury authorized Norman to offer as much as 77s. 10 1/2d.
per standard ounce of gold, an amount that was equal to its legal maximum
selling price.80
These changes in market conditions and in central bank policies brought about
a major shift in the pattern of gold flows, and in so doing not only lightened but
temporarily removed the pressure on London. This was accomplished despite a
huge increase in official demand for gold, especially from countries that were
stabilizing their currencies or were reducing—as was Germany—the foreign exchange component of their international reserves. Excluding the United States
and Britain, the rise in official gold reserves was $458 million in 1927, more than
quadruple the rise during the previous year, and the rate of increase accelerated
substantially during January-June 1928. In meeting this enlarged demand the
United States played the major role. Its gold stock, after increasing somewhat in
the first half of 1927, dropped almost $500 million during the year ended June
1928, especially large sales being made to Argentina, Brazil, Britain, France,
Germany, and Poland.81 The meeting of official requirements was facilitated
also by a major reduction in the "disappearance" of gold into industrial uses
and hoards.
The consequence was that the amount of gold channeled into official reserves

79
Moreau Diary, April 24, 1928, page 546, and Strong letter to Governor Young, October 25, 1927.
Referring to the eleven months immediately
preceding de jure stabilization
of the franc,
Poincare
told the Chamber of Deputies on June 21, 1928 that in the interest of central bank
cooperation
the Bank of France had "voluntarily renounced buying gold on the London market". Annales d e la
Chambre des Deputes, 14me Legislature Debats Parlementaires, Tome 135, June 21, 1928, pages
180-81.
80

Clay, o p . cit., page 237.

81

The governments of Argentina, Brazil, and Poland floated loans in New York, where the proceeds
were used in whole or in part for the acquisition of gold by their respective central banks.
Initially
the Bank Polski indicated that it wished to meet its gold requirements
in London where,
according
to its calculation, the cost would be lowest. However, out of the total of $20 million in gold bought
by the Bank Polski in October-December
1927, only $6 million was taken from the London
market,
the remainder being supplied by the Federal Reserve Bank of New York, partly out of its earmarks
in London. G. E. Roberts, "Gold Movements Into and Out of the United States, 1914 to 1929, and
the Effects", in League of Nations, Selected D o c u m e n t s o n the Distribution of Gold
(Geneva,
1931), page 48. The Bank Polski gold purchases were discussed in cables between Norman and
Strong during the period October 27-November
6, 1927. Board of Governors
of the Federal
Reserve System, Banking and Monetary Statistics, pages
542-55.




133

outside the United States in the eighteen months ended June 1928 was half again
as large as the total of new gold production during the period. Most significant
in view of sterling's key role was the fact that London attracted and retained
some gold on its own account. Not all the regular inflow of new gold production
from South Africa and other British Empire countries flowed out again. Moreover, in December 1927 and again in April-June 1928, when sterling was
especially strong in relation to the dollar, gold flowed from New York to London.
In these circumstances the Bank of England's reserve gains were the largest for
any of the years 1924-31. In the year ended June 30, 1928, its holding of dollars
rose about $70 million, and its gold holdings rose $93 million, which with a
further $18 million gold gain in July 1928 brought its international reserves to
their peak for the six years during which sterling was attached to gold.

THE INTERNATIONAL FINANCIAL SYSTEM BECOMES MORE RIGID

It is ironic that, although the central bank policies of 1927 and early 1928 succeeded in helping sterling, this very success was accompanied by a deterioration
in the relations among the major central bankers and by yet other policy changes
that were to make international financial cooperation more difficult in the future.
The cooling of relations among the major central bankers after the Long Island
conference was complex in origin. In part, it arose because Strong and Norman
were plagued by illnesses that made it difficult for them to keep in touch with the
rapid change of events and with their counterparts abroad.82 In part, it stemmed
from the intense rivalry for financial leadership in Europe that developed between
the British and French central banks as the strength and confidence of the latter
increased. Perhaps most important, it arose from the disillusioning international
monetary experiences of 1926-27 and from the rejection on the part of Strong,
Moreau, and Schacht of the Genoa principles, which Norman had been so
assiduously fostering, in favor of a rather strict and nationalistic version of the
gold standard.

82

Harrison cables to Norman, January 1928; Strong letter to Norman, March 3, 1928; Chandler,
op. cit., page 416. Early in July 1928 Schacht complained to Strong that he had lost contact with
Norman, whom he had not seen for a year. Harry Siepmann of the Bank of England attributed the
deterioration in relations between the British and German central banks to Norman's "illness,
nervousness and preoccupations" (Strong letter to Harrison, July 27, 1928, page 12). Schacht himself,
according to Strong, attributed the difficulties to Norman's personality; although Schacht was very
fond of Norman, he sometimes found him impossible to deal with because of his reserve; he felt
that he never knew what was in Norman's mind (Strong letter to Harrison, July 13, 1928, page 3).

134




In this shift in sentiment, Strong played a significant role. As the years went
by, his correspondence shows increasing skepticism about many of the ideas that
had emanated from the Genoa Conference and especially about the gold exchange
standard. His skepticism that any general gold shortage was impending, his feeling
that national credit structures are less soundly based on foreign exchange than
on gold, his concern lest his own monetary control problems be complicated by
large movements of foreign-owned dollar balances, and his understanding of the
problems that the buildup of foreign balances in London had brought on Norman
and Moreau—all these considerations led Strong to sympathize with and encourage the movement toward the gold standard.
On the other hand, Strong remained convinced that sterling was a key to the
maintenance of a stable international financial structure, and he was pragmatic
enough to soften his orthodoxy with proposals that would protect the position
of the British currency. With this object in mind, he urged the adoption abroad
of monetary laws that would set relatively low minimum holdings of gold against
note circulation.83 The advantage would be, in the case of Britain, that the bulk
of the Bank of England's gold would thus be available to defend sterling and,
in the case of France, that Bank of France gold purchases from abroad in connection with the de jure stabilization would be minimized. If the French were to
adopt a high and inflexible ratio, they would absorb a "mass of unusable metal"
and would play into the hands of those who alleged the existence of a gold
shortage and who were pressing for the extension of the gold exchange standard.84
Similarly Strong's opposition to the gold exchange standard in no way prevented him from feeling that central banks should be equipped to operate in
the exchange markets and, if necessary, to hold foreign exchange on a short-term
basis in the interests of cooperation. It was in this spirit that he obtained authority
in November 1927 to enable the Federal Reserve to rebuild its foreign exchange
balances and that, as the time for the French stabilization approached, he repeatedly urged Moreau to retain his authority to operate in the foreign exchange
markets.85 Otherwise, he warned, there would be the possibility that86

83

Strong memorandum on discussion with the Bank of France, May 27, 1928, page 12; Strong letter
to Rist, June 12, 1928, page 1; and Strong letter to Norman, March 29, 1927.
Strong letter to Rist, June 12, 1928.
85
Minutes of OMIC, November 1-2, 1927; Strong memorandum
on discussions with the Bank of
France, May 27, 1928, page 7.
86
Strong memorandum on discussions with the Bank of France, May 27, 1928, page 7.
84




135

if the franc rose above the gold import point all French bankers who could
import gold at a profit would do so and pump it into the Bank of France, and with
the importation of gold for geographical reasons being cheaper from London than
from any other market, there was danger of discord arising between the two
institutions immediately stabilization was effective.

Strong's efforts to foster a shift to the gold standard, without at the same time
undermining the position of sterling, had only limited success, and this success
endured only as long as the heavy outflow of American capital continued. Indeed,
the 1927-28 experience was strikingly similar to that of 1924-25, when the British
authorities took advantage of thefleetingbuoyancy of. sterling to fix their currency
once more at its prewar parity. Now the removal of the immediate threat to
sterling in the ten months ended June 1928 created a situation in which the
German and French authorities in their turn, marching under the gold standard
banner, adopted policy changes that were difficult to reverse and that increased
the rigidity of the international financial system.
The shift in Reichsbank gold policy, which was already foreshadowed in the
autumn of 1926,87 became evident in the later part of 1927 and early 1928. In
that period, the Reichsbank reconstituted only its working balances of foreign
exchange, and then became a very conservative buyer in the exchange markets.88
Under this policy the mark climbed gradually to the gold import point at which
it became profitable for arbitragers to obtain gold abroad to sell to the central
bank. The beauty of it was that, with sterling strong and the London gold price
high in terms of foreign currencies, this major policy change was accomplished
without putting pressure on London.89 The United States and Russia supplied
the bulk of the $54 million rise in the Reichsbank's gold holdings in the seven
months ended June 1928, a period during which its published foreign exchange
reserve actually declined $8 million.90
In France, too, the cessation of the virtually continuous official exchange operations of 1927 and early 1928 was accomplished with no immediate disturbances,

87

Supra, page

114.

88

Reports of the Agent General, September

89

Brown, The International Gold Standard Reinterpreted, 1914-1934, Vol. I, page

90

Reports of the Agent General, September
Reports of the Reichsbank, 1927 and 1928.

136




1, 1927 to August 31, 1928, Vol. I, page 86.

1, 1927 to August

488.

31, 1928, Vol. I, page 86; Annual

largely because the shift coincided with the June 25, 1928 de jure stabilization
and with the consequent disappearance of hopes that the franc would appreciate.
With the elimination of the speculative element in the exchange markets, both
sterling and the dollar strengthened against the franc; indeed, it was not until
nearly the end of 1928 that the effects of France's new policy began to be reflected in a renewed movement of gold from London to Paris.
While the change was initially painless, it was nevertheless irreversible. The
experience with official management of the exchange market had not been a happy
one for the French authorities. They had succeeded in preventing an appreciation
of the franc but only—as they felt—at the cost of an intolerable threat to domestic
monetary stability. The Bank of France had accumulated some $1.2 billion
equivalent in spot and forward claims on foreign centers during the eighteen
months prior to the de jure stabilization, equivalent to no less than half its total
assets at the end of 1926. When the Bank had attempted to curb the inflow by
converting part of its sterling claims into gold, Moreau had been told that he was
jeopardizing the international financial structure. Although, as already noted, the
Bank of France participated in the international cooperative arrangements to protect London during the year beginning July 1, 1927, the French authorities were
determined by mid-1928 to abandon the policies which they associated with the
embarrassments and difficulties of the pre-stabilization period.91
It must be emphasized, however, that the change was one in Bank of France
policy and not, as many authors have written, in its legal power to operate in
the foreign exchange markets.92 To be sure, the law of August 7, 1926 which
had authorized the Bank to buy gold, silver coin, and foreign exchange at a premium was abrogated when the franc was stabilized de jure. But the June 25, 1928
stabilization law itself made effective once again the authority under which the
Bank of France had operated in the gold and foreign exchange markets before
the war and which had been made inoperative as a result of the subsequent de-

91
Henri Cheron, Rapporteur
General de la Commission
des Finances, Rapport au Senat, June
24, 1928, Annales du Senat, Debats Parlementaires, Session Ordinaire de 1928, Tome 58, page 1127.
92

Virtually all publications on this subject hold erroneously that the June 25, 1928 monetary law
deprived the Bank of France of authority to operate in the foreign market. See, for example,
League
of Nations, International Currency Experience, pages 36-38; Martin Wolfe, The French Franc
Between the Wars, 1919-1939 (New York: Columbia University Press, 1951), page 98; and Hawtrey,
op. cit., pages 20, 30, and 31.




137

preciation of the franc in the foreign exchange markets.93
Hence even after the de jure stabilization, the Bank of France could have cooperated with the other central banks through exchange operations. Indeed, leaders
of the Bank of France indicated that they expected to do so. To Strong's plea for
the retention of power to conduct such operations, Moreau responded sympathetically, if somewhat equivocally, that the Bank of France would continue94
to pursue the policy of the past, namely, freely buying valuta on the market in
order to protect London from the drain, and he expected to continue that policy
with courage so long as the Bank of England made it possible for him to do so.

Similarly, Rist wrote that, if gold inflows into France were to reach "disquieting
proportions", the Bank of France was "quite determined . . . to intervene on the
market by buying devisen, so as to prevent the gold import point being reached."95
However, these laudable intentions were carried into effect only on rare occasions. Indeed, the normally passive posture of the French authorities in the
exchange markets in subsequent years seemed to confirm the general impression
(shared by numerous distinguished scholars) that the Bank of France had in
fact lost its legal power to operate in those markets, and the absence of any
recorded effort by the French authorities to disabuse the public of its error suggests that the Bank of France was content to let the impression stand. It was of
course close to the truth. For, although the bank retained power to conduct exchange operations, the French authorities felt that any further buildup of foreign
exchange holdings by direct and systematic purchases in the open market would
be contrary to the spirit (if not the letter) of the June 25, 1928 stabilization law.96

93
On this point, M. Robert Lacour-Gayet, Director of the Research Department of the Bank of
France, wrote to E. A. Goldenweiser on December 3, 1930 that:
the bank has the power under its statutes to discount bills payable abroad, drawn in
foreign currencies; it likewise has the power to buy and sell gold, and by implication, to
buy and sell foreign exchange representing gold.
The abrogation [in June 1928] of the law of August 7, 1926, did not restrict the bank's
freedom of action; it merely put an end to a temporary system which had grown out of
the inconvertibility of the currency . . .
This is confirmed by an examination of the statutes of the Bank of France. See Goldenweiser letter
to Lacour-Gayet, October 15, 1930 (Federal Reserve Board Papers), and Galantiere letter to Crane,
December 29, 1930.
94
95

Strong memorandum

on discussions

with the Bank of France, May 27, 1928, page 7.

Rist letter to Strong, June 18, 1928.

96
Robert
Papers).

138




Lacour-Gayet

letter to E. A . Goldenweiser,

December

3, 1930 (Federal Reserve

Board

INADEQUATE EFFORTS TO BUTTRESS STERLING

While Germany and France were moving toward a strict gold standard, the
British authorities were taking advantage of the relatively favorable economic
conditions of 1927-28 to push forward long-term plans aimed at the strengthening of sterling. Stimulated by the May 1927 crisis, the authorities intensified their
efforts to improve Britain's international competitive position. Plans were formulated for the consolidation and modernization of industries, such as those producing textiles, whose small, high-cost firms were being squeezed out of
international markets.97 Efforts were made to transfer labor from redundant coal
mines to more promising industries. On the financial side, the central bank was
put in a better position to defend sterling in the exchange markets. Although the
government did not go so far as to accept the view that the main purpose for
which the central bank held international assets was to maintain the external
value of the currency, it went some distance in this direction when it obtained the
enactment of the Currency and Bank Notes Act of July 2, 1928.
Under this act, the Bank of England's fiduciary issue was set at a level that—
had the law become effective in June—would have given the central bank a
margin of "free" gold relatively large in terms of previous British experience:
the bank could have allowed its gold to decline £54 million before it would
have been obliged to enforce a contraction of note circulation.98 Since at that
time the bank also held £45 million in dollar balances, the total of its "free"
international reserves would have been £ 9 9 million, a record for the period
since April 1925 and of course many times larger than the small amount held
before 1914.
The law also provided authority under which, in case of need, the amount
of the fiduciary issue could be changed in order either to release additional gold
for the defense of sterling without necessitating the adoption of deflationary
measures or to absorb gold inflows that might unduly inflate the note issue. The
government's intent was that this authority should be employed not only to deal
with crises but in the ordinary course of events, for example, to prevent the
credit stringencies that would arise if the existing note issue were insufficient to

97
Strong letter to Harrison, July 27, 1928, pages 1-2. See also Sproul letter to Harrison of March 31,
1934 and attached memorandum on "Bank of England and Industrial
Credits".
98
Parliamentary Debates (Great Britain,
1724; Macmillan Report, page 30.




House

of Commons),

Vol. 217, May 22, 1928,

column

139

satisfy the needs of domestic economic expansion or, if gold were flowing out,
to satisfy the stabilization requirements of foreign countries or because, for
any other reason, foreigners decided to hold gold rather than sterling. This
authority could be invoked only at the initiative of the governor of the bank,
but its employment required the approval of the Chancellor of the Exchequer.
Subject to his approval the fiduciary issue could be changed in any amount for
periods up to six months, which could be extended up to two years. Changes for
periods longer than two years required Parliamentary approval."
Thus, the authorities were working to strengthen the British economy on both
the real and financial sides, but progress was slow. Although those who knew
the severity of the May 1927 crisis were keenly aware of the urgent need for
economic adjustment, the general public was at best indifferent; when it was
directly affected, it was often openly hostile to change. The difficulties involved
in official attempts to improve the international competitive position of the textile
industry provide a case in point. After discussing the problem with Harry Siepmann of the Bank of England, Strong wrote to Harrison late in July 1928:100
There are some hundreds of mills, and Siepmann thinks the project will be large
consolidations, but that it will fail because of the old family traditions in many of
the mills, jealousy and inertia and inability to get together and cooperate. There are
something over 1,500 directors in these concerns, and even that is a great obstacle,
as all of them want to continue their positions and fees. He thinks the outlook in
these industries is bad both industrially and financially, and that is especially true
because the big banks which are carrying the financial load all say that they are
indifferent. Credits that are good, they are glad to carry and even increase as
needed. Those that are bad, they have entirely written off or reduced to a point
where they are no factor and they do not care to be troubled.

The arrangements to put the Bank of England in a stronger position to defend
sterling proved quite inadequate. In part, the failure was due to ignorance. The
authorities were aware that foreigners held substantial balances in London, but
apparently they were not aware of the huge magnitude of these liabilities. It was
not until July 1931 that the Macmillan Committee published estimates of London's net short-term liabilities to foreigners101 and even these large figures very

99
Parliamentary Debates (Great Britain, House of Commons), May 14, 1928, columns 744-46;
Macmillan Report, pages 139-40.
100
Strong letter to Harrison, July 27, 1928, page 2 .
101
Macmillan Report, pages 112 and 301. The report estimates London's net short-term liabilities at
£279 million at the end of 1927 and at £302 million a year later.

140




considerably underestimated the true total.102 In mid-1928 these liabilities were
probably three times or more the amount of the Bank of England's total "free"
international reserves. Since in practice the Bank never allowed its free reserves
to drop below the £ 2 0 million level before tightening credit policy, the effective
ratio of short-term liabilities to the bank's free international assets was of
course even higher.103
Another flaw in these arrangements was that the Bank of England was
extremely reluctant to initiate action to change the fiduciary issue.104 The
hitch lay in the danger—foreseen by at least one member of Parliament during
the debates on the Currency and Bank Notes Bill105—that any such action by
the bank would be interpreted as a sign of weakness and would lead to nervousness in the exchange markets at a time when the authorities were seeking to
encourage precisely the opposite sentiment. Summing up its criticism of the
arrangement for changing the fiduciary issue, the Macmillan Committee noted
in its report106
we believe that it would be preferable that the Bank should have control over
a much larger proportion of the total gold stock without calling in aid these provisions. If the object of the fixed fiduciary issue is to tie up the Bank of England,
well and good. This may be inadvisable but it is intelligible. But if its object is not
to fetter the Bank but merely to require it to make a public declaration of what will
be interpreted as weakness, at the most inadvisable moment for creating such an
impression, the disadvantages of liberty and of regulation are, it seems to us,
heaped on one another.

CONCLUSION

By mid-1928, many of the individual problems that would combine to cause the
September 1931 breakdown were already fairly clear and had indeed been recognized as sources of potential trouble. The decisive shift in German and French

102
David Williams, "London and the 1931 Financial Crisis", T h e E c o n o m i c History Review, 2nd
Series, Vol. XV (April 1963), No. 3, page 528; Royal Institute of International
Affairs, T h e
Problem of International Investment (London, 1937), pages 339-40.
103

M a c m i l l a n R e p o r t , page

302.

104

It did so only once, in August 1931, when the fiduciary issue was increased
Macmillan Committee's critical report had been published in July.
105

Parliamentary Debates (Great Britain, House of Commons),

106

by £15 million.

The

Macmillan Report, page 141.




May 22, 1928, column

1724.

141

gold policy was evident; their payments surpluses would in the future be settled
very largely, if not wholly, in gold. Although the Bank of France was still holding
a substantial part of the sterling it had accumulated during the early months of
1927, it was doing so with unmistakable reluctance. It was clear, moreover, that
the demand for gold from Germany, France, and other sources would soon fall
once again on London. The strength of sterling that had reduced the pressure
on the London gold market during the ten months ended June 1928, depended
on the greatly increased outflow of American capital, which was recognized as a
temporary phenomenon. In addition, there was awareness that heavy short-term
liabilities to foreigners had accumulated against Berlin and London, and also
concern lest the international reserves of the Reichsbank and the Bank of England
prove inadequate to deal with any substantial withdrawals.107
But, although each of the major danger spots was recognized individually, no
one viewed them as parts of a single problem. None analyzed their interrelations
or understood how they could act and react upon one another. Strong and
Norman, who were in a position to see the whole problem, were intermittently
ill and, even when well, were distracted by other pressing matters. Below the level
of the Governors, who dominated the international relations of the United States
and British central banks, were Harrison and Lubbock, but they were in less
satisfactory positions to understand the overall problem; what is more important,
the record gives no indication that either of them, or indeed anyone within the
central banks, was asked or volunteered to explore the strengths and weaknesses
of the existing international financial arrangements or to propose means by which
these arrangements could be made more viable. Outside the central banks, there
was no dearth of advice about how the international financial system should be
improved. But outside observers were at best incompletely informed about the
central problems. Even the Chancellor of the Exchequer was kept in the dark
about certain crucial facets of the Bank of England's operations.108
Moreover, Norman did not welcome unsolicited advice or the probing of outsiders into the affairs of the Bank of England. His attitude was only partly a

107

On October 22, 1927, Gaspard Farrer of Barings wrote to Strong that "no amount of gold which
we can afford to keep will suffice if ever serious doubts as to our stability arise—transfer by capitalist
and speculator would be overwhelming".
108
For example, in his memorandum of May 24, 1928 on the Currency and Bank Notes Bill, Strong
registered his "almost consternation" al the discovery that no member of the British government had
been made aware that the Bank of England held more than $200 million in New York.

142




reflection of the traditional feeling that the bank's operations should not be
subject to public scrutiny. As the defense of sterling ran into increasing difficulties, the wisdom of the return to the 1914 parity and the judgment of those
responsible for the return were increasingly brought into question. Perhaps inevitably Norman and Churchill, who of course were the main targets of public
criticism, responded by closing ranks: nothing was fundamentally wrong; the
well-seasoned arrangements of 1844 were working well; the critics were generally
either knaves or fools.109 Hence the authorities rejected the suggestion for official
investigation into Britain's monetary problems, a suggestion repeatedly made by
Reginald McKenna and taken up by the Labor Party.110 Indeed, to head off such
an inquiry, the authorities proposed to appoint a committee of experts who, it was
hoped, would support the views of the Treasury and the Bank of England,111 but
even such a group was apparently never appointed.
Thus, an expert and wide-ranging investigation of British monetary problems
was not conducted in the relatively favorable circumstances of 1927-28, when it
could still perhaps have had useful results, but was delayed until after the election
of the Labor Party in June 1929. By November 1929, when the Committee on
Finance and Industry was appointed, headed by Lord Macmillan, the international economic climate had already taken a decided turn for the worse, and
by July 1931, when the committee's report was finally published, sterling was
on the brink of disaster. The worst fears of those who had opposed the inquiry
were then realized, because the Macmillan Report's estimates of the magnitude
of London's short-term liabilities probably contributed to the crisis of confidence
that was soon to break sterling's link with gold.

109
Norman letter to Strong, March 14, 1927. See also Strong
Bank of England-Bank of France relations, especially page 14.
110
See, for example, McKenna's speech delivered
Review of the Midland Bank for January-February
111

Crane memorandum




on conversation

memorandum

of May 24, 1928, on

on January 28, 1927 and reprinted
1927, page 6.

with Sir Otto Niemeyer,

February

in the Monthly

18, 1928.

143

7. Mounting Strains—Declining Cooperation
. . . reconstruction needs an outside body of some sort which can continue for
many years to exercise a wide and impartial authority behind anything the Central
Banks may do.
Write me your ideas before your fears take effect and competition squeezes out
cooperation.
Norman letter to Strong,
April 17, 1927.

Now that budgets are balanced, debts are less menacing and currencies are reorganized, the principal European nations are being influenced by the feeling of strength
and security which they have obtained, and the old evils of cut-throat competition
are again showing themselves. It even enters into the policies of the banks of issue.
My belief is that the time has arrived to keep out of any complications of the sort
which the European banks of issue are likely to encounter. We have discharged
our moral obligation to Europe, and there seems to me to be no further occasion
for entering into arrangements which might embarrass us in any way, politically or
financially.
Benjamin Strong,
August 17, 1928.

Beginning in the summer of 1928 the international economy became subject to
strains with which the governments in the major countries were unprepared to
cope. Nowhere did the authorities deal effectively with the difficulties in their
domestic economies; what they did do often exacerbated the problems of other
countries, led to retaliation, aggravated existing frictions, and rendered increasingly remote the possibility that their common economic problems could be
tackled on a cooperative basis. The one major international effort of 1929-30
was directed toward the final settlement of reparations on a basis that could have
succeeded only if prosperity in the major countries and equilibrium in the international economy had been maintained. Both of these conditions disappeared in
1930-31. Unsupported by the requisite government policies and equipped with
instruments designed to deal only with relatively minor disturbances, the major
central banks, even when they were willing to cooperate, could do little more
than stand on the sidelines and watch the disintegration of the international
economic system.
144




THE REPARATION SETTLEMENT

In its timing and form, reparation settlement was heavily influenced by problems that had become pressing in Germany and France. In Germany, there was
of course continuous agitation to reduce the annuities which were scheduled to
total at least $596 million equivalent in the year ended August 1929 (two and
a half times the amount paid in 1924-25) and which were contributing to a growing budgetary deficit. Indeed, there was concern that the annuity might go even
higher because, under the so-called prosperity index of the Dawes Plan, the payments required were linked to specified indicators of German economic activity.
At least as important were pressures for the elimination of Allied controls over the
Reichsbank, the German railroads, and the post office and above all for the
evacuation of the Rhineland.
In France, the government wished to settle the amount and duration of the
annuities it was to receive from Germany before it ratified the war debt agreements it had negotiated with Britain and the United States in 1926 and which
called for substantial cash payments to those countries if the ratification were not
completed before August 1929. In addition, the French government wished to
capitalize part of its share of reparations, ostensibly in order to reduce its internal
debt. Probably a more fundamental reason was that the French authorities wished
to shift part of Germany's reparation obligation from the Allied governments to
private creditors on the theory that Germany would be less likely to renege on a
"commercial" debt than on a "political" one.1
As a result of negotiations in Paris during February-June 1929 and at the
Hague in August 1929 and again in January 1930, both Germany and its creditors
partially gained their objectives. The Allied controls over, and occupation of,
Germany were to be ended. Germany's reparation obligation for the year beginning April 1930 was reduced to a little over $400 million. Although the annuities
were to continue thereafter for fifty-seven years, the amount required was appreciably smaller than under the Dawes Plan. Moreover, Germany's immediate
problems were alleviated through the flotation of a $300 million international

1
The plan, which was implemented through the flotation of the Young Loan, was to assign a specified part of the annual reparation payments to the servicing of a bond issue that was to be raised by
the German government in the United States and in the principal financial markets of Europe. The
proceeds of the loan were distributed primarily to France, Britain, and Germany. Moreau Diary,
March 22, 1928, page 519; Thomas Balogh, "The Import of Gold Into France", Economic Journal,
September 1930, page 452; BIS First Annual Report, May 19, 1931; Gustav Stolper, German
Economy: 1870-1940, Issues and Trends (New York: Reynal and Hitchcock, 1940), page 175.




145

loan—called the Young Loan after Owen D. Young who played a prominent
role in the negotiations in Paris and at the Hague. Of the proceeds of this loan
Germany was to receive one third, an amount that was expected both to help
finance its budgetary deficit and also to supply part of the foreign exchange required to transfer reparations in the year beginning April 1930. For their parts,
France and the other creditor countries were assured of reparation receipts that
would at least equal their net war debt payments.2
While the reparation settlement alleviated some international financial difficulties, it aggravated others. Germany lost most of the transfer protection that
it had been accorded under the Dawes Plan. The Young Plan added to Germany's
obligation to service the Dawes Loan the further unconditional obligation to
transfer $146 million equivalent annually to service the Young Loan and
other portions of the reparation annuities that were expected to be capitalized. The remainder—which amounted to three fifths of the scheduled payments in the year beginning April 1930 and to a somewhat larger proportion in
later years3—could be postponed up to two years by the German authorities if
in their judgment "Germany's exchange and economic life may be seriously endangered by the transfer . . . ."4 This was in sharp contrast to the Dawes Plan
arrangements under which (1) a threat to the exchange stability of the reichsmark was grounds for the complete suspension of reparation transfer, (2) no
time limit was specified, and (3) the responsibility for making the suspension
rested with a technical body—the Transfer Committee—appointed by the Reparation Commission and indirectly by the Allied governments. Whereas the
Dawes Plan made transfer protection complete and removed it as far as possible
from the political sphere, the Young Plan not only severely reduced the extent
of transfer protection, but—in an international situation that was still charged
with tension—gave the responsibility for invoking this protection to the German
government. The door was thus opened to recriminations and charges of bad
faith, to further worsening of relations between the debtor and creditor powers,
and to an aggravation rather than an alleviation of the pressures that were to
afflict the reichsmark in the years immediately ahead.

2
Federal Reserve Bulletin (April 1930), pages 172-249; Reports of the Agent General, September
1, 1928 to May 17, 1930, Vol. 1, page 213; and BIS First Annual Report, May 19, 1931.
3

Federal Reserve Bulletin, April 1930, pages 181 and 184.

4

Ibid., page 175.

146




A more positive outcome of the Hague negotiations was the establishment in
mid-May 1930 of the Bank for International Settlements. This institution, designed to facilitate the transfer of reparations and to promote international financial cooperation, was equipped with resources adequate to smooth out the
relatively minor exchange disturbances characteristic of a prosperous and balanced world economy, but was never intended to cope with difficulties of the
magnitude of those which arose in the early 1930's. It started off with a major
handicap. The Federal Reserve System was not permitted to join because membership was thought by the United States Government to conflict with the official
American position regarding reparations. To assure itself of close relations with
the world's major capital market, the BIS therefore invited the participation of a
group of United States commercial banks. In addition, the central banks of
Britain, France, and Germany and of eighteen other countries as well as a group
of Japanese commercial banks participated. Even so, the resources of the BIS
were relatively small. Its capital totaled $100 million, of which only $21 million
was paid up before the end of 1931. Its assets totaled $412 million equivalent at
the end of May 1931, when they reached their peak for 1930-31, an amount that
was about the same as Germany's reparation obligation for that year. And of
course only a small proportion of these assets was available to support any single
currency. The bulk of its assets was held in dollar investments, while another
substantial proportion was placed in sterling. Under extreme pressure the bank
might have provided as much as $100 million in short-term support for a threatened major currency, i.e., support on the same scale as the Federal Reserve was
authorized to give under the November 2, 1927 resolution of the OMIC.

STRAINS DURING THE BOOM: JULY 1928-OCTOBER 1929

Even while the Paris and Hague conferences were still in session, economic
changes occurred that led to the collapse of the reparation settlement and to the
overwhelming of the efforts of the BIS and the major central banks to support
the international financial system.
The first phase of the international economic breakdown coincided with the final sixteen months of the New York stock market
boom, i.e., July 1928 to the end of October 1929. During this period the pull
of the booming American economy on investable funds drove New York money
market rates to record levels and greatly decreased the net outflow of capital
INTERNATIONAL IMBALANCE.




147

from the United States. Whereas during 1927 and probably also during the first
half of 1928, the United States balance-of-payments deficit had been in excess
of $1,000 million annually, it dropped to only $53 million in 1929. Especially
before the October 1929 crash, foreigners added substantially to their dollar
assets, with the consequence that the direction of gold flows shifted dramatically.
While in the year ended June 1928 the United States sold some $500 million
in gold to foreign countries, in the sixteen months ended October 1929 it bought
almost $300 million.
The pressures that were being exerted on the rest of the world by the United
States were augmented by the other major surplus country, France. Although
France's payments surplus was sharply reduced from the $500-600 million levels
of 1927 and 1928, it remained somewhat above $100 million in 1929. Moreover,
the $383 million in Bank of France swaps that were outstanding in mid-1928
matured during the second half of the year. Under the policy followed by the
French authorities after June 1928, foreign exchange acquired by the Bank of
France on account both of these maturities and of the payments surplus was converted virtually entirely into gold. Hence, although the Bank of France did not run
down its large preexisting foreign exchange holdings in the sixteen months ended
October 1929, it raised its gold reserves during this period by $434 million, partly
as a result of domestic dishoarding but mainly by purchases from abroad.5
The pressure emanating from the United States and France was felt throughout
the world. In an effort to counter the pull of the New York market and to replace
capital that was no longer flowing from the United States, monetary policy was
tightened nearly everywhere. Among the hardest hit by the change in the
international economic climate was Germany which had been so dependent
on the inflow of foreign, and especially American, capital in earlier years. To
the extent that it could no longer borrow abroad, Germany was faced with the
need to restrict spending at home in order to expand exports and so fulfill its
reparation obligation. Indeed, German credit conditions had been tight for much
of the year before mid-1928. The Reichsbank discount rate, which had been set
at 7 per cent in October 1927, remained at that level throughout 1928 (see Chart
7), while Berlin market rates, fluctuating on either side of the 7 per cent level,
were well above those of any other major European center. As a result, when
the inflow of American capital declined, Germany was able to recoup much of

5

Quesnay letter to Goldenweiser, November 29, 1928 (Federal Reserve Board Papers).

148




CHART 7

CENTRAL BANK DISCOUNT RATES
UNITED STATES, UNITED KINGDOM, GERMANY, AND FRANCE
DECEMBER 1927-APRIL 1931

Sources: Board of Governors of the Federal Reserve System, Banking and
Monetary Statistics ( Washington, D.C., November 1943 ); R.G.Hawtrey,
A Century of Bank Rate ( London, 1932 ); A. Parchman, Die Reichsbank
(Berlin, 1933).

the shortfall by borrowing in various European markets and especially in London
and Paris.6 The German credit stringency was accompanied not only by a continued high level of capital inflow but also by a rapid export rise, with the consequences that the Reichsbank's published international reserves increased $120
million during the July-December 1928 period. Early the following January the
Reichsbank discount rate was reduced to 61/2per cent. Thereafter, however, the
further tightening of credit conditions both in other European markets and in the
United States severely restricted the inflow of funds to Germany, and a heavy
volume of short-term balances was withdrawn in April when difficulties arose
at the Paris reparation conference.7 Although the Reichsbank discount rate was

6

Sir Henry Clay, Lord Norman (London: Macmillan & Co., Ltd., 1957), pages 244-45; Gardner
memorandum to Goldenweiser, December 14, 1928, pages 4-6 (Federal Reserve Board Papers);
Reports of the Agent General, September 1, 1928 to May 17, 1930, Vol. 1, page 107; and C. T.
Schmidt, German Business Cycles, 1924-1933 (New York: National Bureau of Economic Research,
Inc., 1934), page 71.
7

Infra, pages 165-66.




149

raised to71/2per cent toward the end of April and Germany's trade balance was
strengthened by the curtailment of domestic spending, the Reichsbank's published
international reserves were still 10 per cent lower in October 1929 than they had
been ten months before.
Britain was also hard hit by the change in the economic climate. Much of the
demand for funds that could no longer be satisfied in New York was shifted to
London. Despite an increasingly stringent Bank of England monetary policy, the
amount of new issues floated in London on behalf of overseas countries remained
substantial in the year ended June 1929. When the Bank of England's policy,
together with various market disturbances, finally brought a sharp reduction in
the flotation of new issues after mid-1929,8 many countries found it necessary to
draw down their existing London balances.9 Consequently, sterling came under
considerable pressure. Whereas in the year ended July 1928 the Bank of England's gold and dollar balances10 had increased $163 million to their peak for
1924-31, in the following sixteen months they dropped $298 million to their
lowest level prior to the summer-1931 crisis. In August 1929, when the pressure was especially severe, Norman had told the Committee of Treasury that
unless conditions improved Britain might be forced, along with other countries,
to abandon the gold standard.11
DIFFICULTIES OF COOPERATION. A variety of difficulties prevented the major central bankers from cooperating on any significant scale to deal with these pressures.
Even if they had wished, they could not have isolated themselves from the conflicts in which their countries were engaged. Norman, Moreau, and Schacht all
became involved in greater or less degree in the frequently acrimonious negotiations over reparations.12 At the same time, the determination of the United
States Government to keep out of the reparation controversy minimized Federal

8

Infra, page 159.

9

BIS Ninth Annual Report, 1939, page 81.

10

The gold figures used are those reported by the Bank of England. The dollar balances include those
on the books of the Federal Reserve Bank of New York and those reported to the Federal Reserve
Bank of New York by the only United States commercial bank at which the Bank of England is
known to have held a deposit.
11

Clay, op. cit., page 252.

12

Hjalmar Schacht, My First Seventy-Six Years (London: Allan Wingate, 1955), pages 233-39;
Clay, op. cit., pages 268 and 364; and The Economist, January 19, 1929, page 100.

150




Reserve involvement in European monetary problems. Beyond this, Harrison's
relations with his European counterparts were almost certainly constrained as a
result of the struggle for System leadership that developed between the New York
Reserve Bank and the Board after Strong's death in October 1928.
Monetary policy difficulties. In the literature, this struggle has usually been
discussed as if it took the form primarily of a controversy over the appropriate
means of handling the stock market boom, with the New York Reserve Bank
advocating the traditional discount rate-open market approach and the Board
supporting direct control over the end use of Federal Reserve credit.13 Actually
the controversy also involved a deeper conflict between the needs of the international financial system and those of the United States domestic economy. The
Federal Reserve had been able to cooperate fully and successfully in 1924-25 and
again in 1927 primarily because the monetary policy required to deal with the
domestic recessions of those years also fostered the System's international financial objectives. In 1928-29 this harmony was lacking. While speculative profits
and very high call money rates were drawing funds to New York from all the
world, few other signs of inflationary pressures were visible in the United States.
Industrial production expanded rapidly until July 1929, but wholesale prices
remained stable, and during the previous months significant declines were reported
both in construction and in new orders for durables. The business outlook was
especially clouded by the expectation among many thoughtful observers that the
stock market bubble was certain to burst sooner or later and by the fear that the
speculative collapse would have repercussions throughout the economy.
The policies advocated to deal with this complex situation were poles apart,
Norman and his colleagues being at one extreme, the Board of Governors at the
other, while Harrison attempted a reconciliation that leaned nearer to Norman
than to the Board. Uppermost in Norman's mind, of course, were the international repercussions of New York's high call money rates and the decline in
the outflow of United States capital. These threatened the gold standard generally
and sterling in particular.14 Accordingly, having postponed action during the

13
See, for example, Milton Friedman and Anna J. Schwartz, A Monetary History of the United
States, 1867-1960 (Princeton: Princeton University Press, 1963), pages 254-55; E. A. Goldenweiser,
American Monetary Policy (1st ed.; New York: McGraw-Hill, Inc., 1951), pages 152-53; and C. O.
Hardy, Credit Policies of the Federal Reserve System (Washington, D. C: The Brookings Institution, 1932), pages 131-36.
14

Clay, op. cit., pages 244-49.




151

autumn of 1928 in the vain hope that the pressure would ease after the turn of
the year, Norman traveled to the United States in late January 1929 to urge joint
corrective measures. On February 4, after he had had discussions in New York,
Norman clearly hoped that the Federal Reserve discount rates would soon be
increased, that this move would be accompanied by a rise in the Bank of England
rate, and that Federal Reserve rates at least would be increased further if necessary to break the speculative activity on Wall Street.15 The results that Norman
expected from this course of action are not recorded. However, it is clear almost
beyond doubt that he saw an incisive tightening of Federal Reserve policy and
the breaking of the stock market boom as steps toward both the decline of United
States interest rates to the levels that had prevailed in 1925-27 and also the revival
of the American capital outflow. If the Federal Reserve acted decisively, the strain
on the Bank of England's reserves would be severe but short and would be repaid
in the restoration of an international economic balance that would in the longer
run be favorable to sterling. The British strategy was clearly stated by Walter
Stewart, an American economist who had become Adviser to the Governors of
the Bank of England and had accompanied Norman on the trip from London.
A member of the Board staff, who discussed the strategy with Stewart, concluded
that the British felt Federal Reserve discount rates should16
be raised, at some unspecified time by a full one per cent with a view to breaking
the spirit of speculation, and then subsequently if necessary by another one per
cent, in order to provoke liquidation, and then after a fall in the stock market
similar rate action at the first sign of the next revival. By thus prostrating the stock
market and insuring that thereafter, as in the last three-quarters of 1926, there
would always be stock for sale on rallies, we should be cutting at the root of the
present situation and could rather promptly reduce rates thereafter and buy securities. I take it that such action would be received abroad as earnest evidence of the
right spirit of international cooperation. It is not high rates that any European
country shudders at; it is persistence of high rates.

Without doubt, Norman was prepared to argue that such an incisive and cooperative raising of discount rates as he proposed would serve the long-term
interests not only of sterling and the international financial system generally but

15

Clay, op. cit., pages 246-47; Hamlin Diary, February 4, April 18, and June 28, 1929.

16

Memorandum of February 7, 1929 on visit with Mr. Stewart on February 6, 1929, page 4
(Goldenweiser Papers 1929-31, Library of Congress).

152




of the United States economy as well. Harrison and his colleagues at the New
York Reserve Bank eventually came to share Norman's view but only after a
searching debate and as a last resort. A move in the Federal Reserve discount
rates to 6 per cent and perhaps even to 7 per cent or more would clearly have involved an extraordinary degree of stringency for the American economy. Already
in the early months of 1928 Federal Reserve policy had become progressively
tighter. By August 1, 1928 the discount rates of eight of the banks were at 5 per
cent, and heavy sales from the Open Market Investment Account had squeezed
the money market, with the consequence that member bank borrowings from the
Federal Reserve were equivalent to almost half of member bank reserves.
Returning from Europe at the beginning of August 1928, Strong had seen
the problem in all its complexity. He was deeply concerned to avoid: (1) "a
calamitous break in the stock market, a panicky feeling about money, a setback
to business because of the change in psychology", (2) "a precipitous decline in
the exchanges, especially sterling, which would weaken the bank position abroad",
(3) "embarrassment to the smooth operation of the Dawes Plan", and (4)
"restriction upon our exports". For all these reasons, he felt that there were "many
advantages in having the period of dear money as short as possible". He reiterated his doubts about any attempt to deal with the speculative tendencies in
the stock market: "a gradual unwinding of the situation is quite possible and is
the best bet".17 Two weeks later, when he was back in New York but under
doctor's orders to rest at home, his views had crystallized. He felt that some
of the difficulties that were being experienced might have been avoided if monetary policy had been tightened more vigorously at the beginning of 1928 but now
it was too late to tighten further. On the contrary, he was worried about the lagging effects of monetary policy: high rates now might have effects six months
or a year later in the "curtailment of construction and other business activities
which might affect business unfavorably". And he concluded that the "time has
come to take some of the pressure off the money market without disclosing our
hand more than is inevitable".18
The adoption in August of a System program along the lines indicated by

17
Strong letter to Stewart, August 3, 1928, pages 4 and 6. Most of this interesting letter to Stewart
is published in L. V. Chandler, Benjamin Strong, Central Banker (Washington, D. C: The Brookings
Institution, 1958) pages 460-62.
18

Strong memorandum to files, August 17, 1928, page 8 (Harrison Papers).




153

Strong and the repercussions of this program during the fall and early winter of
1928-29 were major factors in Harrison's subsequent decision to advocate cooperation with Norman. The easing of the pressure on the money market gave new
life to the stock market boom. Call money rates rose to new highs. In Harrison's
view, the threat both to the international financial system and to the domestic
economy had become greater; the longer the boom continued, the worse the
repercussions would be both internally and externally.19
Harrison elaborated the views of the New York Bank to the Federal Reserve
Board in Washington on February 5, 1929. The Reserve Bank's ultimate objective, he stated, was to get to a lower level of interest rates as speedily as possible.
The continuation of the present high rates for very long "would not only have a
directly detrimental effect on our domestic business and commerce, but would
force penalty rates of discount abroad and a possible consequent depression
which would indirectly but seriously affect our export markets". The stumbling
block was that the total volume of credit (in which he included loans extended
not only by banks but also by corporations and individuals) was expanding too
fast. So long as this expansion continued, the System could not achieve its objectives by simply going into reverse, i.e., by reducing discount rates and by buying
securities in the open market. The question that the System had to decide was20
whether we want to let the present situation go along until it corrects itself or
whether we should increase discount rates and through sharp incisive action quickly
control the long continued expansion in the total volume of credit so that we might
then adopt a System policy of easing rates.

Harrison also made clear that, if an increase in some or all Federal Reserve
discount rates were decided upon, it was his feeling that the move should not
be simultaneous with a Bank of England rate increase. Although a simultaneous
move "would have the greatest effect upon the control of credit", Harrison
"feared lest such concerted action might be misinterpreted and severely criticized to the point of militating against our effective cooperation with foreign
banks of issue".21

19

Harrison

letter to Platt, April 17, 1929, especially

pages 3-4.

20

Harrison conversation

with the Federal Reserve Board on February 5, 1929 (Harrison

21

Harrison conversation

with Mellon on February

154




5, 1929 (Harrison

Papers).

Papers).

The crucial question of course was whether the New York Reserve Bank's cure
would be worse than the disease, whether incisive increases in Federal Reserve
and Bank of England discount rates could break the stock market boom without
at the same time doing serious damage to economic activity generally. Although
this question was certainly uppermost in the minds of Harrison and his colleagues
at the New York Bank, their answer is not easy to document, perhaps for the
understandable reason that they were reluctant to have it recorded that the breaking of the speculative boom was a prerequisite for the achievement of their ultimate objectives. In its discussions with the Board, the New York Bank minimized
the immediate effects on business on the ground that the proposed discount rate
increase would be unlikely to raise significantly the already high cost of business
borrowing and argued that business could "better afford to pay a higher rate
for a short time than even present rates over too long a period".22 In a cable
to one of his directors in mid-March, Harrison revealed his views more clearly:23
We have tried everything but the bank-rate and I personally feel that apart from
all other considerations outside [i.e., money market] rates themselves now justify
using that.
With nearly $1,000,000,000 in discounts in the Federal Reserve System I believe
we still have control and that sharp incisive and if necessary repeated increases will
be effective.
Of course such a procedure may be costly and will require courage but will it be
more costly or require more courage than to do nothing in the face of the likelihood
of a long continuation of present or even higher market rates with their inevitable
effect upon economic and monetary conditions both here and abroad[?]

The fact is that Harrison and his colleagues at the New York Reserve Bank felt
that, if an early breaking of the stock market boom had adverse effects (hopefully
short-term ones) on the United States economy, these would be a small price
to pay for avoiding the disastrous consequences for both the domestic and
international economies that would result from a prolongation of the speculative excesses and from the inevitable and violent collapse of the speculative
bubble.

22
Unsigned memorandum
9, 1929, page 4 (Harrison
23

included in a letter from Harrison
Papers).

to Governor

Roy A . Young,

April

Harrison to Young, unnumbered cable, March 15, 1929 (Harrison Papers).




155

A fundamental difference in viewpoint as well as the struggle for leadership
led to the split between the New York Bank and the Board in the early months
of 1929 and to the disapproval of the 6 per cent discount rate established by
the New York Bank on February 14 and on ten occasions during the following
spring. Situated in the political capital, removed from New York's international
financial markets, the Board gave low priority to the needs of the international
financial system and focused its attention primarily on those of the domestic
economy. Thus, to cite only one illustration, the Board's famous direct action
letter of February 2, 1929 ignored the difficulties that were being experienced by
foreign countries as the result of the New York stock market boom and concentrated exclusively on the threat presented to United States trade and industry
by the "extraordinary absorption of funds in speculative security loans".24 The
interaction of the Board's concern with domestic problems and its assertion of
leadership is strikingly illustrated in a memorandum in which a member of the
Board staff summarized a discussion that he had with Governor Roy A. Young
on March 6, 1929. Young was not ready to approve sharp increases in discount
rates to break the speculative boom, according to this memorandum, chiefly
because he was still hopeful that the market would break without the need for
any further System action.25
His inclination against advances is partly because he thinks they do influence business and do not affect speculators, and partly, I am afraid, his desire to do it his
own way and not the New York bank's way. He feels that the New York bank
directors made an incredible blunder when . . . they passed the resolution
that they wanted their discount rate advanced right away, if at all. By doing that
they accomplished what they had never accomplished before—they made the Board
actually turn down their recommendation. Regardless of the merits of the immediate
situation, this is a great accomplishment because it raises the prestige of the
Federal Reserve Board in the system . . . it places the Board in the position of
protecting American business against high rates, which are wanted partly in view of
the international situation.
International concessions are not popular with Congress nor with the people . . . .
The Governor feels that Harrison unquestionably has a commitment (which I
believe probably amounts to nothing more than the desire to play the world game)

24

The Board's letter of February 2, 1929 to the Federal
Reserve Bulletin, February 1929, page 94.

25

Unsigned
memorandum
(Goldenweiser
Papers),

156




on talk

with

Governor

Reserve

Young

Banks

on March

is published

in the Federal

6, 1929, pages

3 and 4

to Norman to use his influence towards the adoption of radical rate advances.
That's the trouble with Harrison. He always presents a plausible story—a good
argument, and when he is all through he leaves an impression that he has an
ulterior motive. I think the impression is even more potent than the reality.

A day earlier Charles Hamlin recorded Young as saying that he felt Harrison
"lived and breathed for Norman".26
The result of the deadlock between the New York Reserve Bank and the
Board was a System policy that achieved the aims of neither. The Board's stand
against discount rate increases did not prevent the reversal of the easier policy
that had been followed in the latter part of 1928. The rate at which the New York
Bank bought 91-day acceptances was increased 1/4 per cent to 47/8per cent on
January 4, 1929 and was subsequently increased in several steps to 51/2per cent
on March 21, 1929. Consequently, Federal Reserve bills "bought" declined much
more than seasonally in January-July and the effect of this decline was accentuated by System sales of Government securities. Since the effect of these changes
was only partly offset by declines in currency circulation, gold inflows, and increased borrowing at the Federal Reserve, member bank reserves were slightly
lower on the average in July than they had been the previous December. Net
borrowed reserves had increased to $1,054 million in July, compared with $806
million in January (the first month for which these data are published). But while
the credit stringency had adverse repercussions on the outflow of capital from
the United States and on construction and other business expenditures, it failed
to dampen the stock market boom.27 The rate of increase in Standard and Poor's
composite stock index was almost as large in the seven months ended July 1929
as in the second half of 1928. Rates on new call loans moved steadily upward,
from an average of 7.61 per cent in the final quarter of 1928, to 8.07 per cent
in January-March 1929, to 8.69 per cent in April-June, and to 9.41 per cent in
July. The combined effects of the mild reversal of Federal Reserve policy and
the continued stock market boom only increased the strength of New York's pull
on foreign and domestic capital.
The increasing credit stringency, together with a growing conviction among the
members of the Federal Reserve Board that "direct action" was ineffective, led

26

Hamlin Diary, March 5, 1929, page 186.

27

See, for example, OMIC, preliminary memorandum, April 1, 1929, pages 4 and 5.




157

to a significant change in monetary policy early in August. For Harrison and
his colleagues at the New York Reserve Bank the objective remained a significant
reduction in the structure of market rates but the tactics were changed. At a meeting with the Federal Reserve Board on August 2, he indicated that the time had
passed when this objective could be attained through incisive increases in discount
rates and proposed an alternative twofold program. In order to satisfy seasonal
credit needs and generally to reduce market rates, the System should pump out
funds through market purchases of securities. At the same time, the System would
discourage any excessive expansion of credit and any impression that the lid
was off by increasing discount rates to 6 per cent. When some members of the
Board, who had previously been opposed to the discount rate increase, indicated
that they would approve provided that the increase was accompanied by a
reduction in the rate at which the New York Reserve Bank bought acceptances,
Harrison accepted the compromise.28 Accordingly the acceptance rate was
reduced 1/8 per cent to 51/8per cent, and effective August 9 the long sought discount rate increase was approved by the Board.
The results of this change in Federal Reserve policy were highly satisfactory
to Harrison when he assessed them early in October. He could hardly have been
surprised that the stock market continued to rise during August and early
September, and he was pleased to note that credit conditions were easing. In the
three months ended in October the System bought some $260 million in bills and,
other transactions largely canceling each other out, member bank discounts
dropped by $211 million while reserve balances rose $52 million. In October net
borrowed reserves were down to $843 million while the average rate for new
call loans at 6.10 per cent was 3.31 percentage points lower than in July. Writing
to Pierre Jay on October 7, 1929 Harrison summed up his impressions:29
the policy which we adopted early in August, of putting out funds through
the bill market under the protection of an effective six per cent rate, has thus far
worked much better than I had even dared hope. Bills have gone up, discounts have
gone down, and the total volume of Federal Reserve credit has expanded only in
proportion to the historic seasonal line. If we can continue this program so that
the total volume of discounts in the System will gradually decline to a figure much

28
Harrison conversations on August 2, 8, and 9, 1929 (Harrison Papers), and unsigned
dum of August 2, 1929 (Goldenweiser
Papers).
29

Harrison letter to lay, October 7, 1929 (Harrison

158




Papers).

memoran-

less than we have averaged during the past year, we will have perhaps taken the
first step looking toward a more natural relationship between discounts and the
total volume of Federal Reserve credit. On August 1, those discounts were over
85 per cent of the total of reserve credit, a higher percentage than at any time since
1920, and only by decreasing this percentage can we hope safely to pave the way
for an easing in the money situation.

In the meantime, the Bank of England had been acting to alleviate the pressure
on sterling. As Norman had forewarned when he was in New York, the bank
rate was raised to 5Vz per cent on February 7, 1929 from the41/2per cent at
which it had stood during the previous twenty-two months. Bank of England
operations kept the London money market tight, bill rates were held close to
the discount rate, and sterling was thus kept out of serious difficulties throughout
most of the first half of the year. In the late spring and early summer, however,
the election of a Labor government, continued conflicts over the reparation
settlement, the transfer of French balances from London to New York, and the
increasing credit stringency in the United States led to a renewed and serious
strain on sterling. The exchange rate against the dollar was at or only a little
above the gold export point from the beginning of June to mid-September, and
the Bank of England's gold holdings declined $133 million. Although Norman
did not wish to raise the bank rate again while the Hague conference was in
session, he intimated to Harrison that a rise was likely after the conference
ended.30 The Hatry scandal31 which broke in London on September 20 provided
a suitable occasion for action: six days later the bank rate was raised to 6 1/2
per cent and during the following two weeks seven other foreign central banks
also raised their rates.
The Bank of England action coincided with and, to an undeterminable but
probably small extent, contributed to the ending of the New York stock market
boom and to a consequent turn of events that was to bring sterling welcome
relief for a few months and deepening difficulties thereafter. The danger signals
had of course been flying in the American economy for some time before
mid-September.32 A 2 per cent decline in the industrial production index for

30

Norman

cables to Harrison,

31

Nos. 209, 211, and 250, August

Details of this complex stock fraud are discussed
562-63, and February 22, 1930, pages 417-18.
32

15 and 16, September

in T h e Economist, September

7, 1929.
28, 1929, pages

Supra, page 151.




159

July was announced to the press on August 31, and on September 22 the New
York papers carried an investment service advertisement warning investors about
"overstaying a bull market".33 In the circumstances, it was hardly surprising that
the stock market turned gently downward during the third week in September.34
Significant declines occurred on the days following the announcements of the
Hatry scandal and the increase in the Bank of England discount rate, but the
market steadied immediately thereafter and the decline was mild until the third
week in October.
In the meantime international payments flows were responding to the change
in market expectations and in interest rate relationships. The margin of the
London acceptance rate above the corresponding rate in New York, which had
been small throughout the previous summer, widened to more than 1 per cent
after the bank rate increase, while the margin in favor of New York on call
money rates narrowed sharply. At the same time, there were indications that
foreign balances were being withdrawn from the New York call loan market.35
The premium on three-month forward sterling, which had appeared in mid-1928
when the pull of the New York market began to be strongly felt in London
and which stood at 1/2 per cent per annum on September 14, 1929, had almost
disappeared by the month end. In October, the premium on sterling had changed
to a progressively growing discount. Spot sterling, which had fallen to a low of
$4.841/2on the day following the announcement of the Hatry scandal, was above
$4.85 a week later and above $4.87 on October 19. On October 24, before the
news reached London that the panic had already begun on the New York stock
market, Norman cabled Harrison to indicate his gratification at the turnaround
in the markets:36
Recent liquidation in your stock market and reduction in call m o n e y rates have
been satisfactory and have helped to re-establish international position.

33

J. K. Galbraith,

The Great Crash, 1929 (Boston: Houghton

Mifflin Company,

34

1955), page 97.

Standard and Poofs Corporation,
Daily Stock Price Indexes, 1926-1957 (New York, 1957),
pages 10-11; C. A . Dice and W. J. Eiteman, The Stock Market (3rd ed.; New York:
McGraw-Hill,
Inc., 1952), page 349, gives the peaks shown by various indexes of industrial shares as September
3, 7, and 19.
35

OMIC, preliminary

36

Norman

160




memorandum,

September

24, 1929, pages 4-5.

cable to Harrison, No. 297, October 24, 1929 (Harrison

Papers).

Market operations to support sterling. The difficulties of monetary policy
cooperation were paralleled in the field of Federal Reserve foreign exchange
operations during the sixteen months ended October 1929. Operations to support
sterling were frequently discussed but very little was done. While changes in the
international political and economic climate and in the leadership of the Federal
Reserve doubtless contributed to this result, central bank policy also played a
role. Early in the period, as previously noted, the Bank of England's international
assets were at their peak for the years 1924-31. Nevertheless, when sterling
came under pressure early in August 1928, Harrison proposed to buy sterling
both to support the British currency and as part of the program to ease United
States monetary conditions. However, the suggestion came to naught. The New
York Reserve Bank insisted that the proceeds be placed in sterling bills which
were chronically scarce in London; on the other hand, the Bank of England was
reluctant to add to the demand for such paper at a time when it was striving
to keep short-term rates as close to the bank rate as possible.37 In addition, Norman
seems to have felt that part of the pressure should be reflected in gold losses
that would impress the public with the continuing need for a strict monetary
policy and that he could afford to absorb the rest of the pressure by running
down the Bank of England's unreported dollar balances which at the time were
ample.38 He may also have wished to conserve the external assistance that was
available to him until the need for it was more urgent. After the turn of the year
another difficulty arose. The Federal Reserve was again squeezing the money
market and was running out of open market ammunition. Government securities
in the Open Market Investment Account were down to $40 million on April 1,
enough to offset the expansionary effects on the cash base of only a modest
amount of sterling purchases.39
As the pressure on sterling mounted during the summer of 1929, the need
to surmount these technical difficulties became urgent. On August 9, Harrison
telephoned to Norman that sterling had weakened somewhat upon the announce-

37
Harrison cables to Lubbock, Nos. 176, 183, and 201 of August 2, 15, and 31, 1928; Lubbock
cable to Harrison, No. 145, August 14, 1928; and testimony of Sir Robert Kindersley,
Macmillan
Evidence, Vol. 1, Question 1269.
38

Norman

cable to Harrison,

No. 165, September

5, 1928.

39

Harrison cable to Norman, No. 15, January 16, 1929 (Harrison Papers); OMIC preliminary
memorandum, April 1, 1929.




161

ment of the increase in the New York Reserve Bank's discount rate, and suggested
the possibility of Federal Reserve purchases of sterling in the amount of $25-50
million. He indicated that, although bills were preferred, he would be satisfied
if the sterling were invested half in bills and half "employed" at interest.40 After
a few days' consideration, Norman accepted Harrison's proposal. The fifty-fifty
formula was agreed upon, and it was decided that any purchases should be made
only when sterling was at the gold export point to New York.41 Harrison was
also concerned about exchange risks, particularly in view of the fact that Norman
himself had repeatedly expressed fears about the future of sterling. When all
other details had been settled, therefore, Harrison cabled that he assumed that
the Bank of England "would be prepared if necessary at any time to repay us in
gold or in dollars at New York for any sterling which we might buy", to which
Norman replied curtly "of course sterling is repayable in gold. That is the gold
standard."42 Apparently Harrison was satisfied by the response, and no arrangements were made to cover the forthcoming sterling purchases in the forward
market.43 Shortly thereafter, the directors of the New York Reserve Bank, acting
under power conferred by the Federal Reserve Act, authorized sterling purchases
of up to $25 million.44 In the event, only £3.3 million ($16 million) was bought
during late August and early September. The rise in sterling following the
September 26 increase in the Bank of England discount rate ended the need for
further operations and the whole amount was repaid on November 22.45
Longer term financing. Norman also discussed the possibility of longer term
financing with Harrison and with various New York commercial bankers during
the spring and summer of 1929. As foreigners turned increasingly to London
for funds, the Bank of England received credit applications that revealed a need

40

Harrison

conversation

with Governor

41

Norman

on August 9, 1929 (Harrison

Papers).

Norman cable to Harrison,
August 14, 1929.

No. 205, August

13, 1929; Harrison

cable to Norman,

No. 224,

42

No. 224, August

14, 1929; Norman

cable

No. 209,

Harrison cable
August 15, 1929.

43

Harrison

to Norman,

conversation

with Young on August 16, 1929 (Harrison

to Harrison,

Papers).

44

Unsigned memorandum
on meeting of the Board with Kenzel and Harrison, August 15, 1929
(Goldenweiser
Papers); Harrison conversations
with Young on August 16 and 19, 1929 (Harrison
Papers).
45

See cables between the Federal Reserve Bank of New York and the Bank of England, August 22September 6 and November 20-24, 1929.

162




not so much for sterling as for dollars. On May 10 Norman wrote Harrison that
he expected such applications to increase, especially if New York interest rates
remained high. As to the means by which the needed dollars could be obtained,
Norman saw several possibilities:46
first, that the advances should as far as possible come direct from you whence,
by the sale of Bonds, they would eventually be repaid: secondly, that the advances
should come from us European Banks which would lead to further increases of
Bank Rates in Europe. Unless you think there is yet another possibility—that the
Bank of France or we or some others should lend the money here and ourselves
directly re-borrow from the Federal Reserve Bank: this obviously raises a number
of questions!

The American response to Norman's suggestion was generally very sympathetic. Although Harrison indicated that Norman's third possibility should be
considered only as a last resort, he suggested that another gold credit might be
arranged along the lines of that negotiated between the Federal Reserve and the
Bank of England in April 1925. In addition, officers of J. P. Morgan & Co. made
it clear that their institution would be more than willing to do anything that it
could to help. But, by mid-August, when these offers of help were made definite,
Norman's situation had changed. He told Harrison that he was not in a position
to consider another 1925-type credit, and he discounted as unfounded all rumors
that the British government was about to obtain a large loan from the New York
market.47
The available records do not indicate what lay behind the shift in Norman's
position, but an explanation may well be found in the attitude of the Labor
government toward the level of the Bank of England's fiduciary issue. A year
before when the Conservative government had set the fiduciary issue at £260
million, members of the Labor opposition had criticized the figure as too low and
had argued in addition that the arrangements for changing the level of the issue
were unduly cumbersome.48 Having come to power at the beginning of June 1929,

46

Norman

letter to Harrison,

May 10, 1929 (Harrison

Papers).

47

Harrison cables to Norman, Nos. 171 and 224, May 23 and August 14, 1929; Harrison
memorandum on Federal Reserve Board meeting, August 14, 1929; unsigned memorandum
on Board
meeting with Kenzel and Harrison, August 15, 1929 (Goldenweiser
Papers); Leffingwell cable to
Grenfell, August 15, 1929 (Lamont
Papers).
48

Supra, page 141.




163

the Labor government apparently decided to deal with the pressures on sterling
not through foreign borrowing as Norman had originally contemplated but by
increasing the fiduciary issue. At any rate, early in August 1929 Norman did ask
the Bank of England's Committee of Treasury to consider whether such an
increase would be advisable in order to release gold for the defense of sterling.49
As events turned out, sterling's subsequent recovery made the adoption of special
measures unnecessary during the rest of 1929 and 1930, but these preliminary
discussions laid the groundwork for the changing of the fiduciary issue and the
credits from the Federal Reserve and Morgan's that were thrown into the unsuccessful attempt to save sterling during the summer of 1931.
THE ROLE OF THE BANK OF FRANCE. The international monetary problems of Britain
and Germany were complicated during the sixteen months ended October 1929
by pressures not only from the United States but also from France, whose
authorities were preoccupied with safeguarding their nation's interests in the
reparation settlement and were reluctant to increase the large balances that they
had accumulated in London during 1927 and early 1928.50
Ordinary financial relations were almost certain to be disturbed by the reparation negotiation in which France and Germany were the principal antagonists,
but there is evidence that the disturbances were aggravated during the latter
part of 1928 and early 1929 by official measures to strengthen the bargaining
positions of the main contenders. Moreau, who had no compunctions about using
financial power to support the aims of his government, did nothing to discourage
the substantial flow of French funds that was responding to the high interest rates
then prevailing in Germany.51 He may even have contributed to the flow through

49

Clay, op. cit., page 252.
Supra, pages 111, 117, and 118.
51
Moreau's attitude toward the use of financial power can be copiously illustrated from his Diary
which, however, ends in mid-1928. Toward the end of January 1928, Moreau was upset because the
Bank of England seemed to be stealing a march on France in negotiations with Yugoslavia. He wrote
in his Diary on January 27:
If the Bank of England takes away from us these customers, whom we are anxious to hold
for political reasons, I shall show my displeasure by buying gold in London.
On February 6, 1928 Moreau discussed with Poincare the conflicts between the Bank of France and
the Bank of England in numerous Continental countries and told him, according to the Diary:
We now have powerful means of exerting pressure upon the Bank of England. Would it
not be advisable to have a serious discussion with Mr. Norman, to try to divide Europe
into two zones of financial influence, allocated respectively to France and to England?

50

164




the transfer of Bank of France funds from New York and London. Large balances
in Berlin would give France leverage over Germany in the forthcoming
negotiation.52
Schacht, too, was preparing. He was aware that much of the inflow to Germany
in the second half of 1928 consisted of French balances,53 and he knew from
experience that the French were unlikely to be squeamish about employing such
financial weapons as they possessed to attain their reparation objectives. The
settlement in gold of the balance-of-payments surplus that Germany enjoyed in
the latter part of 1928 therefore accorded not only with Schacht's gold standard
principles but was a necessary precaution against the possible, if not the probable,
withdrawal of French balances. Of course, the bulk of the gold was taken from
London. Norman, seeking to relieve the pressure, suggested that Schacht increase
his holdings of sterling. But Schacht could only have cooperated at the cost of
weakening his own position.54
The test came during the third week in April 1929 when the Paris conference
became deadlocked over the amount and duration of the annuities to be paid by
Germany. To be sure, the January reduction in the Reichsbank discount rate
and nervousness about the outcome of the conference had been accompanied by
increasing pressure on the mark. The Reichsbank's gold and reported foreign
exchange dropped $40 million during the first quarter. As political tensions rose,
some $60 million more was lost in the first half of April. On April 18, shortly
after the conference breakdown, Pierre Quesnay of the Bank of France told an
American at the conference that $200 million would be withdrawn from Germany
by noon the next day.55 Schacht, who headed the German delegation to the
conference, related that56
the big banks of Berlin received from various French financial institutions
letters, couched in almost identical terms and referring in unmistakable terms to

52
Gardner
Papers).

memorandum

53

December

14, 1928, page 5 (Federal

Reserve

Board

Clay, o p . cit., page 244.

54

to Goldenweiser,

Ibid., page 245.

55

S. M. Crocker, "Notes on the Young Plan", page 147 (Young
Fred Bate, Secretary to Young
Committee.
56
Hjalmar Schacht,
1931), page 89.




The E n d of Reparations (New York:

Jonathan

Papers).

Quesnay

said this to

Cape and Harrison

Smith,

165

the Paris negotiations, announcing a curtailment of the credits which had been put
at Germany's disposal.

How far the withdrawals of balances from Germany were politically motivated,
as Schacht believed,57 or were merely the market's response to the deadlock is
unclear. Whatever the underlying causes, the last two weeks of April saw a
further $110 million drop in the Reichsbank's reported international reserves and
an increase in its discount rate to71/2per cent. However, the storm was quickly
over. By mid-May the deadlock had been broken, Norman was organizing central
bank assistance for the Reichsbank, but already funds were flowing back to
Germany, the Reichsbank's foreign exchange reserves were again rising, and no
help was in fact required.58
Only $128 million of the Bank of France's $434 million gold gain in the
sixteen months ended October 1929 can be attributed to the withdrawal of
French balances from Germany during the Paris conference. Another $177
million was acquired during the latter part of 1928 largely from domestic
dishoarding and from the United States. The remainder was acquired during
July-October 1929 when London was hit by heavy buying from Paris and when
British net exports of gold to France totaled $129 million after having been
negligible throughout the previous year.
The official French view was that the gold movement from Britain was merely
the natural result of the international movement of funds. In its annual report
of 1929, the Bank of France declared:59
From June to December we never took the initiative in acquiring gold by means
of foreign bills. We were obliged, in fulfillment of our obligation to regulate the
currency, to accept all gold of foreign origin which was offered to us over the
counter for francs, but we did not at any time intervene in the exchange market to
accelerate the pace of these gold imports. The bank has opposed no obstacle whatever to the free play of the money market under the regime of the gold standard.

But this was merely gold standard rhetoric which concealed more than it
revealed. In particular, it concealed the extent to which the gold movement arose,

57

Schacht, My First Seventy-Six Years, page 238.

58

Norman cables to Harrison, Nos. 100 and 115, May 6 and 16, 1929 (Harrison Papers).

59

Bank of France, Annual Report, 1929, published in the Federal Reserve Bulletin (March 1930),
page 113.

166




as in May 1927, from French institutional arrangements and financial policies—
from the Bank of France's lack of authority to conduct open market operations,
from the commercial banks' reluctance to discount at the central bank, and from
the piling up at the central bank of surplus Treasury and other official receipts
from the public. A large rise in such official balances did in fact occur during
the summer of 1929, while the expansion of the French economy and increases
in wages and consumer prices led to a significant rise in note circulation during
the whole second half of the year. Since only a small part of the consequent
pressure on the commercial banks was offset by increases in discounts and
advances at the Bank of France, the Paris market acted to avoid stringency by
drawing down its London balances. Sterling was sold for francs, the exchange
rate for the British currency fell to the gold export point to Paris, and French
commercial banks thus met their cash needs by selling gold to the Bank of
France.60
Undoubtedly it was true that the Bank of France was obliged to buy, at a
fixed price, all gold offered to it, but the statement quoted above from the bank's
1929 annual report sidestepped an important question. Since the bank retained
full authority, under the June 1928 monetary law, to operate in the foreign
exchange markets, it had a choice of meeting the Paris market's cash needs
either as it did, through the purchase of gold, or by acquiring sterling and thereby
keeping the British currency above the gold export point. However, the latter was
hardly an inviting alternative for the Bank of France, which considered that it
was already extending too much help to the London market. At the end of
June 1929, indeed, Bank of France holdings of sterling probably stood well
above the £ 70-80 million that Moreau had agreed to hold in June 1927.61
The conflict that developed at the Hague conference between the British and the
Continental delegations and the increasing questioning in the markets about the
future of sterling can only have stiffened the Bank of France's resolve to avoid
any further increase in its sterling commitment. Indeed, this commitment was
reduced, some $100 million having been transferred to New York in the four

60

The factors behind France's absorption of gold during 1929-30 are discussed by R. G. Hawtrey,
The Art of Central Banking (London: Longmans, Green and Co. Ltd., 1932), Chapter I, especially
pages 28-32; Balogh, op. cit., pages 442-60; League of Nations, International Currency Experience,
pages 38 and 39; and W. A. Brown, Jr., The International Gold Standard Reinterpreted, 1914-1934
(New York: National Bureau of Economic Research, Inc., 1940), Vol. II, pages 766-67.
61

Supra, page 119.




167

months to the end of October 1929. Even so, on the latter date the Bank of
France still held £ 9 7 million in London.62

FALSE HOPES: NOVEMBER 1929-JUNE 193O

In the eight months that followed the New York stock market crash, there seemed
a good chance that the vision shared by Harrison and Norman in February 1929
might be realized. Although production and prices were declining nearly everywhere, the contraction was fairly mild in Britain while in France, contrary to
the prevailing trend, the economy continued to expand. United States production,
after declining sharply in the second half of 1929, made a modest recovery in
the early months of the new year. It was clear that a considerable recession was
in progress, but this was the price that Harrison and Norman had felt might have
to be paid in order to eliminate the speculative boom and restore the equilibrating
outflow of long-term capital from the United States.
In fact, the crash brought a dramatic about-face in the exchange and international capital markets. In the following four months, foreign balances in New
York were repatriated in substantial volume. The major foreign currencies swiftly
recovered. Both the French franc and sterling were close to or at the gold import
point during much of November and December, and gold moved once again from
New York to London and Paris. Then in January-June 1930 there was a major
revival of new issues for foreign account on both the New York and London
markets. Germany was one of the main beneficiaries of this revival which,
together with a growing merchandise trade surplus, raised the gold and reported
foreign exchange holdings of the Reichsbank by mid-1930 to its 1924-31 peak.
Moreover, despite the revival in capital outflows, Britain continued to gain gold in
January-June 1930 when the gold inflow into the United States was also resumed.
Unfortunately, as noted below, these British and American gains were largely at
the expense of primary producing countries whose external positions were becoming increasingly desperate.
The revival of international capital flows was sought by the major central
bankers with rare harmony. They agreed that the revival was necessary for the

62
Crane memorandum to Harrison, November 30, 1929, "Report of Trip to Europe", September
11-November 20, 1929, pages 24 and 25.

168




success of the Young Plan and for the maintenance of exchange stability. In
addition, each had special reasons of his own. For Norman, the renewed American
outflow was desirable because it supported sterling and facilitated the reopening
of the London capital market. For the German authorities, foreign funds were
needed to cover the deficits both in the government budget and in the country's
balance of payments. Moreau sought to encourage an outflow of French capital
to enhance both Paris' position as an international financial center and the
political leadership of France in Europe, while Harrison saw the revived outflow
of American funds as a means to foster United States exports and thus counter
the contraction in domestic spending.
The revival in capital flows was facilitated by the virtually simultaneous easing
of monetary policy in the major financial centers. With America slipping into
depression, the Federal Reserve was no longer hampered by the conflict between
domestic and international aims. Within the System, to be sure, some emphasized
the need to liquidate unsound positions that had been built up during the
preceding boom and warned, especially when the stock market was staging an
ephemeral but spectacular recovery in April 1930, against the revival of the
speculative fever. Measures to support foreign central banks were also somewhat
inhibited by continuing suspicions in Washington that Norman, and perhaps
Moreau, too, had undue influence on the New York Reserve Bank.63 Even so,
the reversal of Federal Reserve policy was remarkably swift, the change being
led by the New York Reserve Bank. Its discount rate was reduced to 5 per cent
on November 1, 1929, to41/2per cent two weeks later, and in four additional
steps to21/2per cent on June 20, 1930. At the same time, open market operations quickly changed direction. Over $400 million of Government securities was
bought in the eight months beginning November 1, 1929, more than the amount
that was sold when the markets were being squeezed during January-August
1928. During January-June 1930 a heavy reduction in note circulation and the
renewed rise in the gold stock acted together with open market purchases to pump
out reserves and to facilitate a sharp reduction in member bank borrowings at the
Federal Reserve. In June 1930 net borrowed reserves had dropped to $197
million from more than $900 million a year earlier, member bank reserves were
slightly higher, while the total of currency held by the public and commercial

63

Hamlin Diary, January 20, 1930, page 90.




169

bank deposits was only slightly lower.64
In Europe, too, monetary policy was eased significantly, although the change
was not so large as in the United States. The Bank of France's discount rate,
which had been at 3 1/2 per cent since the beginning of 1928, was reduced to 3
per cent at the end of January 1930 and to 2 1/2 per cent on May 1. Moreover,
the effects of the continued rise in note circulation were more than offset in
January-May by an apparently fortuitous rundown of official deposits at the Bank
of France, which led to a distinct easing of money market conditions.65 In the
fiscal area, the government, at the request of the Bank of France, acted during the
winter of 1929-30 to reduce a variety of taxes that were discouraging the outflow
of French long-term capital.66 Of necessity, the movement toward ease in Germany and Britain was considerably more cautious than in the United States and
France. Both the Reichsbank and the Bank of England acted more or less in step
with the New York Reserve Bank, but their discount rates were kept generally
above New York's. By June 1930, when the discount rates in New York and
Paris were down to 2 1/2 per cent, London's and Berlin's stood, respectively,
at 3 per cent and 4 per cent.
The easing of monetary policy, combined with the sharp decline in economic
activity, brought a major drop in money market rates in all the major centers.
In the capital markets, however, the reaction was less satisfactory. In New York,
yields on United States Government and other prime long-term securities were
significantly lower in January-June 1930 than they had been a year earlier but
yields on lesser quality bonds were virtually unchanged. Elsewhere, the cost of
long-term borrowing was still at about the same level as in the first half of 1929,
even for the British government; in Paris, it was distinctly lower but, despite
official efforts, the market for the general run of foreign issues remained all
but dead.67
Even so, the revival of international lending during the first half of 1930 was
remarkable. The total of new issues for foreign account floated in the United

64
Friedman and Schwartz, o p . cit., pages 712, 713, and 739; Board of Governors
Reserve System, Banking and Monetary Statistics, page 371.
65

Hawtrey,

of the Federal

o p . cit., page 29.

66

Bank of France, Annual Report, 1930, published in the Federal Reserve Bulletin (March 1931),
page 147.

67

The Economist, May 10, 1930 (Banking Supplement), pages 13-14, and August 9, 1930, page 278.

170




States was $701 million, second only to the record amount issued in JanuaryJune 1928. Similarly, in Britain new issues for overseas countries totaled $331
million, an amount that had been exceeded only in the first halves of 1928 and
1929. In both markets, the issues (apart from the Young Loan) were primarily
on behalf of non-European governments, those raised in New York being largely
for the Western Hemisphere, Australia, and Japan, those in London for British
Empire countries.
On two occasions, the Paris market did come to life when it played a significant
but not entirely helpful role in the flotation of securities to support the Young
Plan. In accordance with agreements negotiated earlier in the year, the BIS
shares were issued on May 20 and the Young Loan on June 12 and 13 in Paris
as well as in London and other European centers and in New York. In Paris the
response was very large, both issues being heavily oversubscribed.68 Unfortunately
the ease that had prevailed earlier in the Paris market was already giving way to
renewed stringency, and this stringency was accentuated by large transfers of
funds to the Bank of France in subscriptions for the new issues. As in earlier
periods, the stringency in Paris was relieved by a flow of funds from London so
that, after a three-month lull, gold again began to move from Britain to France
in substantial volume. This being the outcome, there must have been serious
doubts, at least at the Bank of England, whether the benefit of French official
participation in such international efforts was not heavily outweighed by the cost.
sion. Especially in Britain, Germany, and the United States, unemployment

STRAINS DURING THE DEPRESSION: JULY 193O-MAY 1931

DETERIORATION. Hopes that the depression would be mild and that a renewed
outflow of United States capital would restore the international economic balance
were undermined by the further deterioration of the international economy during
the eleven months that followed the flotation of the Young Loan. The mild
recovery of United States industrial production in early 1930 proved abortive;
from April on, output dropped sharply as it also did in Germany. In France, too,
economic expansion gave way after mid-1930 to a gradually deepening depres-

68
The Economist, May 24, 1930, page 1160; The Commercial and Financial Chronicle, June 14,
1930, page 4107, and June 21, 1930, page 4296.




171

became a major problem, even on occasion a threat to political stability. Primary
producing countries, too, suffered from declining exports to the industrial countries and from continued high levels of raw material and food production. The
decline in primary commodity prices steepened and, with the worsening of the
primary producers' terms of trade, the pressure on their international positions
became severe.
As the economic climate worsened, confidence in the international financial
markets was hit by a series of shocks. Defaults on foreign bonds increased. The
currencies of several primary producers were devalued. Eastern European and
South American governments were overthrown by coups.69 In the United States,
the stock market resumed a sharply downward course and import duties were
raised over protests both from abroad and from a large body of American
economists. Confidence was further weakened in the United States, Britain, and
Germany by rapidly rising budget deficits and in the United States and France
by a growing number of bank failures. In Germany, the summer of 1930 saw an
ugly outburst of nationalism, and September, the rise of the Nazis from an
insignificant minority to a major national party. With the ink on the Young Plan
hardly dry, the markets began to hear rumors during the fall that the German
government would ask for a reparation moratorium.
For their parts, the monetary authorities in the major countries, having eased
credit policies very considerably in the eight months following the crash, did
little more. No further discount rate reductions were made until, at the turn of
the year, the New York Reserve Bank and the Bank of France both cut their
rates to 2 per cent;70 thereafter, the New York Reserve Bank went to11/2per
cent early in May 1931 and the Bank of England followed with a reduction in
its rate to 21/2per cent. In both the United States and Britain, the central banks
followed an extremely cautious open market policy. From late June 1930 on,
Harrison vigorously advocated a policy of open market purchases that aimed,
through the pumping of funds into the money market and the further reduction
of short-term interest rates, to support a revival of the bond market.71 But this

69

Report of the Chairman
April 27, 1931, pages 3-4.

of the Open Market

70

Harrison

conversations

71

Harrison

letter to Seay, July 3, 1930.

172




with Harvey

Policy

and Lacour-Gayet

Conference
on December

to the Governors'
22, 1930 (Harrison

Conference,
Papers).

proposal was not accepted. In the eleven months ended May 1931, System purchases of bills and Government securities totaled only $31 million which, together
with gold purchases of $239 million, did little more than offset the drain on
member bank reserves from increased currency circulation. At the end of the
eleven months that saw an increasing number of bank failures, rising currency
hoarding, and a significant decline in bank deposits, member banks still showed
net borrowed reserves of almost $100 million. In Britain, the Bank of England,
which previously had at least partly offset the effects of gold movements on
bank reserves, reversed its policy. During the eleven months beginning July
1930 open market sales reinforced the cash drain that resulted from the outflow
of gold so that bankers' balances at the Bank of England in May 1931 averaged
5 per cent lower than a year earlier.
The monetary policies of the major countries combined with the continued
decline in production and prices and the rise in defaults, devaluations, and
political disturbances to kill the revival of the long-term capital flow. In the year
beginning July 1930, new issues for foreign account fell in both Britain and the
United States to only about one third of the rate in January-June 1930.
The international disequilibrium again became manifest. Despite the enlarged
outflow of long-term capital during the first half of the year and a slight decline
in its current account surplus, the United States payments balance changed from
a $53 million deficit in 1929 to a $598 million surplus in 1930, and the surplus
seems to have run at an even higher rate during January-May 1931. France's
surplus was almost $300 million in 1930, compared with $125 million the year
before. Germany's position, on the other hand, moved into deficit, the Reichsbank's gold and reported foreign exchange balances dropping by about one fifth
in the eleven months ended May 1931. Britain, too, was under severe pressure,
especially from France, to which it exported almost $300 million in gold during
the eleven months, somewhat more than South Africa's production of gold during
the whole of 1930. However, Britain succeeded in passing on much of this pressure to other countries, notably the primary producers which, being in desperate
straits themselves, were heavy sellers of gold not only in New York but also in
London. Consequently, the decline in the Bank of England's gold holdings was
comparatively modest.
CENTRAL BANK COOPERATION: MUCH ACTIVITY; LITTLE ACCOMPLISHED. A l t h o u g h

the

international economy was visibly deteriorating, the threat of an imminent and
major crisis was required to open the way for significant measures of central




173

bank cooperation. In the absence of such a threat, much of the eleven months
immediately prior to the outbreak of the 1931 crisis was given over to bickering
and sparring for position. The Micawberish attitude of the Hoover Administration
and the continuing struggle over leadership within the System prevented any
vigorous initiative on the part of the Federal Reserve. With little freedom of
action, Harrison had to tread carefully.72 The kind of intimacy that had existed
between Strong and Norman disappeared. Although Harrison visited the major
European central banks in April and again in November 1930 and Norman
came to New York in the spring of 1931, the Federal Reserve was not represented at the BIS monthly meetings and remained perforce on the periphery of
the discussions among the European central bankers. And of course these
discussions, being affected by continuing political tensions between Germany and
France, were far from harmonious. Minor irritations added to the strains—as
when the Bank of England failed to mint an adequate supply of fine gold
bars and instead offered standard bars which the Bank of France considered
unacceptable.73
Short-term support for the reichsmark and sterling. The only currency
support operations of which we have any record in the eleven months ended
May 1931 occurred after the September German national elections in which the
Nazis scored large gains. In the month that followed, Germany suffered a heavy
outflow of funds as the result of which the Reichsbank lost some $250 million
in gold and foreign exchange, mostly in the first few days after the election.74
Apart from vigorous intervention by the Reichsbank itself, operations to support
the reichsmark apparently were undertaken only by the BIS. But the form of
this support was curious and the scope small. Although by far the largest share
of the BIS assets was held in the United States, its dollars were not employed.
Rather, despite the delicate condition of the British currency at the time, the

72
Harrison was continually being sniped at from the Board. Some of its members objected on legal
grounds to the New York Reserve Bank's foreign exchange operations. Others felt that Harrison
should not go abroad without the Board's consent and, foreshadowing subsequent amendments to
the Federal Reserve Act, that legislation should be adopted to subject any consultations
between
the New York Reserve Bank and foreign central banks to the supervision of the Board. Hamlin
Diary, October 22, 1930 and March 26, 1931, pages 99 and 199.
73
Moreau letter to Harrison,
pages 7 and 8.

June 13, 1930; Allan Sproul memorandum

74
BIS cable to Federal Reserve Bank of New York, No. 130, September
Sackett, February 27, 1931, page 11.

174




to files, January 21, 1931,
23, 1930; Luther letter to

BIS chose to sell sterling for dollars and to transfer the proceeds into reichsmarks, switching about $2 million equivalent in this manner between September
23 and October 1.75 Although this support was given on a short-term basis,
the continued weakness of the German currency made it impracticable for the
BIS to reduce its reichsmark assets until special arrangements were made for
their partial liquidation during 1932.76
Toward the middle of September 1930 Harrison cabled Norman that sterling
had suffered a sudden drop in New York which was not explicable on seasonal
and other technical grounds alone and raised the question whether the rate could
be most conveniently supported through the Bank of England's sales of dollars
or Federal Reserve purchases of sterling.77 Initially Norman indicated his preference for the first alternative but subsequently agreed that Federal Reserve sterling
purchases would be helpful. Acting under their statutory power, the New York
Reserve Bank's directors on October 9 authorized sterling purchases of up to
£ 5 million. There were the usual difficulties about investing the sterling in
commercial bills which the Bank of England said it could not obtain at the
moment without "breaking the rate".78 On this occasion, the New York Reserve
Bank took a flexible position, and it was agreed that the proceeds would be
placed in deposits at the Bank of England with the understanding that these
would be converted "into bills as and when they can be obtained without inconvenience to you". It was also understood that the operation would not be reversed
until after the turn of the year when sterling was expected to be stronger.79
Under these arrangements, the New York Reserve Bank acquired sterling with
the prime objective, as on earlier occasions, of preventing the London market
from losing gold to New York. Initially the support operations were quite
vigorous, purchases being made at the rate of £ 500,000 daily between October
14 and 17. Purchases were suspended when sterling rose somewhat above the
gold export point to New York but were resumed during the final week of

75
McGarrah cable to Harrison, No. 9, September 23, 1930; BIS cables to Federal Reserve
New York, Nos. 130, 132, 136, September 23, 24 and 30, 1930.

Bank of

76
Roger Auboin, The Bank For International Settlements, 1930-1955, Essays in International
(Princeton: Princeton University Press, May 1955), page 9.

Finance

77

Harrison cables to Norman, Nos. 216 and 218, September 12 and 16, 1930.

78

Harvey cable to Harrison, No. 230, October 9, 1930.

79

Harrison cable to Harvey, No. 244, October 10, 1930.




175

October. By the end of the month, £ 4 . 7 million had been acquired (all of
which had been invested in bills) and authority was obtained to acquire £ 5
million more.80 However, the feeling was strong at both the New York Reserve
Bank and the Bank of England that £ 1 0 million should be the ceiling.81
O. M. W. Sprague, who had succeeded Walter Stewart as Adviser to the
Governors of the Bank of England, told Allan Sproul who visited him at the
Bank of England in early December 1930 that,82
Except in connection with a temporary condition, such purchases merely postpone
the setting in motion of the forces which will really correct the underlying situation.
In this case they would foster the willingness of England to take her difficulties
sitting down, instead of standing against them.

Using its additional ammunition conservatively, the New York Reserve Bank
acquired only £2.5 million during November and December so that at the yearend its holdings totaled £ 7 . 2 million. Although sterling was under considerable
pressure at times during January 1931, no more was acquired by the Federal
Reserve, and with the concurrence of the Bank of England the whole amount
was gradually liquidated during February and March at rates that were only
fractionally higher than those at which it was bought. Even so, the objective of
the operation was attained: London lost virtually no gold to New York during
six months of considerable strain.83
The Bank of France, too, supported sterling during the autumn of 1930, acting
apparently during early November when the failure of several French banks
precipitated difficulties in the Paris market and heavy withdrawals of French
funds from London.84 However, the Bank of France's intervention was limited
to averting what appeared to be a critical situation.85 Once the emergency had

80

Harrison

letter to Young,

81

Harrison

cable to Case, No. 2, November

82

Allan Sproul memorandum to files, January 21, 1931, page 8.

83

Crane letter to Young, March 20, 1931.

84

T h e Economist, November

85

October

17, 1930; Harrison

letter to Seay, October

14, 1930, page 1 (Harrison

31, 1930, pages 3-4.

Papers).

8, 1930 pages 853-54; Crane letter to Black, November

21, 1930.

Apparently
the demand for cash in Paris was so great that it could not be met either by selling
gold to, or borrowing from, the Bank of France. Gold could not be shipped fast enough
from
London. In order to prevent a money panic, the Bank of France pumped funds into the market by
buying sterling
balances.

176




passed, it quietly sold most of the sterling to private bankers and resumed its
passive role in the exchange markets.86 In the three months beginning November
1930, British gold exports to France totaled $178 million.
The search for longer term solutions. The eleven months before the outbreak
of the international financial crisis in the summer of 1931 saw an intensified
search for longer term means to reduce the pressure on the reichsmark and
sterling. In Germany, the conservative Bruning government was convinced that
no solution was possible unless the payment of reparations was stopped and debated only the strategy and timing to attain this objective.87 At the same time,
Luther, who had succeeded Schacht as President of the Reichsbank, carried on
without much hope a desperate attempt to stave off a new financial crisis. The
German government was virtually at the end of its international credit. Between
April and November 1930, it had obtained almost $350 million of long- and
medium-term funds from abroad under the Young Loan and from international
consortia headed by Kreuger and Toll and Lee-Higginson. Despite repeated
attempts at retrenchment, the budgetary deficit was still increasing, and at the
end of 1930 foreign short-term claims on Germany were still more than three
times as large as the Reichsbank's total holdings of gold and foreign exchange,
a ratio that was only slightly smaller than it had been a year earlier.88
From Luther's point of view, the one hopeful aspect was that United States
commercial bankers, who held almost two fifths of Germany's short-term debt to
foreigners in March 1931, 89 were deeply worried about the soundness of their
claims. An official of Lee-Higginson, in Berlin to discuss yet another loan to the
German government, told the American ambassador, Frederic M. Sackett, late
in January 1931 that "the short loans to Germany were now in such volume that
they could not be called or renewals refused without great danger to the financial
situation in the United States".90 A month later, Luther sent Sackett a memo-

86
Allan Sproul memorandum
to files, January 31, 1931, on report of conversation
the Bank of France, pages 31-34.

with Cariguel of

87

E. W. Bennett, Germany and the Diplomacy of the Financial Crisis, 1931 {Cambridge:
University Press, 1962), pages 9-26.

88
Report of Committee
Appointed
on the Recommendation
August 18, 1931), hereafter referred to as Wiggin Report.
89
Wiggin Report, page 3; Board
Monetary Statistics, page 585.
90

of Governors

of the London

of the Federal

Reserve

Conference

System,

Harvard
(Basle,

Banking and

Quoted in Bennett, op. cit., pages 38-39.




177

randum that provided detailed statistics showing the weakness of Germany's
short-term position, emphasized that the country's financial difficulties were
insoluble so long as reparations continued, and proposed, merely as a palliative,
a long-term loan to fund perhaps $350-475 million equivalent of Germany's
short-term liabilities.91 Sackett, deeply disturbed, passed the information he had
received from the American bankers and Luther on to Washington where copies
of at least one of his dispatches were sent to President Hoover and Secretary of
the Treasury Mellon. Luther himself sent a copy of his memorandum to Harrison.
Not surprisingly the proposal for a refunding loan fell on barren ground, but
Sackett's dispatches nurtured the seeds that were to grow during the following
four months into President Hoover's moratorium on reparations and war debts.
Means of strengthening the position of sterling were also under intensive
discussion during the latter part of 1930 and early 1931. In mid-November, when
sterling was under severe pressure from the withdrawal of French balances in
London, Pierre Quesnay, who had become the General Manager of the BIS, was
urging that Britain float a long-term loan in France in order to reduce its
short-term liabilities to foreigners.92 Early in December, Norman discussed the
question with Clement Moret, who in September 1930 had succeeded Moreau
as Governor of the Bank of France. As he related it to Harrison, Moret told
Norman that93
public opinion in France, as well as the Government and the Bank of France,
would look very favorably upon the issue of English private loans in France, or
upon French participation in an English conversion operation, if such a course were
technically possible of adoption. Nor did I hide from Mr. Norman that if he
thought well of it, I should certainly be disposed, as has been done in the past, to
facilitate the rediscount of sterling bills by French banking groups, which bills they
would acquire by agreement with financial institutions in London.
Our conversations did not result [in] any concrete solution: nevertheless I was very
glad, for my part, to talk in all frankness with Mr. Norman, and I hope that what
I said will again have given him the impression that we here are disposed to
cooperate fully and loyally with the Bank of England.

91

Luther letter to Sackett, February 27, 1931.
Sproul memorandum to files, January 21, 1931, page 6.
93
Moret letter to Harrison, December 10, 1930, pages 4-5.
92

178




However, Norman was bent on a different course. Early in February 1931
he presented to the BIS governors a proposal that in some ways anticipated the
International Bank for Reconstruction and Development. This proposal, which
apparently originated with Sir Robert Kindersley, envisaged an International
Corporation with headquarters in Switzerland or Holland, subscribed capital of
£25-50 million, authority to issue bonds up to three times the amount of
subscribed capital, the proceeds of the bond issues to be loaned to various
governments and other official and private entities that could not obtain needed
funds through customary channels and that could offer good security. The crux
of the proposal was that, whereas France and the United States were expected to
absorb the bulk of the bonds issued, the corporation was to be managed, under
the initial version, by a board to which each subscriber of a specified amount
of capital could appoint a director. Under a version that Norman espoused later,
the BIS was to assume the entire responsibility for the organization and management of the corporation and would appoint its president and a majority of its
directors.94
This proposal, as Norman described it to Harrison, was a trial balloon:95
the BIS is already slipping to the bottom of a ditch and in that position seems
likely to do no more than helpfully perform a number of routine and Central
Banking transactions. A change of direction and of spirit is needed in order that
it may also fulfil the hopes for its wider usefulness which were expressed, although
perhaps vaguely, in the Young Plan: to these the German Directors are forever
drawing attention and to their non-fulfillment they may in the future attribute
difficulties which arise.
Some such scheme as the one we have under discussion would serve as a good
occasion to enable the BIS to kill the two birds with one stone. But frankly I can
find little hope of this scheme (or any other designed to open up the Capital
Markets) being entertained by the [BIS] Board: there are a number of reasons for
my doubts but perhaps the principal obstacle is the unwillingness, which we have
seen in certain quarters, to support a scheme of which the control and the funds are
truly international.

Norman had reasons to be gloomy. Except for Luther, the governors of the

94
Unsigned memorandum of February 2, 1931 sent by Norman to Harrison under cover of letter
dated March 3, 1931 (Harrison Papers); McGarrah cable to J. P. Morgan, March 12, 1931.
95
Norman letter to Harrison, March 3, 1931.




179

BIS had shown little interest in the proposal, and both the French and the
Americans were decidedly cool. The reaction of the major Paris commercial
banks, to which the plan had been transmitted by the Bank of France, was that
the scheme was too big, that its chances of realization were very small, and above
all that, as the Paris market was expected to furnish most of the funds, it would
be necessary to have full French control over the institution. This reaction was
entirely shared by the Bank of France.96
The views of the New York Reserve Bank were reflected in a memorandum
to Harrison from J. E. Crane, Deputy Governor in charge of the foreign function.
The proposal, Crane wrote, was impractical; international control raised serious
difficulties and would not "dissipate present reluctance in New York to purchase
foreign securities"; the Corporation would have only "second grade clients". As
for the role of the BIS in the plan, Crane stated that "it has always been felt that
these larger functions were rather visionary and inflationary in character and
should be left severely alone for the present.97 J. P. Morgan and his colleagues
cabled Gates McGarrah, President of the BIS, in a similar vein, adding that
the New York market "is not accustomed to this form of international organization of credit" and would be unwilling to surrender its judgment to such an
institution which would, in any case, be open to political influence abroad and
to invite attack from politicians in the United States. Morgan's was opposed to
"the intervention of artificial agencies" and felt that the best results would be
obtained in the long run if new foreign issues "have to meet the tests of the
market at the time of issue on their own merits.98
While the considerations behind the failure of the Kindersley proposal are
clear enough, the reasons why Norman did not take up Moret's offer are largely
a matter for conjecture. Our records reveal nothing about the conditions attached
to Moret's offer except that the French insisted, not unreasonably, that any
British bonds issued in Paris be denominated in francs. This condition the British
authorities were unwilling to accept,99 presumably because it implied doubts

96

Lacour-Gayet

97

letter to Crane, February

Crane memorandum
12, 1931.
98
99

J. P. Morgan,

to Harrison,

T. W. Lamont,

Hamlin Diary, December

180




24, 1931.

"Kindersley

Plan—Bank

for International

and S. P. Gilbert cable to McGarrah,

20, 1930, page 132A.

Settlements'*,

March 13, 1931.

March

about the position and future of sterling. They doubtless felt also, on the basis
of the May-June 1930 experience, that any sizable flotation in Paris would only
increase both the withdrawal of French balances from London and the movement
of gold from Britain to France and thus offset much, if not all, the benefit that
might otherwise be obtained from a loan. Although these problems would not
have arisen under the alternative proposals100 that Moret discussed with Norman,
it would be surprising—in view of the moves that the German government was
making at the time toward a reparation moratorium and a customs union with
Austria—if France had not employed all its bargaining strength to obtain firm
commitments of political support from Britain.101 In London, however, the government was unwilling to give up its freedom of action, and the proposal for an
International Corporation that, as the memorandum presented to the BIS
indicated, would grant loans "on a purely business basis" seems to have been an
attempt to obtain at least indirect support for Britain's financial position while at
the same time avoiding new political commitments.
One final puzzle remains. Available records give no sign that, during the eleven
months before the outbreak of the crisis, Norman discussed the possibility of
raising long-term funds in New York—where they could almost certainly have
been obtained on terms that would have been acceptable to the British authorities.
Although Norman was in the United States from March 27 to April 14, 1931,
there is no indication that the subject was mentioned. The explanation seems to
be that by mid-March, when it was clear that the Kindersley proposal was dead,
the international financial climate seemed to be improving. Norman was more
hopeful about the German outlook.102 Both sterling and the reichsmark were
strengthening somewhat on the exchange markets, and the heavy movement of
gold from Britain to France had temporarily stopped. In the four months
following their low point at the end of January 1931, the Bank of England's
gold and dollar reserves increased by $59 million. It would have been easy that
spring to hope that the worst was over.

100

Supra, page 178.

101

Bennett, op. cit., pages 44-62 and 83-85.

102

Harrison letter to Stimson, April 21, 1931 (Harrison Papers).




181

8. The 1931 Crisis
INTRODUCTION

The brief respite enjoyed by sterling and the reichsmark in the spring of 1931
belied the dangers ahead. The foundations on which the international financial
system had been built were already weak, and the forces of destruction were
poised. The network of financial claims that bound the international economy
together was under severe strain as a result of the world depression. This network
had been built on confidence among investors—mainly in the United States and
Britain—that, if they made allowances only for ordinary business risks, they
could obtain the relatively high earnings offered abroad without placing their
capital in jeopardy. However, as prices, production, and incomes declined, the
basis for this confidence became increasingly tenuous. Public authorities, whose
revenues were declining sharply, defaulted on bonded indebtedness. Banks, which
had borrowed abroad to enlarge their investable funds, found their assets frozen
in firms on the verge of bankruptcy. All over the world those who held bank
deposits, both in their own countries and abroad, became nervous about the
safety of their funds. The authorities, in their attempts to stave off financial panic
and to restore confidence, increased interest rates and/or cut government spending in a futile attempt to balance budgets but succeeded only in making matters
worse by smothering any remaining incentive to invest and by further reducing
already diminishing incomes.
Even before the spring of 1931 the deepening deflationary spiral had been
accompanied, as noted in Chapter 7, by an increasing number of defaults and
devaluations. However, these had occurred mainly among the primary producing
countries. The major shock to the system came with the breakdown in the
industrial heart of Central Europe where both United States and British investors
had especially heavy commitments. Locked into loans to firms afflicted by the
depression, the Credit-Anstalt,1 Austria's largest commercial bank, had short-term
liabilities to foreigners totaling $100 million on May 11, when its difficulties
were announced to the public.2 The bulk of these liabilities were to British and

1

Abbreviation for Oesterreichische Credit-Anstalt fur Handel und Gewerbe.

2

Morgan & Cie. to J. P. Morgan & Co. (cable) message from L. Fraser for S. P. Gilbert, May
26, 1931.

182




United States interests, but significant amounts were also owed to Germany,
France, and other European countries.3 Thus all the major countries were involved, but the most susceptible to trouble was Germany which itself had shortterm liabilities to foreigners that probably totaled between $2Vi-3 billion at the
end of March 1931 and whose largest creditors (as in the case of Austria) were
the United States and Britain.4 The danger was plain. A chain of deposit withdrawals, panicky runs on the banks, and moratoria could spread from Austria
to Germany and even to Britain and the United States. The threat to exchange
stability was especially grave because in Austria, Germany, and Britain shortterm liabilities to foreigners were a multiple of total official gold and foreign
exchange reserves, because all but relatively small amounts of these reserves
were tied up as cover for liabilities of the central banks, and because—given the
psychology of businessmen at the time—recourse to special measures to release
additional reserves for the defense of the currency was more likely to aggravate
than to alleviate market fears and uncertainties.

GERMANY AT LOGGERHEADS WITH FRANCE

The authorities in the major countries recognized almost as soon as the CreditAnstalt difficulties were announced that they were confronted with a serious
international financial crisis, but their efforts to deal with this crisis cooperatively
were thwarted by a fundamental political conflict, i.e., by Germany's drive to
throw off the obligations imposed on it under the Versailles Treaty and by
France's insistence that these obligations be fulfilled.
By the spring of 1931 immense pressures had built up in Germany to throw
off what were regarded as the shackles of the Versailles settlement on which all
the country's ills were blamed. Industrial production in the second quarter was
down one third from the level of two years before; almost two fifths of the industrial labor force was unemployed; street fights between extremists of the right and
left were increasing. The Bruning government which, backed by the army, had
been ruling by decree since October 1930 was desperately seeking diplomatic

3
Morgan Grenfell & Co. cable to J. P. Morgan & Co., May 20, 1931; McGarrah cable to Harrison,
No. 13, May 27, 1931.
4

Wiggin Report, pages 1-10.




183

victories to increase its domestic political support.5 Accordingly, it risked
increased French hostility by announcing in mid-March arrangements under
which Germany and Austria proposed to form a customs union and subsequently
by pressing for immediate and substantial relief from reparations.
In both Britain and the United States, there was understanding and some
sympathy for the problems and aims of the Bruning government. Indeed, ever
since the Balfour note of 1922 British governments had advocated an all-round
cancellation of reparations and war debts. However, bankers in both London and
New York were worried lest German invocation of the relief provisions of the
Young Plan6 precipitate heavy withdrawals of foreign-owned deposits from
Germany and lead to the freezing of their reichsmark claims. This prospect was
especially frightening to the authorities in the United States where bank failures
had been rising sharply during the spring of 1931. 7 The solvency of some of
New York's great banks that were heavily committed in Germany might become
suspect and thereby add to the already severe strain under which the American
banking system was operating. It was to avert this new threat to the already
depressed United States economy and to take the initiative in the fight against
the depression that President Hoover on June 20 proposed the one-year postponement of payments of interest and principal on intergovernmental debts and
reparations—which became known as the Hoover moratorium.
The authorities in London and Berlin immediately accepted the President's
proposal, but in Paris the announcement from Washington was received with
consternation and anger. The French reaction was only partly attributable to
the fact that the United States Government had announced a proposal under
which France would lose almost $100 million during the coming year (i.e., the
excess of its reparation receipts over its war debt payments) and had done so
unilaterally and virtually without consulting the authorities in Paris. The reaction
also stemmed from bitterness that the United States Government, in its haste
to protect American interests, had given away a major bargaining counter, i.e.,

s

Hans Luther, Vor dem Abgrund 1930-1933 (Berlin: Propylaen Verlag, 1964), pages 159 and 163.

6

Supra, page 146.

7

Milton Friedman and Anna J. Schwartz, A Monetary History of the United States, 1867-1960
(Princeton: Princeton University Press, 1963), page 309; McGarrah letter to Fraser, June 9, 1931;
Harrison cable to Norman, No. 181, May 27, 1931; E. W. Bennett, Germany and the Diplomacy of
the Financial Crisis, 1931 (Cambridge: Harvard University Press, 1962), page 138; and Luther,
op. cit., page 164.

184




the granting of reparation relief which the French themselves might have employed in their efforts to force Germany to drop the proposed customs union
with Austria. More fundamentally still, the reaction arose from France's deepseated belief that, to the extent that Germany threw off the obligations of Versailles, French security would be increasingly threatened.8 France, as we shall
see, would fight intransigently to safeguard its security against a resurgent Germany; it would also go to great lengths to offer financial cooperation to maintain
a status quo that Germany now found intolerable.

CENTRAL BANK COOPERATION:
HALF MEASURES TO SUPPORT AUSTRIA AND GERMANY

In dealing with the 1931 crisis the major central bankers were hobbled not only
by fundamental political conflicts but also by a number of other handicaps. As
no comparable international financial difficulties had occurred before, the
authorities had no previous experience by which to guide themselves. They were
still only partly aware, moreover, of the magnitude of the potential trouble. It
was not until May 20 that Norman and Harrison had roughly reliable data on
the extent of London's and New York's claims on the Credit-Anstalt. Harrison
had fairly accurate data on the short-term claims of the major New York banks
against Germany, but he did not learn the magnitude of London's short-term
claims against Germany until the latter part of July.9 The first figures he saw
on London's short-term liabilities to foreigners were from the Macmillan Report
which was published on July 13. But more than mere information was lacking.
A conception of the kind of dramatic cooperative action that was required to
restore confidence and thus stop the mounting financial crisis had yet to be
developed. The Bank of England, from which leadership might ordinarily have
been expected, was in no position to organize an adequate cooperative effort.
As already noted, the French authorities, who had large resources at their
command, would make them available only on conditions that were politically
unacceptable to Germany and Austria. The Federal Reserve, too, had large
resources, but in May 1931 Harrison was still guided by the stabilization expe-

8

Bennett, op. cit., pages 169-72.

9

Cable received on July 21, 1931 from "responsible friends" in London by the Grace National Bank
and made available to Harrison.




185

riences of the 1920's. It was not until the end of July, when sterling was coming
under increasing pressure, that Harrison began to grope toward action on a
scale approaching the magnitude required to deal with the crisis.
These many difficulties were reflected in the handling of both the Austrian
and German crises. Although central bankers understood the close monetary
links between Austria and Germany, the difficulties in the two countries were
not treated as an organic whole but separately. However, a country-by-country
approach could have been successful only if the assistance to Austria had been
given in sufficient amount and with sufficient promptness to restore full confidence
in the stability and convertibility of the currency and had thus prevented the
crisis from infecting Germany. Neither condition was fulfilled. Instead, heavy
reliance was placed on arrangements under which Austria's foreign creditors
agreed not to withdraw their outstanding claims on the country. Unfortunately
the standstill arrangements—as they came to be called—had an effect contrary
to that intended by the central bankers. By freezing the creditors into their
Austrian assets, the standstill arrangements created doubts about the ability
both of the creditor institutions to meet their own deposit liabilities and also
of the creditor countries to convert their currencies into gold and foreign
exchange at the existing exchange rate. Consequently, Austria's difficulties were
transmitted to its weakest creditor, Germany, whose difficulties were similarly
treated with inadequate central bank credits and standstill arrangements, thereby
insuring that the trouble would spread to the next weakest creditor, Britain,
from which the infection spread, after the severing of sterling's link with gold,
to the United States.
All the weaknesses that were to lose the
battle against the 1931 crisis were evident in the international attempt to support
the Austrian schilling. By almost every standard, the amount of the central bank
credits was too small. Arranged under the leadership of Gates W. McGarrah,
President of the BIS, and granted by the BIS and the central banks of Britain,
France, Germany, the United States, and seven other countries, they totaled
only $14 million.10 This was of the same order of magnitude as the stabilization
SUPPORTING THE AUSTRIAN SCHILLING.

10
The Federal Reserve agreed to take up to $1.4 million of the credit but, because more central
banks agreed to participate than had been expected, the share that the Federal Reserve was actually
asked to contribute totaled only $1.1 million. Harrison letter to Norris, May 22, 1931; BIS cable to
Federal Reserve Bank of New York, No. 348, June 14, 1931; and Memorandum of Agreement
between Federal Reserve Bank of New York and BIS, May 30, 1931.

186




credits that had been granted to the various Central and Eastern European
countries during the 1920's but, considering that the foreign short-term claims
on the Credit-Anstalt alone were $100 million at the beginning of May and
that perhaps $20 million was withdrawn in the first two weeks of the crisis,
credits of this magnitude were clearly inadequate to restore confidence in the
circumstances of May 1931.11
Moreover, there was a three-week delay between the onset of the crisis on
May 11 and the making available of the credits to the Austrian National Bank
on May 31. It is true that almost as soon as the crisis broke McGarrah began
to sound out the major central banks about aid to Austria.12 By May 14, when
he approached Harrison, the Reichsbank, the National Bank of Belgium, and
the Bank of England had agreed to participate.13 On the following day, in
response to Harrison's request, the Federal Reserve Board approved Federal
Reserve participation up to a maximum of $3 million.14 That the credit was
delayed for another two and one-half weeks is probably attributable only in
small part to the negotiations to obtain the participation of seven other central
banks, including the Bank of France. The delay was attributable mainly to the
need to arrange the standstill among the Credit-Anstalt's principal creditors in
Berlin, London, New York, and Paris and to persuade the Austrian government
to guarantee the eventual repayment of the Credit-Anstalt's foreign obligations.15
From one point of view these were reasonable precautions. Having agreed to
extend assistance insufficient in amount to restore confidence in the schilling,
the central bankers were exposed to the danger that their credits to the Austrian
National Bank would in effect be used to finance the withdrawal of foreign
short-term loans to Austria. However, while the haggling over the terms of
the standstill and the Austrian government guarantee continued, the pressure
on the schilling, which had virtually ceased during the third week in May, was

11

The stabilization credits granted during the 1920's to Czechoslovakia, Hungary, Poland, and Rumania were in the $10-25 million range. Morgan & Cie. cable to J. P. Morgan & Co. for S. P. Gilbert,
May 26, 1931 and Morgan Grenfell & Co. to J. P. Morgan & Co., May 20, 1931.
12

Fraser cable to Harrison, No. 3, May 11, 1931.

13

McGarrah cable to Harrison, No. 6, May 14, 1931.

14

Meyer letter to Harrison, May 15, 1931.

15

BIS cable to Federal Reserve Bank of New York, No. 319, May 28, 1931; Austrian
Bank letter to Federal Reserve Bank of New York, May 28, 1931.




National

187

renewed on a substantial scale.16 In the last statement week of the month the
Austrian National Bank lost one-third again as much foreign exchange as it obtained from the central bank-BIS consortium. Some three fifths of the credits were
drawn as soon as they became available at the end of May, but the pressure on
the schilling continued albeit with somewhat diminishing intensity.17 The remainder was drawn on June 5, and on June 8 the Austrian National Bank
requested a second $14 million credit.18 Under BIS leadership this too was
arranged by June 14, but subject to the condition that the Austrian government
negotiate a $21 million, two- to three-year foreign loan, the proceeds of which
would be employed to strengthen the position of the Credit-Anstalt.19
Negotiations for this loan, of which a substantial part was expected to be
taken up in Paris, were initiated but promptly bogged down when it became
evident that the French made the abandonment of the customs union a condition
for cooperation.20 The situation was further complicated by stiff opposition
within the Austrian cabinet to the guarantee of the Credit-Anstalt's foreign
liabilities. The fall of the Austrian government was imminent. Meanwhile, rumors
of these difficulties had reached the exchange markets and the schilling again
came under heavy pressure. On June 16 Norman telephoned Harrison that the
position in Vienna was critical, that unless the Austrian government immediately
obtained the loan and the Austrian National Bank the second credit, the Austrian
commercial banks would close on June 18, and that the moratorium would
probably spread to adjoining countries including Germany. The situation, Norman said, was too pressing and serious to wait upon the ordinary placing of
the Austrian government bonds. Accordingly, the Bank of England had that
day, on its own account, extended the entire amount of the loan to the Austrian
government. Norman expressed the hope that this emergency credit, which was
for seven days and renewable on a week-to-week basis, would "avoid a crisis in

16

Morgan

&. Cie. letter to J. P. Morgan

17

BIS cable
Arrangements

& Co., May 26, 1931.

to Federal Reserve Bank of New York, No. 324, May 31, 1931. Draft
in connection with Austrian Situation, Basle, June 3, 1931.

18
BIS letter to Federal Reserve Bank of New York,
Norman on June 8, 1931 (Harrison
Papers).
19
McGarrah
letter to Fraser,
No. 348, June 14, 1931.
20

June 9, 1931; BIS cable to Federal

Gannon letter to Wiggin, June 13, 1931; Bennett,

188




June 5, 1931; Harrison

op. cit., pages

Reserve

149-50.

Report

on

conversation

with

Bank of New

York,

Central Europe . . . afford the Austrian government some time to turn around
and to negotiate for the sale of its bonds" and "save the ship before she sinks".21
The immediate objective of the central bankers was achieved. The Austrian
banks remained open, the parity of the schilling was maintained and, although
the Austrian government fell on June 16, the succeeding government respected
the guarantee extended to the Credit-Anstalt's foreign obligations. For a time
the standstill arrangements were fairly well observed. But these arrangements
led to other forms of direct control over Austria's financial markets. The agreement of foreign banks to maintain their credits was followed almost inevitably
by pressure for, and the eventual adoption of, exchange controls in order to
prevent capital flight by Austrian nationals, to collect Austrian foreign exchange
receipts, and to ration foreign payments in order to facilitate the reduction of the
country's external obligations.22 The first steps were thus taken to break up,
and fundamentally change, the system of unrestricted multilateral payments that
had been restored during the 1920's.
THE FAILURE OF EFFORTS TO SUPPORT THE REICHSMARK.

T h e Cooperative effort

to

support the reichsmark fell into three distinct phases. The first effort followed
immediately upon the announcement of the Hoover moratorium proposal on
June 20, was complicated to only a minor extent by political conflicts, and was
rewarded with a definite but brief success. The second, stimulated by the renewed
attack on the reichsmark that began at the end of June, involved an attempt
to organize central bank support on a hitherto unparalleled scale, was thwarted
by the prevailing political conflicts, and led the central bankers to pass the
problem of supporting the German currency on to their respective governments.
In the third phase, the problem was tackled at the highest governmental level.
However, since the political differences were found irreconcilable, the problem
was passed on to an international committee of experts who in due course returned
it unsolved to the governments.
First phase: the central bank credit to the Reichsbank. The central bankers'
fears that Austria's difficulties in May and June would infect Germany proved

21

Harrison conversation

with Norman on June 16, 1931 (Harrison

Papers).

22

Morgan Grenfell & Co. cable to J. P. Morgan & Co., May 28, 1931; H. S. Ellis, Exchange Control
in Central Europe (Cambridge: Harvard University Press, 1941), pages 33-39. Laws establishing
exchange control were enacted in October 1931. H. C. Hageman memorandum to Crane, March
9, 1933.




189

CHART 8

CENTRAL BANK DISCOUNT RATES
UNITED STATES, UNITED KINGDOM, GERMANY, AND FRANCE
MAY-DECEMBER 1931

Sources: Board of Governors of the Federal Reserve System, Banking and
Monetary Statistics ( Washington, D.C., November 1943 ); R.G. Hawtrey,
A Century of Bank Rate ( London, 1932 ); A. Parchman, Die Reichsbank
(Berlin, 1933).

only too well founded. On May 26, when it was still uncertain whether Austria
would be forced to close its banks and when the central bank credit to the Austrian National Bank was still pending, rumors spread through the exchange
markets that a German bank moratorium was imminent. Berlin stock market
prices declined sharply, and in the foreign exchange market the reichsmark
dropped below the gold export point.23 Subsequently it became known that, despite the government's earlier retrenchment efforts, the national budget was still
showing a substantial deficit. New taxes were decreed and new cuts were announced in official salaries and in payments to the four million unemployed.24 On
June 13 the Reichsbank increased its discount rate to 7 per cent from 5 per cent

23
Norman cable to Harrison,
Bennett, o p . cit., page 117.
24

No. 152, May 28, 1931; The N e w York Times, May 27, 1931; and

Ellis, o p . cit., page 165; League of Nations,

190




Monthly Bulletin of Statistics, December

1931.

(see Chart 8). However, these official measures led to intensification of domestic
political agitation and to widespread expectations both that the German government would soon declare that it could no longer meet its reparation obligations
and that this declaration would precipitate the long feared financial crisis.
In the resulting uncertainty, German capital moved abroad and foreigners
hastened to reduce their commitments in Germany. The Reichsbank lost some
$250 million in gold and foreign exchange in the first three weeks of June. On
June 19 when it was clear that the Reichsbank's eligible international assets
would be insufficient to meet its note cover requirements for the June 23
statement,25 Luther approached Norman for a credit from the Bank of England,
the Federal Reserve, and/or the BIS. Although the announcement of the Hoover
moratorium on the following day brought respite from the pressure on the
reichsmark, Luther indicated on June 22 that he still needed $80-100 million
in order to meet his legal minimum during the period June 24-July 16 when
the bank was faced with substantial temporary demands for currency and foreign
exchange.26 Norman, who had kept Harrison informed by daily telephone calls
and cables, indicated that he was reluctant to lend additional funds since the
Bank of England was already heavily committed in Austria and Hungary27 but
that, since the situation was "critical and dangerous", he would agree to a $100
million credit to the Reichsbank against commercial bills provided the Federal
Reserve participated on a fifty-fifty basis.28
Harrison was in full agreement with the need for such a credit but insisted
that the participation be broadened, particularly to include the Bank of France.
Long-standing Federal Reserve policy would in any case have required that an
invitation be extended to the Bank of France, but there was another even stronger
reason for Harrison's position. He had learned from Governor Meyer of the
Board that the Washington administration, including apparently the President,
was dubious that any credit should be extended to the Reichsbank on the grounds

25
The statement
each week.

was published

four times each month and not necessarily

26
Harrison conversations with Norman
23, 1931 (Harrison
Papers).

on the same day of

on June 19, 23, 24, 1931 and with Lacour-Gayet

on June

27

The Bank of England had participated with the Federal Reserve, seven other central banks, and
the BIS in a $10 million, three-month credit to the Hungarian National Bank. The Bank of
England's share, like that of the Federal Reserve, was $2 million. The agreement on the credit was
dated June 18,1931.
28

Harrison conversation




with Norman

on June 23, 1931 (Harrison

Papers).

191

that such assistance might remove the pressure on the French government to
accept the proposed moratorium on intergovernmental debts and reparations.
Harrison himself was skeptical that the extention[extension]of the proposed credit would
have such an effect but indicated that he would make it a condition for Federal
Reserve participation that the Bank of France participate in substantial amount.
However, even after he had been reassured on this point, Meyer remained hesitant. According to Harrison's notes of a telephone conversation on June 23,
Meyer asked29
whether I thought that the credit would be repaid. I said my own judgment
was, and Norman was firmly of the opinion, that the credit would be repaid on
July 16, unless, of course, negotiations relative to the debts completely broke down,
in which event I said that he must realize that we ran the risk of a frozen credit in
the Reichsbank. I stated, however, that even if the credit should become frozen,
I did not feel that there was much risk of ultimate loss. In any event, I felt that the
risks of not participating were greater than the risks of doing so . . . .

Meyer then asked whether it would not be better for the Board to authorize
participation up to only $35 million rather than the $50 million for which the
New York Reserve Bank had asked and Harrison explained that, although he
did not expect to extend more than $35-40 million, he wanted some leeway.
Governor Meyer then asked what was the hurry and why couldn't he take it up
with the Board tomorrow or give them more chance to think it over. I told him
that I only knew what Norman told me, that it was essential that they have the
credit as of tomorrow, and if that was so we had to act tonight, and I asked him
if he would not try to please arrange for approval or disapproval tonight so that
I might cable Norman.

Meyer then agreed to consult the members of the Board. At 7 p.m. on June 23
Meyer advised Harrison that the Board had approved a credit of up to $50
million. At the same time that Harrison was making these arrangements, he was
also in touch with Lacour-Gayet of the Bank of France and with McGarrah
of the BIS. The upshot was that by June 24 the $100 million credit to the
Reichsbank had been put together with the BIS and the central banks of Britain,
France, and the United States participating to the extent of $25 million each.30

29

Harrison conversation with Meyer on June 23, 1931 (Harrison Papers).

30

Norman cable to Harrison, No. 191, June 24, 1931.

192




Even so the credit was not available in time to be included in the Reichsbank
statement for June 23, but the $5 million deficiency in the reserve requirement
was made good by an overnight deposit in that amount made with the Reichsbank by the Bank of England.31
Second phase: the attempt to raise a second central bank credit for the Reichsbank. The announcement of the central bank credit to the Reichsbank on June
25 reinforced the effect of the Hoover moratorium proposal on the exchange
markets. The reichsmark strengthened, and for a day or two the Reichsbank
apparently lost no reserves in exchange operations.32
However, the calm was short-lived. As June drew to a close, uncertainty increased whether the French would accept the President's proposal. Moreover, on
July 2-3 there was news of the failure of a major German textile concern to
which the Darmstadter und Nationalbank (Danat Bank) had large loans outstanding.33 A run on the Danat and other banks began, and with it a renewal
of the pressure on the reichsmark in the exchange markets. The story is told
in the use of the central bank credit. The first $4 million drawing was not made
until June 27; by June 29, no more than $12 million had been drawn. In the
next three days $64 million was drawn. Even so, the Reichsbank's eligible international assets were below the 40 per cent minimum ratio on July 3.34 On
July 4, the credit was exhausted and the Reichsbank was obliged to draw the
full amount of a $50 million line of credit that had been arranged in the twenties
by its affiliate, the Gold Discount Bank, with a group of New York banks under
the leadership of the International Acceptance Corporation.35 These drawings
enabled the Reichsbank to show a $69 million reserve gain in the two weeks
ended July 7 but, despite the announcement on July 6 of an agreement between
the United States and French governments on the Hoover proposals, the Reichsbank lost some $71 million in gold and foreign exchange during the following
week, with the consequence that its July 15 statement showed reserves substantially below the legal minimum.

31

Harrison conversation with Norman on June 23, 1931 (Harrison Papers).

32

Luther, op. cit., pages 173 and 178.

33

Bennett,

o p . cit., page 220.

34

See various cables between the Federal Reserve Bank of New York and the BIS, the
and the Federal Reserve Board between June 26 and July 4, 1931.
35

Reichsbank

cable to Harrison,




Reichsbank,

No. 195, July 4, 1931.

193

The imposition of direct controls over the exchanges, on which major reliance
was eventually placed, played a secondary role in the initial efforts to deal with
this renewed outbreak of the crisis. True, Harrison was in touch with the major
New York commercial bankers on July 3 or earlier and was able to tell Luther
on July 7 that the New York banks were generally maintaining the amount of
their credits to Germany at the level outstanding on June 20.36 It seems probable
that parallel standstill efforts were being made by Norman in London, but on
this the evidence is not entirely clear.
At this stage, however, the main effort lay in hectic negotiations to raise a
new central bank credit for the Reichsbank. On July 8 Norman telephoned Harrison that he had heard from Luther that37
the position in Berlin is now desperate; that they lost another $10,000,000 in
exchange today and that Luther is going to London tomorrow by aeroplane just
for one hour; that they would travel to Dover together and that Luther would then
go on to Paris to see Governor Moret, and he, Norman, to Basle.
Norman said that judging by his conversation with Luther this afternoon, Luther
will . . . ask that the central bank credit shall be renewed, and will further ask for
an additional credit in "an unlimited amount" otherwise, he will say, Germany will
collapse . . . . Norman then mentioned that while Luther had referred to an
unlimited credit it was, of course, out of the question and that he gathered from
the conversation with Luther that they would need accommodation anywhere from
$500,000,000 to $1,000,000,000 to avoid a complete breakdown. Governor Norman
mentioned that he would be willing to join in such a credit, which Luther said
must be for at least six months, provided we and the Bank of France participated,
and that he felt that the character of the credit or the terms must be made to suit
Paris and New York, inasmuch as the powers of the Bank of England are so much
broader than ours, and that he was therefore able legally to participate in any
arrangement that would be adaptable to our own needs.

In the ensuing conversation Harrison stressed that, if any such large credit were
contemplated, he would consider it only on condition that (1) the private bankers
in the major international financial centers agreed to maintain their existing
credits to Germany and to take participations in the new central bank credit and
(2) that the new credit be guaranteed by the Reichsbank, the Gold Discount
Bank, and the German government. With all these points Norman agreed. Finally
Harrison told Norman that

36

Harrison conversation with American bankers July 3, 1931 (Harrison Papers); Harrison
Luther, No. 107, July 7, 1931.
Harrison conversation with Norman on Germany, July 8, 1931 (Harrison Papers).

37

194




cable to

I was very skeptical about the idea of a credit in any event; that I felt the chief
difficulty was a flight from the reichsmark by German nationals and that the
Reichsbank should resort to much more drastic credit control than apparently was
the case. Governor Norman said he thought Luther was now rationing credit very
strictly, but he was not sure how effective or how broad it was. He thinks the
difficulty is that they need Dr. Schacht there; that Luther does not seem to have the
force necessary to do the trick. In any event, I insisted that these credit restrictions
should be adopted, in my judgment, before we could fairly be asked for a sizable
new credit. Norman generally agreed with this position.
Governor Norman mentioned that the next published statement of the Reichsbank
will be as of Tuesday, July 14, and that if anything was to be done it should be
done before that time.

Later the same day Harrison dispatched cables to Norman, Luther, and Moret
in which he reiterated his view that much of the pressure on the reichsmark
was attributable to German capital flight which he felt could be curbed by a
firmer Reichsbank credit policy.
In the negotiations of the next five days both Harrison and Norman changed
their positions regarding the second Reichsbank credit. As the prospects for a
collapse of the Germany currency increased, Harrison worked desperately for
the adoption of a program that would enable him to recommend that a second
credit be granted. On the other hand, Norman, after initially favoring the credit,
turned flatly against it and in this was supported by Moret.
Moret's position reflected that of his government. Having surrendered one
major bargaining counter when they agreed to the Hoover moratorium, the
French authorities were determined to exploit the reichsmark crisis to force
Bruning to agree to France's terms. Repeatedly in the conversations of July 10-21,
French officials urged the German government to agree (1) to drop the customs
union proposal, (2) to resume the full payment of reparations at the end of
the moratorium year, (3) to refrain in the meantime from using for military
purposes any of the funds that in the absence of the moratorium would have
been paid in reparations, and (4) to suspend agitation for the revision of Germany's eastern frontier. If Germany would do so, the international political
atmosphere would clear.38 Such an improvement, the French pointed out, not
only would reduce the pressure on the reichsmark but also would put the French
government in a position to come to the aid of Germany with substantial financial

38

Luther, op. cit., page 187; Bennett, op. cit., pages 230, 231, and 264-65.




195

assistance. However, until Germany accepted France's terms, the Paris authorities were prepared to wait in the belief that Bruning would find that he had
no alternative and in the knowledge that France's position was strong—the
French banks had withdrawn most of their credits to Germany during the
previous fall and winter. What the French failed to foresee was that Germany
would escape the acceptance of France's terms and would at the same time avoid
a complete currency debacle through the adoption of exchange controls.
While the considerations behind the French position were clear from the
beginning, those behind the shift in Norman's position became apparent only
gradually. When Harrison telephoned on Thursday, July 9, the day Luther was
to arrive in London, Norman reported that the Bank of England directors were
doubtful about the granting of a second credit to the Reichsbank because they,
like Harrison, felt that the pressure on the reichsmark was due more to the
outflow of German capital than to the withdrawal of foreign funds; because
they were uncertain whether the Reichsbank could or would adopt a more
stringent monetary policy; the proposed credits were "altogether too big and political rather than financial"; the French would require political conditions to
which the British could not agree; and because the Bank of England did not
wish to participate in a credit that would facilitate the continued payment of
reparations to France.39 Norman's final objection was especially significant. In
view of the United States-French agreement three days earlier to suspend intergovernmental payments for a year, it could only have meant that he was
opposed to a credit granted on the understanding that reparations would be resumed at the end of the moratorium year.
In the course of the following week indeed it became unmistakably clear that,
whereas French policy aimed to perpetuate the status quo, Norman backed by
Snowden was working for a new order. Both the Governor and the Chancellor
felt intensely that the whole Versailles system of reparations and war debts was
one of the fundamental causes of the crisis and that, unless the governments
could agree to eliminate the system forthwith, there was little that the central
banks could do to alleviate the pressure on the reichsmark.40

39

Harrison conversation

40

with Norman

on Germany,

July 9, 1931 (Harrison

Papers).

Harrison conversations with McGarrah and Fraser on July 14 and with Norman on July 16, 1931
(Harrison Papers); Cochran cable to Castle, Papers Relating to the Foreign Relations of the United
States, 1931, Vol. 1, page 255.'

196




However, Norman and Snowden were deceived in hoping that the deepening
of the crisis would bring France and the United States to accept the need for
a fresh start.41 Such a reversal in the French position was not in the cards. It is
true that the involvement of the New York banks in Germany gave the United
States authorities an interest in helping to support the reichsmark, but they would
not have done so without the participation of both Britain and France whose
positions, as we have seen, were irreconcilable. Above all, any suggestion of
going beyond the one-year moratorium was flatly rejected by the Hoover administration. At this stage indeed the position of the United States was much
closer to that of France than to that of Britain.
Luther's flying trip to London and Paris was thus in vain. News of its failure
intensified the pressure on the reichsmark.42 On July 11-12 Harrison received
desperate messages from the German authorities appealing for large credits and
for a Federal Reserve statement of confidence in the reichsmark.43 To all of
these, Harrison's response was that he could do nothing until the German
authorities had presented a program to support their currency that was satisfactory to Norman and Moret who on July 12-13 were to meet in Basle to discuss
the German crisis.44
In view of the diametrically opposed positions of the major countries, the
Basle meeting could hardly have avoided a deadlock. Moreover, Luther was
exhausted from his trip to London and Paris and from hectic discussions in
Berlin that began on Sunday and continued until 3 a.m., Monday, July 13 when
he flew to Basle.45 Once there Luther indicated that the Reichsbank was doing
all it could to support the currency and reiterated his request for an additional
loan of unspecified amount. However, Leon Fraser, the BIS's vice president,
felt Luther had presented no program worthy of the name; he had indeed been
"quite helpless" during the entire meeting. For their parts, Norman and Moret
were prepared only to renew the first central bank credit to the Reichsbank. They
indicated that any new loan to Germany would have to be supplied primarily

41

Luther, o p . cit., page 186.

42

The Economist, July 18, 1931, page 104.

43
Kiep to Harrison with copy of cable from Berlin, July 11, 1931; Harrison
at 10 a.m. and 12 noon on July 12, 1931.
44

with

Kiep

Harrison cable to Luther, No. 116, July 12, 1931.

45

conversation

Luther, o p . cit., page 193.




197

by governments. If such a loan could be arranged, they would be prepared to
support it with additional central bank credits. Relaying this information to
Harrison by telephone, McGarrah asked whether the United States Government
would participate in such a loan and if so for how much.46
Harrison's personal reaction was that the proposal for a government loan
was "futile" and that the United States Government felt it had already "done a
great deal" under the Hoover moratorium. He emphasized that47
time is the essence and it would undoubtedly take all the governments a long
while; that our own government, certainly, could not make such a loan without the
approval of Congress and there would be no session of Congress until December,
and that any plan proposing government loans would seem to me to be disingenuous,
and would be considered as an effort to pass the buck back to the United States,
realizing it is likely that we would not be able to make the loan anyway. McGarrah
agreed with this and said he was quite conscious of the fact that it might be an
effort to "put the baby on our doorstep".

However, Harrison also felt that, unless prompt and substantial support were
received from abroad, the German currency would collapse, and this view was
shared by McGarrah. Accordingly, Harrison suggested to McGarrah that it
might be possible to avoid the delays and political dangers inherent in a
government loan by arranging a private loan to the Gold Discount Bank from
the commericial[com ercial]banks in the major financial centers and to supplement this
private loan with an additional central bank credit to the Reichsbank. McGarrah
indicated that he felt this to be a constructive suggestion and that he would
discuss it immediately with Luther and perhaps also with the others at the
meeting.48
By 5:20 p.m. New York time (11:20 p.m. Basle time) Harrison had heard
nothing further from Basle. He then telephoned Norman. Called from the meeting, which was still in session, Norman indicated that the deadlock was unbroken.
In Harrison's record of the conversation, Norman then said49

46
Harrison conversations
with McGarrah
at 10:35 a . m . and 3:45 p . m . on July 13, and
McGarrah and Fraser on July 14, 1931 (Harrison
Papers).
47

Harrison

48

conversation

Harrison
Papers).

conversations

49

conversation

Harrison

198




with McGarrah

at 3:45 p . m . on July 13, 1931 (Harrison

with McGarrah
with Norman

Papers).

at 10:35 a.m. and 3:45 p . m . on July 13, 1931

at 5:20 p . m . on July 13, 1931 (Harrison

with

Papers).

(Harrison

that he was leaving in an hour to go back to London and that the board was going
to adjourn. I said, if that is true, what is going to happen? He said nothing is going
to happen—absolutely nothing. I said I thought it was a great pity that they should
adjourn before something was done. He said: "Well, what have you to suggest?"

Harrison then repeated the proposal that he had made earlier in the day to
McGarrah to which Norman's response was merely to reiterate that the German
problem was
too big for the central banks, and the governments must interest themselves in it,
and he did not see anything that the central banks could do at this time. He said he
had to go back to the meeting as they were still in session and that he was leaving
in an hour for London. He seemed tired, disgruntled and discouraged.

Third phase: the London Conference. There was ample ground for discouragement. On July 13—the very day of the Basle meeting—the failure of the Danat
Bank was announced. Thereupon the German government decreed a two-day
holiday for the Berlin bourse and all commercial and savings banks. When the
financial markets and banks reopened on July 16, the Reichsbank discount rate
was increased to 10 per cent, and the first of an ever-growing number of exchange
control regulations had already been promulgated.50 Until the last moment, moreover, it was uncertain whether a conference of the major governments could be
arranged because Pierre Laval, the French Prime Minister, and President Hoover
would not agree to send representatives until they had received assurances from
Prime Minister MacDonald that the conference would deal only with the immediate emergency and would avoid (contrary to the hopes of Norman and
Snowden) any discussion of "fundamental questions", i.e., reparations and
war debts.51
As discussions about the terms of reference for the conference continued,
the French government made a major bid for the preservation of the status
quo. It proposed on July 16 that a $500 million ten-year government-guaranteed
loan be floated on Germany's behalf on the Paris, London, and New York markets. It was also made clear that the Bank of France would participate in additional central bank credits to the Reichsbank pending the flotation of the

50

Ellis, op. cit., page 166.

51

Stimson conversation
with Hoover, July 17, Papers Relating to the Foreign Relations of the
United States, 1931, Vol. 1, page 273; Harrison conversation with Mills on July 16, 1931 (Harrison
Papers); and Bennett, op. cit., page 256.




199

bonds.52 The reception for the loan proposal was completely negative. Hoover's
reaction was that he would not call a special session of Congress to obtain
authority for such a loan and that, even if he did, Congress—which had come
under the control of a coalition unsympathetic to his administration as the result
of the November 1930 election—would not grant the authority.53 Norman
rejected the proposal on the ground that, if the British government guaranteed a
loan for Germany, it could hardly avoid extending the same favor to India,
Australia, and other British dominions.54 Beyond this, sterling was already feeling
the devastating repercussions of the German crisis and was in no position to
stand the strain of additional assistance to support the reichsmark. Even if
sterling had been strong, however, it is virtually certain that the British authorities
would not have participated in a loan whose aim was to perpetuate the status
quo. In any event, Germany would not make the necessary political concessions
as became evident in the discussions that took place in Paris between Bruning
and Laval on July 18-19.55
In view of this response, no more was heard of the French loan proposal when
the representatives of the major powers met in London on July 20-23. Indeed,
this conference was unable to move significantly beyond the positions that had
previously been reached by the central bankers. The conference recommended
that the central bank credits to the Reichsbank, which had already been renewed
until August 6, be extended for three months beyond that date.56 It put the sanction of the governments behind the efforts that Harrison and Norman were already
making to prevent the withdrawal of outstanding private credits to Germany,57
and it recommended that the BIS establish a committee of experts "to enquire
into Germany's immediate further credit needs and to study the possibility of

52
Harrison
256-57.

conversation

with Mills

on July 16, 1931 (Harrison

Papers);

Bennett,

o p . cit., pages

53

Stimson memorandum
of July 15, Papers Relating to the Foreign Relations of the United States,
1931, Vol. 1, page 261; Bennett, op. cit., pages 259-60.

54

Harrison

conversation

55

Bennett,

op. cit., pages

with Norman

on July 16, 1931 (Harrison

Papers).

263-74.

56

See particularly
the stenographic
notes of the London Conference, July 20-23, Documents on
British Foreign Policy, 1919-1939, 2nd Series, Vol. 11, pages 435-84; BIS cable to Federal
Reserve
Bank of New York, No. 429, July 15, 1931.
57

Harrison conversation
with Norman at 9:30 a.m. on July 15, 1931 (Harrison
cables to Norman, No. 231 and No. 234, July 15 and 16, 1931.

200




Papers);

Harrison

converting a portion of the short-term credits into long term".58 A month later
this committee, of which Albert Wiggin of the Chase National Bank was chairman, made a report that in effect handed the problem of Germany's credit
needs back to the governments. But by this time Germany was learning to live
without additional foreign loans behind a mounting barrier of exchange controls.
The center of the crisis had moved to London.

THE ATTACK ON STERLING

Britain was in a weak position to resist the attack on sterling largely because the
Labor government was unable to resolve the conflict between the need for
domestic economic recovery and the requirements of external stability. The
Labor Party itself was split. Snowden was as fully committed to the defense of
sterling's gold parity and to the balancing of the government's budget as his
Conservative predecessors but, in seeking these objectives, he was obliged to
deal with strong opposition from Labor Party colleagues led by Arthur Henderson, the Foreign Minister. While this opposing group did not advocate the
abandonment of gold, it refused on domestic welfare grounds to accept the
policies that Snowden insisted were necessary for the protection of sterling. Many
within the Labor Party were suspicious that the well-being of the workers was
being sacrificed to the international interests of the City of London, and all were
highly sensitive to Britain's growing industrial distress. In the two years that
Labor had been in power, industrial output had declined about onefifthand
unemployment had more than doubled to 22 per cent of the labor force in June.
Snowden had nevertheless rejected proposals for greatly expanded public works.
Even so, government expenditures rose. Moreover, with the continued contraction
of economic activity in 1931, revenues declined, the consequence being that
Snowden's hopes for budgetary balance were disappointed. The upshot was
that Snowden satisfied neither those who favored spending to relieve unemployment nor those who felt that budgetary balance was essential to avoid financial
disaster.
At the same time Britain's external position failed to show any fundamental
improvement. In the two years ended the second quarter of 1931, imports had

58
Report of the Committee appointed on the Recommendation of the London Conference, 1931,
published in Documents on British Foreign Policy, 1919-1939, 2nd Series, Vol II, page 485.




201

dropped sharply but the decline in exports was sharper still. Long-term capital
outflow was reduced but so were earnings from shipping, tourism, insurance,
and investments abroad. Although London's short-term liabilities to foreigners
had been reduced, they were still in the neighborhood of £600-700 million in
mid-1931, four to five times as large as the Bank of England's gold reserve.59
THE FIRST ATTACK. The first phase of the attack on sterling began on Monday,
July 13, when the central bank governors were deadlocked at Basle and when
the failure of the Danat Bank was announced. On that day sterling, which had
been at or close to parity with both the dollar and the French franc for the
past two months, dropped sharply on the exchange markets. On July 15 the
rate was $4.84 1/4, well below the gold export point to New York. On the
following day, Norman telephoned Harrison that "the situation now looks very
bad . . . the whole thing is 'boiling up' ".60 In the two weeks ended July 29
the Bank of England lost $200 million in gold and dollars, one quarter of its
international reserves.
The major immediate cause of the attack was the closing of the Berlin banks,
accompanied by the virtual freezing of foreign assets in Germany where British
interests had short-term claims of around $300 million.61 Although the magnitude
of these claims became generally known only later, it was understood in the
markets in mid-July that London's commitment in Germany was large and that
the merchant banks, which typically financed a substantial proportion of their
foreign investments with funds borrowed from abroad, were heavily involved.62
To make matters worse, estimates of London's large short-term liabilities to
foreigners also became available to the public for the first time on July 13 when
the Macmillan Report was published. The market was thus flooded with doubts
not only about the solvency of individual British banks but also about whether the
authorities would be able to maintain the convertibility of sterling at the existing
parity. Beyond this there were fears on the Continent that the runs suffered by

59

David Williams, "London and the 1931 Financial Crisis", The Economic History Review, Vol. 15
(April 1963), page 528.
60
Harrison conversation with Norman at 3:50 p.m. on July 16, 1931 (Harrison
Papers).
61
Report of the Committee appointed on the Recommendation
of the London Conference, Annex V,
August 18, 1931; Parliamentary Debates (Great Britain, House of Commons), September 21, 1931,
column 1294.
62
David Williams, "The 1931 Financial Crisis", The Yorkshire Bulletin, Vol. 15 (November
1963),
page 107.

202




the Austrian and German banks would spread to yet other countries, and these
fears led commercial banks in Belgium, France, Holland, Sweden, and Switzerland to attempt to bolster their liquidity by drawing down their London balances.63 Doubts and fears quickly snowballed. Rumors spread that Norman was
demanding the imposition of a moratorium on payments by British banks, and
that the government would levy a tax on capital to deal with the crisis. These
rumors not only aggravated the withdrawal of foreign balances but stimulated
some outflow of British funds as well.64
The distraction of the German crisis and the split in the
Labor Party resulted in a hesitant initial response by the British authorities to
the attack on sterling. On July 16, when the Bank of England had already lost
£ 7 million in gold but when it was still possible to be hopeful about the forthcoming London Conference, Norman cabled Harrison:65
BRITAIN'S RESPONSE.

We were perplexed this morning about change in bank rate because exchanges are
still so disorganized that 21/2per cent is hardly justified. But to cope with such
disorganization would require 7 per cent or 8 per cent rather than 31/2per cent
and as no more gold is being withdrawn we hope to ride at least another week.

But, although the London Conference failed and the pressure on sterling
intensified, the bank increased its discount rate only 1 point on July 23 and
a week later by another 1 point to41/2per cent, at which level it remained
unchanged until after the suspension of the gold standard. Meanwhile, the government did nothing more than promise on July 30 that it would "take every possible
step to ensure that the proud and sound position of British credit shall be in no
way impaired". Thereupon Parliament adjourned in the expectation that it would
not reassemble until October when the government planned to present its
budgetary retrenchment program.66

63

Willard Hurst, "Holland, Switzerland,
and Belgium and the English Gold Crisis of 1931", T h e
Journal of Political Economy, Vol. 40 (October 1932); Snowden statement on July 21, 1931 to
London Conference and Tyrrell cable to Vansittart, No. 174, July 16, 1931, published in Documents
on British Foreign Policy, 1919-1939, 2nd Series, Vol. 11, pages 451 and 203.
64
Hugh Dalton, Call Back Yesterday: Memoirs 1887-1931 (London: Frederick Muller, Ltd., 1953),
Vol. 1, page 255; G. L . Harrison, "The Credit Crisis of 1931", dated November
20, 1931, memorandum prepared for the Governors Conference in December 1931.
65
Norman cable to Harrison, No. 223, July 16, 1931.
66
Parliamentary Debates (Great Britain, House of Commons),
Vol. 255, Statement to the House of
commons by the Chancellor of the Exchequer, July 30, 1931, columns
2512-15.




203

At this point, the cause of inaction was partly pure physical exhaustion. Norman
collapsed on July 29 and was almost entirely out of action until after the crisis
was over. Partly it was because Snowden was waiting for the report of a committee, headed by Sir George May, which had been appointed the previous
February to review the government's finances. Snowden hoped apparently
that this report, which predicted an enlargement in the budgetary deficit that was
regarded as huge, would convince his reluctant colleagues to accept drastic
financial retrenchment. At the same time, the authorities hesitated to borrow
abroad, partly because they felt that such borrowings would remove the pressure
on the government to eliminate the budget deficit, and partly because they were
loath to approach France, which Harrison insisted should participate in any
credits to support sterling.67
CENTRAL BANK CREDITS TO THE BANK OF ENGLAND. It is doubtful whether, in the

circumstances, sterling could have been saved by even the most drastic combination of measures that the authorities of the day could have adopted, but the
program that actually was adopted was definitely a second best, even by the
standards of mid-1931. While the total amount of the credits obtained was considered very substantial, they were not granted in a single impressive package
but in two separate arrangements. The first set of credits was unaccompanied by
any additional measures to check the pressure on sterling, and the second set,
which was accompanied by the formation of a new government pledged to the
elimination of the budgetary deficit, was only announced after it was known in
the markets that the first credits had been virtually exhausted.
The handling of the crisis was a source of frustration and disappointment to
the central bankers. Harrison and apparently also Moret had hoped that the
central bank credits could be granted simultaneously, or at least in conjunction,
with large market loans to the British government.68 But the latter could hardly
be floated in New York or Paris until the British government had dealt with the
budgetary deficit. To Norman this was painfully clear. When toward the end of

67

McGarrah cable to Harrison, No. 27, July 28, and Norman cable to Harrison, No. 247, July 28,
1931; Harrison conversation
with Norman on July 29 and with Lacour-Gayet,
August 7, 1931
(Harrison Papers); Sir Henry Clay, Lord N o r m a n (London: Macmillan & Co., Ltd., 1957), page 385;
and Bennett, o p . cit., pages
281-82.
68

Harrison

conversations

204




with Lacour-Gayet

on August

6 and 7, 1931 (Harrison

Papers).

July Harrison telephoned to ask about the possibility of such a loan, Norman
replied that69
there was absolutely nothing to it; in the first place, Parliament was adjourning
and there was no time for action; in the second place, the home political situation
was not satisfactory and no such loan was possible . . . . you cannot make a loan
unless your house is in order.

In the meantime, however, the pressure on sterling was increasing and the
need for action was becoming urgent. Shortly after the failure of the London
Conference, the Bank of England approached both the Federal Reserve and the
Bank of France to obtain market support for sterling. On July 23 the directors
of the New York Reserve Bank authorized purchases of sterling of up to
£ 5 million on terms similar to those adopted under earlier arrangements.70
Apparently because the main pressure was coming from Continentals who were
offering sterling in their own markets, the opportunity to operate under this
arrangement was not great; by August 8 when the last purchase was made, only
£2.1 million ($10.2 million) had been acquired by the New York Reserve
Bank.71 In contrast, Bank of France sterling purchases, which were undertaken
against dollars pledged by the Bank of England, were far larger, apparently
totaling some $60 million equivalent in the period up to August 7.72
The Bank of England also overcame its reluctance to approach foreign markets
for loans. Sir Robert Kindersley was dispatched on July 25 to Paris where he
discussed with Moret the possibility of a £20-30 million rediscount credit for
six to nine months against sterling bills.73 As Norman described it, the credit might
have been either between the two central banks directly or between groups of
London and Paris commercial banks or a combination of both. On July 28,
Norman indicated that he had expected to ask the Federal Reserve Bank of New
York to make a similar arrangement in favor of the Bank of England but for a

69

Harrison

conversation

with Norman

on July 29, 1931 (Harrison

Papers).

70

Harrison cable to Norman, No. 243, July 23, 1931; McClelland letter to Crane, July 24, 1931; and
Harrison memorandum
for files, July 24, 1931 (Harrison
Papers).
71
Compilation
(apparently by Sproul) of Bank of England dollar sales and Federal
of New York sterling purchases, July 23-September
18, 1931.
72

Bank

Harrison conversation with Lacour-Gayet on August 7, 1931 (Harrison Papers).

73

Reserve

Norman cables to Harrison, No. 243 and No. 247, July 27 and 28, 1931.




205

larger amount and for a longer period.74 However, the inability of the British
government to agree on a budgetary retrenchment program, the complications
of raising a loan from the Paris commercial banks, and above all the need for
immediate action swept these proposals aside. On July 30 Harrison was urging
that the Federal Reserve and the Bank of France act alone. He telephoned
Lacour-Gayet:75
We are the two banks with the greatest gold supplies in the world, and a credit
granted by us, in which we each take equal shares without any other participants,
would have an electrifying effect, in my judgment, and would do more than anything else to convince the world of our constructive cooperation in the interest of
the gold standard . . . .

The Federal Reserve credit to the Bank of England was approved on July 30
by both the New York Reserve Bank's directors and by the Federal Reserve
Board. Although Sir Ernest Harvey, the Deputy Governor who had taken over
the leadership of the Bank of England after Norman's collapse, had hoped for a
substantially larger amount and was supported in this by Harrison and by at
least one of the directors of the New York Reserve Bank,76 the amount granted
was $125 million. Harrison made it clear to the Bank of England that this was
the maximum that could be obtained from the Federal Reserve. If more were
required, the British authorities should approach their New York commercial
bankers who would, Harrison indicated, promptly provide additional funds. At
this stage, however, Harvey indicated that the Federal Reserve credit, together
with an equivalent one from the Bank of France, would be enough.77 The
assistance was in the form of lines of credit on which the Bank of England could
draw dollars and francs as required. As each drawing was made, the Bank of
England was to deposit in the account of the central bank concerned prime
sterling commercial bills of not more than ninety days' maturity, having at least
two satisfactory bank signatures and the guarantee of the Bank of England. The
credits were for three months, with the understanding that they could be renewed
for a further period of three months, and carried interest on the amounts drawn

74

Norman cable to Harrison, No. 247, July 28, 1931.
Harrison conversation with Lacour-Gayet
on July 30, 1931 (Harrison
Papers).
76
Memorandum
on meeting of board of directors of the Federal Reserve Bank of New York,
30, 1931 (Harrison Papers); Hamlin Diary, July 30, 1931, page 115.
77
Harrison conversation
with Harvey on July 30, 1931 (Harrison
Papers).
75

206




July

at the rate of 33/8per cent per annum. Provisions were made safeguarding the
lenders against exchange risks and against the erection of any legal obstacles to
repayment in gold at maturity.78
The $250 million credits were announced on Saturday, August 1, in singularly
unpropitious circumstances. On the previous day the May Committee published
its report indicating that under existing policies the British budgetary deficit in
the 1932-33 fiscal year would be £120 million, the largest since 1920, and that
the prospective deficit in the following fiscal year would be substantially larger.
At the same time the Bank of England fiduciary issue was increased by £15
million which, in conjunction with the gloomy May report, was interpreted in
the markets as a sign of inflation while the credits themselves were interpreted
as a sign of weakness.79
To make matters worse, when the London market opened on Tuesday, August
4, after the long Bank Holiday weekend, the Bank of England failed to draw
on the credits or even to support sterling firmly with its own resources. This
lapse was apparently due to the feeling at the bank that the strain on sterling
should be primarily reflected in gold losses that would jolt the government into
action and that would perhaps provide a basis for a further increase in the
discount rate.80 It was also due to the insistence of the Bank of France that the
Paris commercial banks participated in each drawing on the French tranche of the
credit.81 Since such participation would reveal to the Paris market the extent to
which the credit was being utilized and since the agreements with the Bank of
France and the Federal Reserve provided that the two lines of credit be used
pari passu, the Bank of England felt unable to draw on either of its new facilities.
In the absence of firm official support, the sterling exchange rate declined
slightly on August 4 and on the following day broke badly in London, New York,
and Paris82 (see Chart 9). Rumors flooded the markets that the Bank of France
had withdrawn its support for sterling and that it would not cooperate as long as

78

Agreement between Federal Reserve Bank of New York and the Governor and Company of the
Bank of England, August 1, 1931; agreement concluded on August 1, 1931, between the Bank of
France and the Bank of England.
79

Clay, o p . cit., page 386.

80

Harrison conversation with Lacour-Gayet on August 7, 1931 (Harrison Papers).

81

Crane cable to Harrison, No. 4, August 10, 1931.

82

The Economist, August




8, 1931, page 254.

207

CHART 9

SUPPORTING STERLING IN CRISIS: A PARTIAL VIEW
JULY 13-SEPTEMBER 21,1931

Note: Support operations include Bank of England dollar sales, as reported in daily
cables to the Federal Reserve Bank of New York, and purchases of sterling by the
New York Reserve Bank. All told, these operations amounted to somewhat over
$400 million in the period July 23-September 19,1931, roughly two fifths of the
more than £200 million ($972 million) reserve losses officially stated to have been
sustained by the British authorities during the two months before sterling went
off gold. The chart does not include other support operations, particularly those
conducted on the basis of the credits extended by the Bank of France and the
Paris market. The sterling exchange rate shown in the chart is the average of noon
buying rates for cable transfers in New York as certified by the Federal Reserve
Bank of New York to the Secretary of the Treasury.

The gold export point from London to New York, roughly approximated by the line
in the chart, was defined as that exchange rate below which it was advantageous
for holders of sterling to purchase gold from the Bank of England, ship the metal
to New York, and sell it to the United States authorities in order to acquire dollars.
t Bank holiday weekend.

208




Norman remained head of the Bank of England.83 Although vigorous support
operations were quickly resumed under the market arrangements that had been
concluded toward the end of July, negotiations to resolve the problem of drawing
on the Bank of France credit dragged on until August 7 when it was decided that
the entire $125 million equivalent would be drawn at once in the hope that the
Bank of England could by this means conceal its operations from the Paris
market.84 In the meantime, however, as Lacour-Gayet lamented, the "moral
effect" of the credits was being lost.85 Pressure on sterling, which had been light
on August 4, became heavy in the days that followed. A fair, though incomplete,
notion of the hectic developments in the exchange markets is given by the daily
total of support operations in the form of Bank of England dollar sales and
Federal Reserve sterling purchases, comparable data for support operations in
other forms being unavailable. These support operations, which amounted to only
$0.5 million on August 4, mounted to $9 million on August 8 and to $22 million
on August 10. They slackened somewhat thereafter, but they mounted to a peak
of $45 million on August 24, the day the Labor government resigned.86
Most of the story of the raising
of the loans to the British government from New York and Paris commercial
bankers in the latter part of August has already been told, most notably by Sir
Henry Clay's biography of Lord Norman and in Reginald Bassett's Nineteen
Thirty One.87 Accordingly, it is not necessary to recount the details of Snowden's
and MacDonald's conversion to the view that additional loans would have
to be raised in New York and Paris, the approach to private bankers in those
centers, the communications from the bankers to the effect that loans could
be raised only if the British government were to adopt a program of budgetary
reform that had the full approval and support of the Bank of England and the
THE BANKERS' LOAN TO THE BRITISH GOVERNMENT.

83

Papers Relating to the Foreign Relations of the United States, 1931, Vol. I, page 18;

Sproul

conversation with Catterns, August 5, 1931; and Hurst, op. cit., page 640.
84
The agreement was facilitated when the Federal Reserve waived its right to pari passu
drawings
with the Bank of France. The Paris commercial banks eventually participated
to the extent of half
the credit. Crane cable to Harrison, No. 4, August 10, 1931; Bank of France A n n u a l Report, 1931
published in Federal Reserve Bulletin (March 1932), page 162; and Harvey cable to Harrison, No.
280, August 7, 1931.
85
Harrison conversation
with Lacour-Gayet
on August 6, 1931 (Harrison
Papers).
86
Crane cables to Harrison, No. 292, August 11, 1931, and No. 24, August 24, 1931; Clay, op. cit.,
page 386; Reginald Bassett, Nineteen Thirty-One (London: Macmillan & Co., Ltd., 1958), page 155.
87
See especially Clay, op. cit., pages 387-95; Bassett, op. cit., Chapter 9.




209

City of London generally, the refusal of the Henderson faction to accept
Snowden's budget balancing proposals which included a 10 per cent cut in relief
payments to the unemployed, the resignation of the Labor government, the
subsequent bitter allegations that Snowden's budgetary proposals had been
adopted under pressure from foreign bankers, and particularly that the Federal
Reserve Bank of New York had informed the British government that further
credits would be granted to the Bank of England only if economies in the dole
were enforced and the British government's subsequent flat denials that these
allegations contained any truth.
The part of the story that has not hitherto been published relates to Harrison's
efforts both to obtain substantial additional funds to support sterling and to
avoid entanglement in the bitter political controversy that, as he was fully aware,
was then raging in Britain. It has already been noted that, repeatedly during the
latter part of July, Harrison had urged Norman and Harvey to encourage the
British government to approach the New York market for a long-term loan.
Although Norman was dubious, Harvey on July 30 indicated not only that he
heartily agreed with Harrison's suggestion but that he had already raised the
question with the British Treasury.88 Throughout early August the Bank of
England exerted its utmost efforts to hasten the formulation of the government's
proposals to reduce expenditure and balance its budget and, foreseeing that its
efforts might soon bear fruit, the bank approached Harrison on August 13 for
advice on the raising of a foreign loan for the British government.89 Harrison's
response was that such a loan "should be arranged jointly in Paris and New York
since it is essential that Paris and London should work together", and that it
should be arranged "promptly after the announcement of a strong budget program". Harrison himself did not venture to say how much might be raised but
other senior officers of the bank guessed that "perhaps $250,000,000" could be
obtained in New York if sterling were kept strong and if the government were
to announce "a very convincing budget reform".90
On August 22 Harrison received from Harvey an outline of Snowden's proposal
for balancing the budget, and on the following day he discussed these proposals
with Norman who was recuperating at the Chateau Frontenac in Quebec.

88

Harrison conversation with Harvey, July 30, 1931 (Harrison
Papers).
Crane cable to Harrison, No. 4, August 10, 1931; Burgess memorandum
90
Burgess cable to Crane, No. 10, August 14, 1931.
89

210




to files, August 4, 1931.

Harrison told Norman that Morgan's had received messages from London
indicating that MacDonald would resign if either Morgan's or Harrison himself
felt the program inadequate. Harrison then told Norman, according to his record,
that:91
both Morgan's and I thought that was not a fair question to put up to us; that
while we were certain that a very definite budgetary program must be put through
if they are to float a loan here, nevertheless it was impossible for us at this distance
to determine whether the proposed program was adequate enough . . . . Norman
agreed with this one hundred per cent . . . . He said that was obviously a question
to be determined in London; that we could not do so here. I told him that Morgan's
had finally cabled them to ask whether the program was satisfactory to the Bank
of England and the City, including their own partners; that I thought that was the
best way they could leave it. To this Norman also agreed. I then mentioned I had
asked Harvey several times whether he thought the program was adequate and while
he generally implied he thought it was, I got no definite answer from him. Norman
intimated that I probably would not get a definite answer from him but that he,
Norman, felt that the program was inadequate; that we must not fool ourselves
now; that any inadequate program would cause trouble in a year or so and that
it is essential that we must force an economic adjustment now and not in a year
or so from now; that the program, in his judgment, must be sufficiently drastic to
place the cost of output and wages on a competitive basis with the rest of the world
and unless that were done he was certain the program would not be adequate;
that if the Government attacked the situation courageously and did enough by
way of drastic readjustment then, in his judgment, they would not need a credit
at all. I said the difficulty with that was that the time was so short and the central
bank credit so nearly exhausted, it was a risky thing to take the chance on even a
drastic budgetary reform checking the withdrawals. Norman allowed there was
something in this point. The sum and substance of our conversation indicated pretty
clearly, however, that Norman felt the proposed program was not sufficiently severe
to avoid trouble later on and that there was no use in doing the thing halfway.

Inadequate though Snowden's proposals may have been, they proved too much
for Arthur Henderson and his followers to swallow. The Labor government split
and resigned, effective August 24. Late the same day a "national" government
was formed composed of members of the Conservative and Liberal Parties and
led by MacDonald (who remained Prime Minister) and Snowden as Chancellor
of the Exchequer. The stage was thus set for the announcement that the new
government would implement Snowden's proposals and for the raising of the
loans in the New York and Paris markets. By August 25 the British authorities
were assured that $200-250 million could be obtained in the New York market

91

Harrison conversation with Norman on August 23, 1931 (Harrison Papers),




211

if a similar amount could be raised in Paris.92 On that day Harrison called
Lacour-Gayet. He learned that Harry Siepmann of the Bank of England had
discussed with Moret a loan of $120-235 million but that Moret felt the amount
should be nearer the lower than the higher figure. For his part Harrison emphasized that this time the job "must be done once and for all and not piecemeal",
and that the loan should be adequate in amount "to do the trick" and long enough
to avoid any "embarrassing maturity". He then asked Lacour-Gayet93
how quickly it would be possible to place the credit in Paris. He said he
thought three or four days. I told him that I thought it ought to be done immediately . . . . He, figuratively, threw up his hands and said it was not possible, that
it was now 5 o'clock in Paris and that Moret has not even yet been authorized to
speak to any of the private bankers. I was astounded at this, and Lacour-Gayet said
that neither he nor Moret got any impression from Siepmann that there was any
imminence about this credit; that he had come to talk in the most general terms;
that he was now flying back to London, and that until they were authorized to
talk to the private bankers there was nothing more he could do about it. I asked
whether, if they got immediate authority to approach the private banks, it would be
possible to talk to some of the bankers tonight. He said possibly, but that many of
them have already left for the day and gone out of town. I urged upon him what I
considered the imminence of the British position, especially as the Bank of England
has only £1,000,000 left in the [central bank] credit. Lacour-Gayet said he could
not understand that as there was $8,000,000 not used in the French tranche. I said
I got the figures from Crane last night; that he indicated there was only £1,000,000
left in all, and that I felt in any event it was necessary to do something by tomorrow
night or Thursday morning at the latest. I asked him whether, if they got authority
to go to work at once, it might not be possible to do it by that time. He said he
supposed it might be but it would be difficult to act in such a hurry. On the whole,
he was most sympathetic and helpful in his attitude, but somewhat surprised at our
thought that there was any such rush . . . .

Harrison and Lacour-Gayet discussed the loan further on August 26. Both
agreed that it would be desirable to raise at least $200 million in each market.
Harrison emphasized that to raise $100 million in each market would be misunderstood and would cause more questioning and doubt than it would do good.
With this Lacour-Gayet agreed but felt nevertheless that, without a public issue,
the Paris banks might be able to raise only $100 million. The French private
banks, he said, were not so organized to take up as much as $200 million by

92
J. P. Morgan & Co. cable to Morgan Grenfell & Co., August 24, 1931; Harrison conversation
with Lacour-Gayet on August 25, 1931 (Harrison Papers).
93

Harrison conversation with Lacour-Gayet on August 25, 1931, pages 4-5 (Harrison Papers).

212




themselves on short notice. On the other hand, a public issue would require
considerable time to arrange and might lead—as had happened on earlier occasions—to stringency in the Paris market which would attract balances from
London to Paris, the very thing that the loan was aimed to discourage.94
In the end, it proved possible to raise a one-year $200 million loan in Paris
but only half the amount was obtained from the banks, the other half being
provided by a public issue which was not floated until September 14. Nevertheless, agreement to provide the loan was announced on August 28 at the same time
as the $200 million loan from a group of New York banks headed by J. P.
Morgan & Co.95 Thereafter Parliament reconvened on September 8, and on
September 10 Snowden, as promised, presented an emergency budget under which
revenues were increased and expenditures cut to the extent necessary to change
the deficit anticipated in the government's accounts into a small surplus.96
Although the formation of the "national" government, the
announcement of the New York and Paris loans, and the emergency budget were
all well received in the principal foreign markets, the pressure on sterling was
only temporarily relieved. Basic problems remained unresolved. In particular, the
bitterness in British domestic politics was far from reassuring. The "national"
government had been formed to deal only with the crisis; an election was promised
as soon as the emergency had passed. The hostility of the opposition to Snowden's
budgetary reforms intensified; allegations of a Bankers' Ramp spread despite the
official denials.97 A manifesto, published on August 27 jointly by the Labor
Party and the Trades Union Congress, predicted that cuts in wages and salaries
would "bring about embittered conflict and industrial chaos".98 There was much
private discussion within the Labor Party and elsewhere about whether sterling
THE FINAL ONSLAUGHT.

94

Harrison

95

Crane cable to Harrison,

conversation

with Lacour-Gayet
No. 19, August

96

Parliamentary Debates (Great Britain,
Exchequer
to the House of Commons,
September 12, 1931, pages 460-62.

on August

26, 1931 (Harrison

Papers). Supra, page 171.

28, 1931, and Clay, o p . cit., pages
House of Commons), Statement
September
10, 1931, columns

397-98.

by the Chancellor of the
297-313; T h e Economist,

97

The phrase "Bankers' Ramp" was used by dissident Labor Party members and others in Britain to
describe the pressure that was alleged to have been exerted on the Labor government
by foreign
financiers. In rejecting these allegations in an August 25, 1931 speech, Prime Minister
MacDonald
used the phrase as synonymous with conspiracy. In Webster's Dictionary (third international)
"ramp"
is described as obsolete slang meaning swindle, confidence game, dupery.
98

The text of the manifesto




was published

in T h e N e w Y o r k Times, August

28, 1931, page 8.

213

would or should be devalued although little of this was reflected in the press."
Disturbances in the Royal Navy, arising from pay cuts made under the emergency
budget, were magnified in the foreign press into a serious mutiny.
Long before the so-called mutiny, of course, Continental bankers had been
drawing down their sterling balances in order to reinforce their cash positions and
so safeguard themselves against the threat of new currency panics. Withdrawals
from London by Dutch and Swiss commercial banks were particularly heavy
because they, like their London and New York counterparts, had substantial
commitments in Germany which were frozen under the standstill arrangements
that were announced September 9. But there can be little doubt that the so-called
mutiny, announced on September 15, precipitated the final onslaught against
sterling. In the days that followed, prices on the London stock exchange and
on the Paris, Amsterdam, and Zurich bourses declined sharply and continuously.100 The British 5 per cent War Loan, widely held by foreigners, which had
been quoted at l00 1/2 on September 9, was down to 971/2ten days later. Especially in the Dutch financial markets, conditions approached panic as the banks
curtailed credit and repatriated funds from London.101
The scanty evidence that exists suggests that only a small part of the pressure
against sterling was on official foreign account. Central bankers were nervous but
generally seemed to have refrained from withdrawing the bulk of their sterling
balances.102 At least one western European central banker approached the Bank
of England about the beginning of September for a gold guarantee of his sterling
balances, was refused, and invited to take gold then and there if he wanted it
but after considering the matter decided not to withdraw his funds.103 On the
other hand, another important European central bank, which apparently held
significant short-term balances in London, is known to have converted all but
£300,000 into gold before September 20.104

99
Crane, "Notes on Visit in England" included in Confidential Memorandum
September 16, 1931, and T h e Economist, September 19, 1931, page 504.
100

T h e Economist, September

101

Hurst, o p . cit., pages
Hurst, o p . cit., pages

Visit,

656-59.

102

of European

643-47.

26, 1931, page 554.

103

Crane cable to Harrison, No. 20, September
Harvey on October 2, 1931 (Harrison
Papers)
104

Crane, Report

214




of European

Trip, November

3, 1931; Harrison
21, 1932.

conversation

with Norman

and

The information on the behavior of the French banks during the crisis is conflicting. Crane was told in the course of various visits to London and Paris that
toward the end of August 1931 the Bank of France still held about £ 8 0 million
in short-term sterling balances, that French private banks held about £40 million,
and that by September 21 the Bank of France's balances had been cut to £65
million, a reduction equivalent to about one fifth the amount of the Paris
credits.105 Market gossip was that the Paris private bank balances were down to
£10 million by the time sterling went off gold.106 On the other hand, in his
September 21 post mortem, Snowden went out of his way to chide the critics of
the French banks, saying that his information was that "these banks have not
played any part in the recent withdrawals from London, but have maintained
their balances practically intact. . . ."107
As the pressure against sterling built up in the early days of September, the
Bank of England held the spot rate within a narrow band around $4.86, roughly
1/2 cent above the gold export point to New York. It also supported sterling in the
forward market especially after September 8 when Harrison was informed that
the London clearing banks would thereafter be free sellers of forward exchange.108
These exchange market tactics, which drew severe criticism in New York, were
based on the view that any decline in the spot rate or widening in the forward
discounts would sap such confidence in sterling as remained, would impair the
confidence-making effects of the budgetary measures to be announced by the
Chancellor on September 10, and would jeopardize the success of the Paris
market issue on September 14.109 For a time, Harrison deferred to this view
which was held by the authorities in both London and Paris but, when the same
tactics were continued after the flotation of the Paris issue, he strongly emphasized
to Harvey that the feeling in New York was that the maintenance of such a high
rate was encouraging sales of sterling, that no one was "satisfied with continued

105
It seems clear that the £80 million, which was held almost entirely at London commercial banks,
did not include the sterling counterpart of drawings on the credit from the Bank of France. Under
the loan agreement the sterling counterpart was to be held at the Bank of England. Crane cable to
Harrison, No. 14, August 23, 1931. See also Crane conversation with Moret, August 26, in his
Confidential Memorandum of European Visit, September 16, 1931, and Crane, Report on European
Trip, November 21, 1932.
106

Hurst, op. cit., page 642.

107

Parliamentary Debates (Great Britain, House of Commons), September 21, 1931, column 1297.

108

Harvey cable to Harrison, No. 411, September 8, 1931.

109

Ibid.; Harvey cable to Harrison, No. 402, September 4, 1931.




215

use of the credit to maintain sterling at such a high point", and that sentiment
generally "was questioning and restless". He also reported that Norman, whom
he had just seen, was110
much disturbed about the way the exchange situation was handled and he
agreed with us that it was most important not to peg it, that every peg in past
history had proved disastrous, that he saw no occasion for it at this time and that
the chief reason he was hurrying home before he was quite ready to do so was to
see if he could not make some headway in handling the exchange situation more
satisfactorily.

However, it is doubtful whether, at this stage, any change in exchange market
tactics would have made an appreciable difference. The pressure was again
snowballing fast. By Wednesday, September 16, almost half of the New York and
Paris loans had been either sold in the spot market or committed against forward
contracts.111 That day the Bank of England's gold and foreign exchange loss was
£ 5 million; it was £ 1 0 million on Thursday, nearly £ 1 8 million on Friday, and
£ 10 million on Saturday when the markets were open only half a day.112 At the
close of business on September 19, the Bank of England's holdings of gold and
foreign exchange exceeded by only £ 5 million the total of its obligations under
forward exchange contracts, the central bank credits, and the New York and
Paris loans.113 More than £200 million of gold and foreign exchange had been
lost since July 13.114
The final onslaught came, as Snowden said on September 21, with "appalling
suddenness". The balancing of the budget, which had been considered the key
to the defense of sterling and for which MacDonald and Snowden had accepted
the breakup of the Labor government, was in vain. Less than a week after the
budget message, the British authorities were again confronted with an urgent need
for new measures to save sterling. Without much hope, they approached Washington about the possibility of some new initiative on the part of the Hoover

110

Harrison conversation with Harvey on September 16, 1931 (Harrison Papers).

111

Harrison conversation with Harvey on September 16, 1931, page 1 (Harrison Papers).

112
Parliamentary Debates (Great Britain, House of Commons), Snowden statement in Parliament,
September 21, 1931, columns 1293-99.
113

Harrison letter to Glass, March 23, 1939 (Harrison Papers).

114

Parliamentary Debates (Great Britain, House of Commons), Snowden statement of September
21, 1931 to the House of Commons, column 1295; Clay, op. cit., page 398; and Harvey cable to
Harrison, September 20, 1931.

216




administration toward the permanent elimination of reparations and war debts.
The question whether additional credits might be obtained was also raised with
both the United States and French governments. On neither count were the
responses encouraging.115 Nor did the British authorities feel there was anything
they themselves could do. From New York and the Continent they had received
advice that a further increase in the Bank of England discount rate would only
increase the distrust of sterling and that a further tightening of credit could not
operate quickly enough to stem the reserve losses.116 The possibility of mobilizing
British private holdings of foreign securities was considered and rejected on the
ground that the resources that could thus be obtained would have been inadequate
to check the huge reserve drain. On similar grounds, it was decided that the Bank
of England should conserve its remaining gold which amounted to £136 million
on September 16. Foreign balances in London were still a multiple of the Bank's
gold holdings, and it was believed that the release of even a large proportion of
these holdings would not have stopped the drain but would, as Snowden said,
only have benefited those who were showing the least confidence in sterling.117
At 6:30 a.m. New York time on Saturday, September 19, Harvey telephoned
Harrison to say that, in the absence of any alternative, the British authorities had
come to the conclusion that they would suspend gold payments effective Monday,
September 21. After some discussion of market operation, Harrison asked118
whether there was not anything left to do, especially in view of the fact that
withdrawals today seem to have let up so much as compared with yesterday. He
said he was afraid not, that they had reached the end of their tether and that so far
as he could see there was nothing left to do. But, I said, you have $75 million left
in the Paris credit and with that as a backlog and possible change in conditions
and some restrictions, might they not be able to control the situation. He said that
the $75 million left in Paris was but a drop in the bucket considering the rate of
withdrawals yesterday, and that it was hard to tell at the time he was talking how
much has been disposed of today; and that furthermore there was still a half hour

115

Harrison conversation with Harvey on September 16, 1931, page 4, and Harrison conversation
with Harvey at 8:35 a.m. on September 19, 1931, page 2 (Harrison Papers). Osborne cable to the
Marquess of Reading, No. 578, September 20, 1931, Documents on British Foreign Policy, 19191939, 2nd Series, Vol. II, 1931, page 262; and Clay, op. cit., page 398.
116

Clay, ibid., page 387.

117

Parliamentary Debates (Great Britain, House of Commons), September 21, 1931, columns 1295-96.

118

Harrison conversation with Harvey at 6:30 a.m. on September 19, 1931, pages 3-4 (Harrison
Papers).




217

left to go without any assurances as to what the total already is. I asked him whether
with some evidence of slowing up today there were not some restrictions which
might be put in together with the $75 million leeway that might turn the tide.
Harvey was of the definite impression that there was nothing left for them to do;
that in 1914 they had many powers which they do not now possess and that he did
not see any way of effectively stemming the tide. I said it seems a great pity to let
it go and asked whether there was anything we could do within reason. Harvey
replied that he thought we had already done a great deal and that he saw nothing
else which we could do to help, that there was no alternative left.

The same day the Bank of England addressed a formal letter to the government advising that the bank be relieved of its obligation to sell gold at a fixed
price under the provisions of the Gold Standard Act of 1925. This advice was
formally accepted at a Cabinet meeting on Sunday, September 20. The suspension
of the gold standard was announced to the public the same evening and on the
following day, with the stock exchange closed, legislation was enacted by both
houses of Parliament giving the government's decision the force of law.119

EPILOGUE

Britain's abandonment of gold came as an unexpected shock after two years
during which numerous other difficulties had already weakened and depressed
the international economy, and was a major factor in the final plunge of the
depression to the depths of 1932. The sequence of events in this further twist of
the deflationary spiral is clear. When the flight from sterling forced the British
currency off gold, the repercussions on international liquidity were immediate.
Those foreigners—and there were many—who elected to continue to hold
London balances found the gold value of these holdings reduced by 30 per cent
as a result of the drop in the sterling exchange rate between mid-September and
the end of 1931. Others, stopped from obtaining gold in London, shifted into
dollar balances in order to obtain the metal in New York. At the same time,
other holders of dollar balances, mainly foreigners, were also converting their
funds into gold. The United States gold stock consequently declined by 11 per

119

The text of the communications between the Bank of England and the government regarding the
suspension of the gold standard is printed in Bassett, op. cit., pages 237-39. See also Clay, op. cit.,
page 398.

218




cent in the four months ended December 1931. Federal Reserve gold holdings,
which for years had been far in excess of the amounts required, dropped close
to the legal minima.120 Despite the depressed condition of the United States
economy, the Federal Reserve felt obliged sharply to increase discount rates, the
rate of the New York Reserve Bank being increased 2 percentage points to
31/2per cent, and to squeeze member bank reserves during the remainder of
1931. For similar reasons, similar policies were pursued abroad. The Bank of
England increased its discount rate by 1 1/2 per cent to 6 per cent upon the
announcement of the abandonment of gold while the Reichsbank, after having
reduced its rate from the 15 per cent established at the beginning of August 1931,
maintained the rate at 8 per cent in the three months ended December 10. Even
in France, in which much of the flight capital sought refuge and where the
authorities were already pursuing a very conservative policy, the central bank
signaled the beginning of an even more cautious course by increasing its discount
rate early in October.
Under the impact of the sharp decline in international liquidity, more stringent
monetary policies, and deepening business pessimism, financial difficulties multiplied and production and trade contracted further. Comparing the second half
of 1931 with a year earlier, United States commercial bank suspensions rose
86 per cent while defaults on foreign dollar bonds increased seventeenfold.
In most industrial countries the decline in industrial production which had slowed
during the spring of 1931, accelerated during the following autumn and winter
and unemployment reached new highs.121 In the external sphere, international
commodity prices continued to decline. Moreover, Britain's abandonment of
gold inaugurated a period of fluctuating exchange rates that contributed, together
with declining national income and expenditures and rising tariffs and other
barriers, to the stifling of international trade. The upshot was that the value of
world exports in 1932 was about one third lower than it had been the year before
and only two fifths that of 1929. The collapse of a key currency thus signaled
the end of the international financial system established in the 1920's and contributed substantially to the disruption of the international economy.

120
E. A . Goldenweiser,
pages 121-23, 158-60.
121

League of Nations,




American Monetary Policy (1st ed.; New York: McGraw-Hill,
Statistical Year-book, 1931-32, pages 58-59, 176-86; 1932-33, pages

Inc., 1951),
146-56.

219

9.

Conclusion

Judged by its overall accomplishments in 1924-31, central bank cooperation
deserves only mixed marks. For the period up to June 1928, the record has
considerable merit. Cooperation facilitated the stabilization of several major
currencies and numerous minor ones. The strains that arose after these stabilizations were manageable. Deficit countries such as Britain and Germany adjusted slowly, but the pace of economic change elsewhere was not so rapid or
abrupt as to make these adjustment efforts hopelessly inadequate. The acute
monetary difficulties that arose were handled, not without hard bargaining, by
compromise and accommodation. Had world trade continued to expand and had
the United States economy remained reasonably prosperous, exchange stability
might have been maintained at a cost in terms of unemployment no greater than
the deficit countries were willing to pay. It is even conceivable that, had the
economic weather remained fair, some of the deficiencies in the cooperative
arrangements—for example, the rather narrow limitations on the amount of
official credit available or the one-sided burden that the adjustment process
placed on deficit countries—might in time have been at least partially remedied.
On the other hand, the record of central bank cooperation in the period after
mid-1928 must be judged a failure. This failure stemmed not so much from the
deficiencies of central bank cooperation itself as from the inability of the authorities—including particularly those in the United States—to manage their domestic
economies successfully. For the breakdown of the American economy immensely
complicated the problems of central bank cooperation. In addition to destroying
the opportunity for any gradual improvement in cooperative techniques, the
collapse greatly accelerated the pace of economic change and intensified the
pressures to which the international financial system was subjected. The rate at
which incomes and prices dropped, export markets narrowed, trade barriers
were erected, and the direction of capitalflowsshifted far outstripped the capacity
and willingness of individual countries to adjust. And from these basic economic
changes arose major political and monetary difficulties with which the fairweather cooperative arrangements of the 1920's were completely unable to cope.
But part of the failure also stemmed from rigidities in the views of the
authorities themselves. They, in common with most economists and the public
generally, failed to understand until it was too late that the economic breakdown
of 1929-31 had created problems that were entirely different from those of the
early and mid-1920's. They, therefore, followed policies which had been success220




ful in earlier years but which—applied in the circumstances of the Great Depression—sometimes aggravated the problems they were intended to solve or, when
they did work in the right direction, were inadequate to deal with those problems.
The latter was certainly true of the central bank credits that were arranged during the 1931 crisis. Even so, it is difficult to conceive of any extension or modification of central bank cooperation—within the realm of what was then
considered possible—that would have been sufficient to maintain exchange rate
stability in the circumstances of political and economic disintegration that
prevailed. Indeed, the conclusion seems inescapable that the failure of central
bank cooperation in 1929-31 was only part of the larger failure of the Western
democracies to deal successfully with the economic and political problems of
the time.
The requisites for successful international financial management in the complex conditions of 1924-31 were many. The most basic need was to develop
economic tools that would insure stable growth without inflation in the major
countries. But stable growth—although necessary—was not sufficient for success.
It was also necessary for each national component to mesh with the rest of the
world economy, and to do so on the basis of stable exchange rates and without
resort to restrictions on trade and payments. Hence there was need to develop
(1) more effective policy instruments by which each country could foster
adjustments to economic changes elsewhere in the international system, (2)
better arrangements under which—if need be—international credit could be
provided to finance such adjustments as well as to maintain orderly conditions
in the exchange markets and to deal with unforeseen emergencies, and (3) a
generally agreed conception of the international economy that could provide
criteria for the administration of adjustment instruments and credit facilities.
Meeting these various needs would not, of course, have eliminated all the difficulties of international economic management in 1924-31. But their fulfillment
would have improved the chances for handling the problems that arose under
relatively calm conditions rather than under the overwhelming pressures of world
depression and financial crisis.




221

Appendix

CHRONOLOGY OF IMPORTANT EVENTS, 1919-31

1919:
June 28

1920:
January 16

Versailles Treaty is signed by representatives of the Allied and
Associated Powers and Germany. It is declared in force,
January 10, 1920, when the majority of those powers—but
not the United States—ratified it.

The League of Nations holds its first meeting in Paris.

February 23

The Reparation Commission meets in Paris. It is a successor
to another similar body formed during the Peace Conference
in 1919.

July 16

The Spa Conference of Allied and German ministers agrees
upon percentages for division of German reparation payments
by the Allies.

1921:
May 11

August 25

1922:
February 9

February 15

222




Germany accepts total reparation obligation of $33 billion, as
set by the meeting of the Allied Supreme Council in London.
German-American Peace Treaty is signed in Berlin by representatives of German and American presidents.

United States War Debt Refunding Act, which provides a basis
for negotiations on war debts with the Allies, is signed by
President Harding.
Permanent Court of International Arbitration established at the
Hague with thirty-three member countries, but not the United
States.

May 19

Twelve Resolutions of the Financial Commission of the Genoa
Conference, which include calling a central bankers' meeting
and establishing the gold exchange standard, are approved by
the conference.

August 1

Lord Balfour announces that Great Britain, while favoring the
surrender of its share of German reparations and the writingoff of the whole body of inter-Allied indebtedness, will seek to
collect debt payments from the Allies and Germany only
sufficient to pay its debt to the United States.

September 21 President Harding signs the Fordney-McCumber Tariff Act,
which raises rates on many agricultural and manufactured
products to protect the farmers and certain wartime industries.
The act also provides for adjustments by the President to
equalize costs at home and abroad.
December 29

1923:
January 11

Secretary of State Hughes in New Haven proposes an investigation by experts of Germany's capacity to pay reparations.

French and Belgian troops occupy the Ruhr, following the
Reparation Commission's declaration of Germany's default on
reparation payments.

February 28

British-American war debt agreement, as approved by the
Senate, is signed by President Harding. It is ratified by Great
Britain, June 19, 1923.

October 16

Rentenbank is established as a means to halt German inflation.
The currency is stabilized on November 15, 1923 on the basis
of one rentenmark equal to one trillion paper marks.

1924:
January 14




First Committee of Experts, with Dawes as chairman, begins
to work to determine ways of balancing Germany's budget and
reinforcing the stabilization of the German currency.
223

January 21

Second Committee of Experts, with McKenna as chairman,
begins to estimate the amount of German capital abroad.

April 9

Two Committees of Experts publish their reports to the Reparation Commission.

August 16

Protocol of the London Conference, to put the Dawes Plan
into operation, is signed by government representatives. By
August 30, 1924 the necessary German legislation is passed,
including that for the Reichsbank.

September 1

Dawes Plan for payment of German reparations goes into
operation.

1925:
January 14

United States and Germany sign a separate agreement for payment by the latter of American costs of occupation and mixed
claims. Conference of Allied Finance Ministers in Paris sign
agreement for division of reparation receipts.

April 28

Churchill announces that Great Britain is returning to the
gold standard at prewar parity.

July 31

French and Belgian troops evacuate the Ruhr, according to the
London Protocol of August 16, 1924.

December 1

Locarno nonaggression pacts are signed by European government representatives. They are ratified by sufficient countries
to be declared in force February 14, 1926.

1926:
January 30

April 29

224




British forces evacuate Cologne, as provided for in the London
Protocol.
Secretary of the Treasury Mellon and Ambassador Berenger
sign French-United States debt agreement. It is ratified by
France, July 26, 1929 and by the United States, December
18, 1929.

May 1

British coal strike, followed on May 4 by general strike, in
protest of wage cuts by mine owners and the Royal British
Coal Commission Report. The coal strike lasts seven months
and the general strike nine days.

July 12

Churchill
and Caillaux sign British-French war debt agreement,
which comes into force on July 27, 1929 after it is ratified by
France.

September 8
1928:
June 25

August 27

Germany is admitted to the League of Nations.

French law, stabilizing the franc at one fifth the 1914 parity
and establishing the gold standard in France, becomes effective.
Kellogg-Briand Peace Pact, renouncing war as an instrument
of national policy, is signed by fifteen government representatives. It is ratified by France, March 29, 1929, by the United
States, January 16, 1929, and declared in force by President
Hoover, July 24, 1929.

November 22 Currency and Bank Notes Act, adopted July 2, 1928, comes
into effect establishing final form of gold standard in Great
Britain.
1929:
February 11

Committee of Experts, headed by Young, meets in Paris to
setfinalplan for German reparation payments.

June 7

Young Committee submits its plan to the Reparation Commission.

August 31

End of first Hague Conference to draw up plans for putting
Young Plan into operation.

September 20 Hatry stock scandal in Great Britain occurs as the New York
stock market prices begin to weaken.




225

October 24
1930:
January 20

New York stock market panic starts.

Protocol is signed at second Hague Conference on German
reparations, to put Young Plan into operation. Necessary German legislation is passed and countries ratify by May 17, 1930.

May 17

Young Plan goes into operation, Bank for International Settlements is established, and Dawes Plan ends.

May 19

The Reparation Commission is liquidated in accordance with
the Young Plan.

June 17

Hoover signs Hawley-Smoot Act which raises United States
tariffs to new highs.

June 30

The Rhineland is evacuated by Allied forces, according to the
Hague Protocol.

November 6

Oustric stock scandal in France ruins several banks and many
small investors.

1931:
March 21

Germany announces plans for an Austro-German customs
union. Plans are abandoned, September 3, 1931.

May 11

Credit-Anstalt, largest bank in Austria with international investments, reports heavy losses.

June 20

Hoover proposes a year's moratorium in intergovernmental
debt payments.

July 13

Macmillan Report is published revealing large foreign shortterm claims on London.

July 13

Danat Bank is closed in Berlin, followed by closing of Berlin
bourse and promulgation of exchange controls.

226




July 23

Seven Power Conference in London, to consider ways to provide capital for Germany, adjourns.

July 31

May Committee issues report on needed economy in British
government expenditures.

September 15 British sailors "mutiny" in protest of economy measures of
new national government.
September 21 Gold Standard Amendment Bill is passed in Great Britain to
suspend the gold standard.




227

Index:
Page
Arbitrage
See In spot and forward markets under
Exchange Rates, Foreign.
Australia
Austrian Credit-Anstalt
Austrian Government

80

182,
185,187-89,226
188, 189

Austrian National Bank
Austro-German Customs Union

186-88
181,
184, 188,226

Balance of Payments
Effect of sterling area trade on British
93
France
112, 120, 123,148, 173
Germany
173
Great Britain
103, 173
Statistics
72
United States
22, 71, 92,
97, 128, 148, 173
Balfour, Lord Arthur James
10, 223
Bank for International Settlements
Credit to Austrian National Bank.... 186-88
Credit to Reichsbank
192
Currency support operations
174
Organization
147
See also McGarrah, Gates W., and
Quesnay, Pierre.
Bank of England
Credits to
28, 77, 78, 81-85, 204-9
Credit to Austrian government
188
Credit to Austrian National Bank
186-88
Credit to Reichsbank
58, 191-97
Currency support operations
58,
207-9,215,216
Discount rate
87,89,101,102,149,
159, 170, 172, 190,203,219
Dollar holdings
105, 134, 139, 142n
Fiduciary issue
139-41, 163, 207
Gold buying price
90, 133
Gold holdings
39, 90, 99, 102, 105, 108,
134,139,159,173,203,217
Gold payment suspension
218

228




Page
139, 141, 142,
150, 181,202,216
Market operations
128
Monetary policy
86, 88, 89, 99-102, 128,
159, 170, 172, 173, 203, 218, 219
Role in stabilization efforts
62
Securities holdings
89, 101
See also Harvey, Sir Ernest M.; Lubbock,
Sir Cecil; Niemeyer, Sir Otto E.; Norman,
Montagu C ; Siepmann, Harry A.; Sprague,
Oliver M. W.; and Stewart, Walter W.
Bank of England Act of 1844
28
Bank of France
Credit to Austrian National Bank
186-88
Credit to Bank of England
204-9
Credit to Reichsbank
191-97
Currency support operations
176, 205
Discount rate
149, 170, 172, 190, 219
Foreign exchange holdings
111, 119, 120
Foreign exchange operations
110,
136-38, 148, 167
Forward exchange operations
121, 122
Gold holdings
119, 120,
148, 166, 171, 177
Gold policy
32, 108, 117,
118, 136-38, 166
Gold purchases
117, 130, 132
International assets
119
Monetary policy
111,112,169,170,219
On Kindersley's proposal
179
Open market operations
43
Repayment of British loan by
111
Sterling holdings
39, 44,
116, 118,167,214
Swaps
44, 121, 122, 148
See also Lacour-Gayet, Robert; Moreau,
Emile; Moret, Clement; Quesnay, Pierre;
and Rist, Charles A.
Bankers' Ramp
213
International reserves

Bradbury, Lord John
British Government
Credit to

10, 73
81-85

Page
108, 139, 140,

Fiscal policy

201,203,209,213
Gold embargo
78, 82
Gold payment suspension
218, 227
Loans to
209-13, 215-17
On bank's fiduciary issue
163
On reparations and war debts
20, 216
Political policy
21, 112
War debt payments to United States
90
See also Balfour, Lord Arthur James;
Churchill, Winston; Henderson, Arthur;
MacDonald, J. Ramsey; and Snowden,
Philip.
British Navy Mutiny

214, 227

Brown, William A., Jr
Bruning, Heinrich

34
10, 177, 184, 196

Capital, Foreign Short-term
Austrian standstill on
186, 188
Changes in
26, 93, 130, 164
In Austria
182, 185
In France
110
In Germany
149, 165,
177, 183, 185,202
In London
104, 131,
140, 185,202,214-18
In New York
97, 168, 218
Capital Issues
Yields

87,170

Capital Issues, Foreign
British control over
77, 104
In Great Britain
92, 104, 150, 170
Major sources
20
In United States
92, 97, 128, 170
In United States and Great Britain
25,
168, 173
Capital Movements

23, 62, 106, 108, 109

Case, James Herbert
Central Banks
Conflicting policies
Cooperation on interest rates
Cooperation principles




10, 97
42
76
33-44

Page
17, 28, 77-78, 81-85,
186-88, 191-97,204-9
See also Currency support operations
under particular central banks.
Discount rates
87, 149, 190
See also under particular central banks.
Gold policies
108, 114, 136
Relations
17, 41, 220
Stabilization conditions
41
See also Bank of England, Bank of France,
Federal Reserve Bank of New York,
Federal Reserve System, Reichsbank,
and the officers of these banks.
Credits to

Central Bankers
Meeting
National interest
On capital
Relations

117, 123
29
flows
168
17, 41, 108,
115,123, 134, 174
Views
34, 142
See also particular officials of the
central banks.
Churchill, Winston
Career
Information provided to
Role in banking policy
Views on
Discount rate
Gold standard
Clementel, Etienne

10
142
73, 119, 143
102
81,84,224
10, 54

Committee on the Currency and Bank of
England Note Issues
73, 78
Coolidge, Calvin
Crane, Jay E

10, 57
10, 180, 212

Credits
To Austrian government
To Austrian National Bank
To Bank of England

188
186-88
28, 77-78,
81-85,204-9
To central and eastern European
central banks
187n
To Reichsbank
58, 191-97
Cunliffe Committee

73, 102

229

Currency and Bank Notes Act

Page
139,
141,225

Darmstadter und Nationalbank
(Danat Bank)

Federal Advisory
193, 199, 226

Dawes, Charles G

10, 47

Dawes Plan
Agent General
Annuity terms
Control terms
Exchange provisions
Experts' Committees
Experts' Report

49, 56
48, 145
50-54, 145
49, 146
47, 223, 224
48-49, 51,
63-64, 66, 224
Governments' approval
51
International loan
49, 50-56, 67-70
London Conference of 1924
51, 53, 224
On Reichsbank's reserves
60-67
Position on the Gold Discount Bank.. .58-60
Transfer Committee
49, 52, 56
See also Gilbert, S. Parker; Kemmerer,
Edwin W.; Kindersley, Sir Robert;
Morgan, J. P., & Co.; Young, Owen D.;
and bank officers and government
officials.
Discount Rates
See Interest Rates and under Central Banks
and particular central banks.
Doumergue, Gaston
10, 111
Economic Conditions
Changes in
Depressed
Expansion
Recession in
Stock market boom

18-21, 23-26, 45, 97
171, 182, 219
127
126, 168
147, 159

Exchange Equalization Account

105

Exchange Rates, Foreign
Dollar-sterling
35, 66, 89, 102, 103,
130, 159, 160,202,208,215
In spot and forward markets
93, 94, 101,
121, 122, 128, 129, 160,215
Of franc
104, 110
Of franc, mark, pound, and reichsmark
in New York
19
Of reichsmark
45, 70n, 113

230




Page
17,20
18, 221

Stabilization of
Stability of
Council

64

Federal Reserve Bank of New York
Acceptance rates
157, 158
Conflict with Board
151
Credit to Bank of England
82, 83
Currency support operations
See under Federal Reserve System.
Discount rate
85, 87, 88, 99, 149,
156,158, 169,172,190,219
See also Case, James Herbert; Crane,
Jay E.; Harrison, George L.; Jay, Pierre;
Sproul, Allan; and Strong, Benjamin.
Federal Reserve Board
Direct action policy
151, 156
See also Goldenweiser, E. A.; Hamlin,
Charles S.; Meyer, Eugene; Platt, Edmund; Stewart, Walter W.; and Young,
Roy A.
Federal Reserve System
Bill holdings
86, 158
Bills and securities holdings
83, 157, 173
Credit to Austrian National Bank
187
Credit to Bank of England
206
Credit to Reichsbank
192
Currency support operations
42-44,130,
161,162,175,205
Discount rates
99, 124, 153, 158
Gold holdings
86, 130, 173, 219
Monetary policy
85, 86, 88, 97, 98, 124,
151-59, 169, 172,173,219
National interest
151
Reserves
124
Securities holdings
98, 158
See also Federal Reserve Bank of
New York, Federal Reserve Board,
Open Market Investment Committee,
and particular officials.
Foreign Capital Issues
See Capital Issues, Foreign.
Foreign Exchange Rates
See Exchange Rates, Foreign.
Foreign Short-term Capital
See Capital, Foreign Short-term.

Fraser, Leon
French Banks
On Dawes loan
Sterling balances
Loan to British government

Page
197
68
121-22, 215
213

French Government
Fiscal policy of
47, 111, 170
Morgan loan to
68
On Austro-German customs union
181,
184, 195
On war debts and reparations
45, 51, 53,
145, 195, 199
Political policy
21, 112, 181
See also officials, Clementel, Etienne;
Herriot, Edouard; Laval, Pierre; and
Poincare, Raymond.
Genoa Conference
34-35, 223
German Government
Exchange controls
113, 194, 199
Fiscal policy
113, 190
Loans to
177
See also International loan under Dawes
Plan.
On Austro-German customs union
181,
184, 195,226
On reparations
21, 45, 172,
177, 183, 184
See also Bruning, Heinrich, and Luther, Hans.
Gilbert, S. Parker
10, 57
Gold
Australian imports of
95
Central bank holdings
133, 142, 214
See also under particular central banks.
Dollar parity
17
Indian imports of
95
Prices in New York and London
132
United States stock
23, 24, 26,
90,95, 133, 148,218
See also Gold holdings under Federal
Reserve System.
Gold Discount Bank
.58-60, 193
Gold Standard
Concepts




27-29, 71

Concepts vs. practices
For Germany

Page
29-32, 33, 43, 106
61, 62

Gold Standard Act of 1925

218

Goldenweiser, E. A

10, 16

Grigg, Sir James

73

Hamlin, Charles S

10, 16, 157

Harding, Warren

36,223

Harrison, George L.
Career
Papers
Relations with central bankers

11
16
151,
154, 157, 174
Relations with Federal Reserve Board.... 151
Trips abroad
,
174, 210
Views on
British fiscal policy
211
Credit to Bank of England
163, 206
Credit to Gold Discount Bank
198
Credit to Reichsbank
191, 194, 197
Currency support operations
161, 162,
175, 186,215,217
Discount rate policy
153, 154, 155, 158
German standstill efforts
194
Loan to British government
204,
210,212
Loan to Germany
198
Open market purchases
172
Stock market boom
154
Harvey, Sir Ernest M
11, 102,
206,210,217
Hatry Scandal
159, 160, 225
Henderson, Arthur
Herriot, Edouard
Hoover, Herbert
Hoover Moratorium
Hughes, Charles E

11, 201, 210, 211
11, 51, 57, 68
11, 174, 178,
184, 199, 200, 226
184, 193,
197, 198, 226
11, 47, 55, 223

Interest Rates

23, 85-89,
99, 101, 121, 128, 157
See also Discount rates under Central
Banks, and particular central banks.

231

Page
179

International Corporation
Jay, Pierre

11, 69

Kemmerer, Edwin W

11, 61, 63

Keynes, John Maynard

11, 73

Kindersley, Sir Robert

11, 47,
66, 179, 205
177

Kreuger and Toll
Lacour-Gayet, Robert

11,
138n,209,212

Lamont, Thomas W

11, 16, 48,
52, 54, 57, 67

Laval, Pierre

11,199

League of Nations
36, 37, 41, 45, 222
See also Salter, Sir Arthur.
Lee-Higginson

177

Loans
See Capital Issues, Foreign, and
under particular central banks.
Logan, James A

11, 57

London Conference of 1924
See under Dawes Plan.
London Conference of 1931
Lubbock, Sir Cecil
Luther, Hans
MacDonald, J. Ramsey
Macmillan Committee
Macmillan Report
May Report

200,
203, 227
11, 79

11, 16, 44,
67, 177, 194, 197
11, 52, 53,
57,199,209,211
32, 39,
74, 102, 143
140, 141,
143,185,202,226
12, 204, 207, 227

McDougal, James B

12, 126

McGarrah, Gates W

12, 186, 187, 198

McKenna, Reginald
Mellon, Andrew W

232




12, 47, 73,143
12,28, 75,
77, 178,224

Page
Meyer, Eugene
12, 192
Mills, Ogden L
12, 123
Monetary Policy
See under particular banks.
Moreau, Emile
Career
12
Diary
15
Gold and dollar policy
117
National interest
30, 115
Relations with central bankers
115
Views on
British policy
.. 118
Foreign balances
112, 119
Foreign exchange operations
138
Franc stabilization rate
112
Moret, Clement
12, 178, 180, 195, 212
Morgan Grenfell & Co
48
Morgan, Harjes & Co
48
Morgan, J. P
12, 48, 54, 67, 68, 180
Morgan, J. P., & Co.
Credit to British Treasury
82
Role in Dawes Plan
47
Views on
Agent General and Transfer
Committee
57
British fiscal policy
211
Dawes Loan
50, 53
French loan
68
Loan to Great Britain
163,213
See also partners, Lamont, Thomas W.;
Morgan, J. P.; and Morrow, Dwight W.
Morrow, Dwight W
12, 55
National Bureau of Economic
Research
126
Niemeyer, Sir Otto E
12, 79
Norman, Montagu C.
Bank policy
36
Clay's biography
16
Career
12
Collapse
204
Gold standard practices
36, 100, 103
Relations with central bankers
41, 58, 75,
84,115,134,151

Page
Relations with Churchill
102, 143
Relations with J. P. Morgan
48
Role in Dawes Plan
47, 59, 62, 68
Trip to Canada
210
Trip to Paris
117
Trips to United States
76, 77, 79,
123, 152, 174, 181
Views on
Bank for International Settlements'
functions
179
Bank of England's difficulties
99, 117,
142, 150, 160, 164,202
British fiscal policy
43, 211
Central bankers' meetings
32, 36, 123
Credits to the Bank of England
77,
83, 163,205
Credit to Reichsbank
58, 191, 194, 196
Currency support operations
58, 131,
161, 162, 175,216
Dawes Plan
53-57, 65, 67
Discount rates
86, 88, 152, 203
Gold standard
32, 34
International banking authority
144
Loan to Austria
188
Loan to Germany
197, 199, 200
Loan to Great Britain
162, 178,
179, 181,204,205
London as a financial center
34, 62
National interest
30, 32, 73
Reichsbank's reserves
62, 65,
66,114,165
Reparations and war debts
34, 196
Sterling stabilization
73, 74, 76, 78
Open Market Investment Committee
National interest
125
Records
16
Securities holdings
43, 83, 85,
88, 153, 161, 169
Platt, Edmund
Poincare, Raymond

12, 127
12, 51,
110,118, 119
24, 35, 75, 90, 91, 96
12, 165, 178

Prices
Quesnay, Pierre
Reichsbank
Contribution of Dawes Loan to




49

Page
58, 192, 193
58, 174
127, 148-50,
190,199,219
114, 132
114,136,165
46, 66, 95,
114, 136,165
International reserves
60-66, 109, 115,
136, 149, 150, 165, 166, 168, 174, 191, 193
Open market operations
43
Sterling balances
113,114
See also Schacht, Hjalmar H. G.,
and Luther, Hans.
Reparation Commission
47, 52, 53,
57, 69, 222, 226
Reparations
German payments
21
German views on
21, 55, 145,
172, 177, 183
Settlement
144
See also Dawes Plan and Young Plan.
Reparations and War Debts
20, 47,
145, 150, 180
Credit to
Currency support operations
Discount rates
166, 170,
Gold buying price
Gold policy
108,
Gold reserves

Rist, Charles A
Sackett, Frederic M
Salter, Sir Arthur
Schacht, Hjalmar H. G.
Autobiography
Career
At central bankers' meeting
Relations with central bankers
Role in Dawes Plan
Views on
Currency support operations
Dawes Loan
French short-term funds in
Germany
German foreign borrowing
Gold Discount Bank
Gold standard
Reichsbank's reserves

12, 123, 138
12, 177
13, 38-40
16
13
123
58,
112, 115,134n
47
59
54, 67
165
110
60
63
63, 66, 109

233

Siepmann, Harry A

Page
13, 140, 212

Snowden, Philip
Career
13
Views on
Dawes Plan
53
Fiscal policy
201,204,209,211,216
French withdrawals
215
Gold reserves
217
Reparations and war debts
196
South Africa

80

Sprague, Oliver M. W

13, 176

Sproul, Allan

13, 176

Stewart, Walter W

13, 152

Strong, Benjamin
Calling of central bankers' meeting
123
Career
13
Chandler's biography of
16
Death
151
Key currency approach
40
Papers
16
Relations with central bankers
41,
84, 115,134
Role in Dawes Plan
48
Trips abroad
64, 75, 76n,
115,116, 140, 153
Views on
Bank of England's dollar balances
100
British fiscal policy
140
Capital movements
76
Central bankers' meeting
39, 107
Cooperation among central banks
31,
33,40,41,72, 116, 144
Credits to Britain
77, 85
Currency support operations
59, 123,
125, 131, 135
Dawes Plan
57, 65
Discount rate
31, 88, 98, 127, 153
French foreign exchange
operations
135
Genoa proposals
36-40
Gold exchange standard
38, 135
Gold standard
28, 33, 37, 39, 64, 135
London as a financial center
65

234




Page
National interest
30, 31, 98, 106, 144
Prices
37,75
Sterling stabilization
72, 74, 79, 80
Stock market speculation
127
Sweden
Theunis, Georges

80
13, 68

United States Government
Isolationist policy
34
On government loans
56, 198, 217
On reparations and war debts
20, 47,
147,150, 191,217
See also Coolidge, Calvin; Hoover,
Herbert; Hughes, Charles E.; and
Mellon, Andrew W.
Vissering, Gerard

13, 59

Warburg, Paul M

13,61

Wiggin, Albert
Williams, John H

201
40

War Debts
Payments to United States

21,
90, 222-24
See also Hoover Moratorium, Reparations
and War Debts, Young Plan,
and particular governments.
Young, Owen D.
Career
Papers
Role in Dawes Plan
Role in Young Plan
Views on
Dawes Loan
Reichsbank's reserves

13
16, 50
47, 57
146
69
63

Young Plan
Annuity terms
145, 146, 165
Bank for International Settlements
147,171
Exchange transfer protection
146
Hague conferences
167, 225, 226
International loan
145, 171
Relief provisions
184
Young, Roy A

13, 156, 157


Federal Reserve Bank of St. Louis, One Federal Reserve Bank Plaza, St. Louis, MO 63102