By Eric Vandenbroeck
Bankrupting The Enemy
While previously we mentioned that Japanese textbooks
generally referred to a U.S. financial siege of Japan "before" Pearl Harbor,
US documents released in the context of the Holocaust Era Records Act between 8
and 10 years ago show among others that from 1937 to 1940 a dozen experts in
U.S. financial agencies analyzed Japan's balance of payments, gold production
and reserves, scrap gold collection, liquidation of foreign investments, and
other financial data. They also predicted when Japan would be bankrupt and have
to stop the war in China, always six to eighteen months in the future. It was a
comforting thought to policy makers, but the analyses were wrong. From
1938 to the summer of 1940 the Bank of Japan secretly accumulated $160 million
in the New York agency of the Yokohama Specie Bank. It began with funds removed
from London. It was a sum equal to three years' of oil purchases from the U.S.
The YSB did not report the deposits, as required by law. Bank examiners
discovered the fraud in August 1940. Japan raced to withdraw the money during
the first half of 1941. The fraud accelerated thinking in Washington toward a
dollar freeze, instead of commodity embargos, as the most deadly sanction. The
freeze order was drafted in March 1941. As is well known, It was imposed on 26
July 1941 when Japan occupied southern Indochina. After the freeze, Japanese
diplomats and agents proposed many ideas to unfreeze dollars in order to reopen
trade. Dean Acheson, the effective manager of the freeze policy, rejected them
all. In August 1941 the Japanese government, through Mitsui, made an
extraordinary offer to barter $60 million of silk for $60 million of US
commodities, mainly oil. Barter would not require unfreezing dollars, they
thought. Strangely, they chose as spokesman a Roosevelt-hating
lawyer named Raoul Desvernine.
Acheson and Vice President Wallace rejected the scheme on 15 September, about
the time the Japanese government was deciding for war. Furthermore features of
the prewar situation from records that were not classified but that are omitted
from other history books, thus there are a large number of vulnerability
studies, of Japan's foreign trade written in April 1941 by committees of trade
experts of U.S. agencies under direction of the Export Control Administration.
Reviewing the oil and tanker situation in both Japan and the U.S. for example
one will here find that FDR blamed an imaginary shortage of gasoline in America
as a reason for the freeze of Japan. There was no shortage despite the loan of
20% of US tankers to Britain. Two-thirds of Japan's dollar earnings were due to
exports of raw silk to America. (When war began in Europe all other currencies
became blocked and inconvertible.) The Great Depression and rayon substitution
destroyed the silk dress industry. After 1930 nearly all Japan's silk went for American
women's stockings, which was a strong market despite the Depression. On 15 May
1939, however, DuPont introduced nylon stockings at the New York Worlds Fair. They were a great success. By 1941 nylons
gained 30% of the market, and were on track to 100% in 1943 if no war. The
market loss of $100 million per year would have been a disaster for Japan if it
had not gone to war. Yet what the following case study furthermore shows is
that Roosevelt's idea of a financial siege of Japan in fact backfired by exacerbating
rather than defusing Japan's aggression. And for sure the attack on Pearl
Harbor was not the result of a deliberate Roosevelt strategy (as right-wing
conspiracy theorists claim), but a Roosevelt miscalculation.
U.S.Exports to Japan 1939 (in Millions of Dollars)
U.S.Exports to Japan 1940 (in Millions of Dollars)
U.S.Exports to Japan 1941 (in Millions of Dollars)
The American
perception of Japan's economic and financial vulnerability dated back to a time
thirty-five years before Pearl Harbor. President Theodore Roosevelt grew
concerned after the victory over Russia in 1905 that Japan would seek to
dominate China in contravention of the U.S. Open Door policy, which championed
independence and free trade for China. Japan would perceive that policy, and
U.S. bases in the Philippines and Hawaii, as barriers to building an empire.
Roosevelt asked the U.S. Navy for a plan to fight Japan, if and when necessary.
The result, War Plan Orange, was fundamentally an economic strategy in both
origins and outcome. (Japan was code-named Orange, the United States Blue.) The
godfathers of the plan, Admirals George Dewey and Alfred Thayer Mahan, had
served as young officers enforcing the Union's Anaconda Plan against the
Confederacy, an "island" vulnerable to economic blockade. They and
later disciples in the War Plans Division of the Navy demonstrated a fierce
mindset favoring vigorous action, a mindset echoed by civilian bureaucrats who
advocated a nonviolent economic and financial "war" against Japan in
the crisis years before Pearl Harbor.
While U.S. military
planners assumed Japan's war aims would be limited --a surprise attack, victory
in naval battle, and a negotiated peace ceding dominance of East Asia--their
aim was a crushing defeat of the enemy, an aim demanded by an aroused public.
They understood that Japan, an island nation poor in natural resources,
depended on overseas trade for the sinews of war and its very economic life.
The Japanese Empire produced food enough, but industrialization and conquests
led to voracious needs of metals and fuels. The planners designed a strategy of
siege. After initial losses, Blue forces would fight back island by island,
sink the enemy fleet, seize bases near Japan, starve it of vital imports, and
ultimately force it to capitulate. Japan's financial destitution would be
ensured by "coercive pressure" on world lenders to deny funds such as
Wall Street provided during the Russo-Japanese War. Plans rang with confidence
that the United States could enforce "final and complete commercial
isolation" (1906), leading to "eventual impoverishment and
exhaustion" (1911) and "in the end ... economic ruin" (1920). As
air power came of age, bombing of industry and transportation intensified the
siege plan. In 1941 Plan Orange morphed into global Plan Rainbow Five, and in
1942-45 it was executed in most major respects. The Pacific war culminated in
unconditional surrender after devastation of Japan's economy, including, at the
end, the deployment of atomic bombs.
In the 1930s
peaceable internationalist governments in Tokyo gave way to military-dominated
regimes. The anticipated violations of the Open Door unfolded in the invasion
of China and designs against colonies of the Western powers. The helplessness
of Japan, if isolated economically and financially, evolved into an axiom at a
time when the U.S. government was averse to fighting a war. When national
policy to deter Japanese aggression took root, the United States gradually
deployed its vast economic and financial powers to strangle Japan by means
other than ships and bombs. It was a Plan Orange strategy in peacetime. The
story now turns to the U.S. strategy of achieving the nation's foreign policy
aims, without combat, by bankrupting Japan.
When Japan invaded
China on 7 July 1937, U.S. government financial experts reckoned the aggressor
could not wage a long war because it lacked hard currency to purchase essential
commodities abroad. As Herbert Feis, the economic adviser of the State Department,
wrote, "Warfare requires many vital raw materials which Japan does not
possess at all or in sufficient quantities, and which must by
purchased with foreign exchange." Therefore, the "ability of Japan to
carry on a protracted war depends ultimately upon her actual and potential
foreign exchange resources." 1
And as early as 1935,
it sought to regulate its position as the world's most important supplier of
materials for war, it was in the summer of 1941 the United States decided to
deploy its most powerful economic sanction against Japan, dollar freezing. The
first steps, taken at a time when Japan was not yet an aggressor in China, were
aimed at keeping the United States from being drawn into another foreign war.
Later steps were taken by so-called voluntary means and by executive orders to
ensure that the nation's exports did not support Japanese aggression in China.
In fact sanctioning to impose a nation's will on others in peacetime was also,
not a novel concept, as it was based on international law governing trade
relations among sovereign states laid down by the European jurists Hugo Grotius
and Emeric de Vattel in the seventeenth and eighteenth centuries.
Thus in the autumn of
1937 Franklin Delano Roosevelt brooded about deterring foreign military
dictatorships from attacking peaceful nations. On 5 October, thirteen weeks
after Japan invaded China, the president delivered his famous
"quarantine" speech. Likening the spreading aggressions to an
epidemic disease, he suggested that law-abiding countries ought to quarantine
the aggressors. When pressed by reporters, he denied that he meant economic
sanctions, calling sanctions "a terrible word to use." "There are,"
he said, "a lot of methods in the world that have never been tried
yet."! Roosevelt was not sure what he meant until December, after Japanese
bombers sank a U.S. gunboat in China. He turned to his energetic secretary of
the treasury, Henry Morgenthau Jr., for a "modern" weapon to wield
against Japan. Treasury experts unearthed the perfect device: a relic of the
First World War known as Section 5(b) of the Trading with the Enemy Act (TWEA),
a single paragraph that empowered the president to paralyze dollars owned by
foreign countries, whether enemy or not. Denial of U.S. dollars, a key reserve
currency of the world and indispensable to Japan for waging war, could dissuade
Japan from belligerence. That surviving section of the act had arisen from an
obscure spat in 1917 between government agencies that foreshadowed the
bureaucratic grasping for power in Washington during 1937-41 as the United
States groped toward invoking its great financial powers to render Japan
effectively bankrupt in the world.
And finally the first
months of 1941 then, marked a turning point in the will of the United States to
advance from a patchwork of export restrictions to full-blooded financial
warfare against Japan. A spurt of work from January through March established the
nature of the financial punishment it would mete out when the time came. Above
all, that the levers of control would be manipulated not by learned economists
and banking technicians, nor by moderate diplomats seeking bargaining leverage,
but by truculent lawyers determined to show Japan no mercy.
In common English
usage, "bankruptcy" is a synonym for "impoverishment."
Japan was cast into international bankruptcy, a condition of absolute
illiquidity, by the U.S. financial weapon. The choke hold of the relentless
freeze rendered its dollars and gold worthless for national survival. It was a
strange sort of bankruptcy. Japan's reserve of gold and dollars exceeded $200
million in late 1941, enough to buy, for example, four years of U.S. oil at
pre-freeze shipment rates, yet it was rendered useless. Thereafter Japan piled
$60 million of new gold and $15 million of silver into the useless reserve
every year until it suspended mining of precious metals in 1944. Japan had
invested in gold mines, collected gold ornaments, and nurtured dollar-earning
exports.1 It had husbanded its reserve for future purchases of resources for
its war economy, inflicting "the curse of gold" on its people by a
"wholesale attack on the standard of living." At the end of the war,
although the economy was in a shambles, the government of Japan was awash in
gold amounting to twice its hoard in 1941. After the surrender the gold and
silver in the Bank of Japan and other government vaults was sequestered by
MacArthur's occupation forces. As to Japan's frozen dollars, they were never returned.2
A contemporary
journalist summed it up. He could not have known that for four years, U.S.
Treasury and Federal Reserve analysts had predicted Japan would soon exhaust
its assets and necessarily abandon its aggressive policies. However, he wrote,
"Japanese leaders exerted every possible effort to avoid this outcome.
They succeeded .... Gold production was stimulated A vast foreign exchange
reserve was maintained." Although U.S. forecasts of empty vaults proved
false, Japan was plunged into the international bankruptcy they predicted
because of a stroke of a pen in Washington.3 Two views, one American and one
Japanese, illustrate the attitudes about Japan's bankruptcy on the eve of war.
When Dean Acheson
arrived at the State Department in January 1941 he rediscovered the prodigious
powers of Section 5(b) of the 1917 Trading with the Enemy Act. He and
colleagues of like mind promoted its deployment against Japan, then twisted a
cautious squeeze designed "to bring Japan to its senses, not its
knees" into strangulation. Acheson, an officer of the department charged
with peaceful solutions through diplomacy, boasted to Cordell Hull on 22
November 1941 that financial crippling had proven far more devastating to Japan
than embargoes. The freeze administrators thwarted Japan from removing its
dollars from U.S. control as it had been doing for a year. Their actions
slashed U.S. exports to zero despite Japan's valid export licenses for oil, and
other licenses it would have been entitled to for cotton, lumber, and
foodstuffs. Nor could Japan pay the mineral-rich nations of North and South
America or the Dutch Indies, which demanded dollars, while the sterling bloc
joined in the freeze. U.S. markets abruptly closed to Japan. Washington refused
to allow Japanese trading companies to receive dollars, even if paid into their
blocked accounts, hastening the ascendancy of nylon, which devastated silk
farmers and demolished Japan's largest renewable flow of dollars.4
U.S.Forecasts of Japan’s International “Bankrupty,”
1937-1941
In Japanese eyes the bankruptcy
was a lethal threat, an assault on the nation’s very existence. After the war,
Koichi Kido, lord keeper of the privy seal and adviser to Emperor Hirohito,
delivered an eloquent statement through his American defense counsel, William
Logan Jr., before the International Military Tribunal for the Far East. (Kido
was found guilty of war crimes and sentenced to life imprisonment but was
released in 1955.) Kido styled his defense “Japan Was Provoked into a War of
Self-Defense.” Allied charges of war crimes defined aggression as “a first or
unprovoked attack or act of hostility.” Kido argued that strangling an island
nation dependent on foreign resources was a method of warfare more drastic than
physical force because it aimed at undermining national morale and the
well-being of the entire population through starvation. A nation, he concluded,
had the right to decide when economic and financial blockade was an act of war
that placed its survival in jeopardy. Kido thereby harked back to the dawn of
international law three centuries earlier when jurists held that refusal to
sell to another nation might well be a valid casus belli under extraordinary
circumstances such as starvation.The defense added,
“We know of no parallel case in history where an eco
nomic blockade … was enforced on such a vast scale with such deliberate,
premeditated, and coordinated precision …. Responsible leaders at that time
sincerely and honestly believe[d] that Japan’s national existence was at
stake.” Because sanctions “threatened Japan’s very existence and if continued
would have destroyed her,” the “first blow was not struck at Pearl Harbor.”
Indeed, Lojan continued, the “Pacific War was not a war of aggression by Japan.
It was a war of self defense and self
preservation.”5
The OSS Document
Unfortunately for
Japan, its leaders chose a war that brought upon it far more economic
devastation than any sanctions, along with great loss of life and untold
misery. Although struggling along under bankruptcy without going to war was a
dreary prospect, a third course was open to Japan: renouncing imperial
aggression in return for thawing of the freeze. One may wonder, what if Japan
had endured the freeze long enough to ascertain that Germany could not win and
had then abandoned the Axis, perhaps even joined the Allied side as it had in
1914? It would have prospered mightily by selling ships, machinery, and other
goods to the Allies. It would have emerged after the war as the strongest
regional power, with a world-class navy, an overflowing treasury, and a zeal
for industrial modernization, just as colonial empires in Asia were crumbling.
It might have shored up China against communism. A cooperative Japanese
commercial "empire" in East Asia, economically buoyant and trading
internationally on a grand scale a generation sooner, could have changed the
course of history in the twentieth century and beyond.
Arthur B. Hersey titled his study for the Office of
Strategic Services and the Department of State "The Place of Foreign Trade
in the Japanese Economy; an analysis of the external trade of Japan proper
between 1930 and 1943 focusing on possible or probable postwar development."
He comprehensively analyzed Japan's external trade with both the yen bloc and
the rest of the world in three representative prewar years: 1930, 1936, and, to
a lesser extent, 1938.6 His goal was to outline a possible range of conditions
of Japan's economy about five years after the end of World War II to assist
U.S. planners contemplating post-surrender and occupation policies. The
methodology was complex. To link actual past to hypothetical future years
Hersey converted physical units (pounds, bales, square yards, calories, etc.)
to a common denominator of "constant yen," a proxy for physical units
that also allowed him to adjust erratic prices of internationally traded goods
into more comparable units.7
During the war many
economists expected a return of the global Depression after a brief postwar
boom. Hersey believed Japan's future in international trade to be especially
bleak. Its appetite for imports of minerals, industrial crops, machinery, and
even foodstuffs was almost unbounded. But its capacity to import would be
limited to the hard currency it could earn from exporting goods and services
and from gold mining. (He considered foreign loans unlikely.) In the 1930s
Japanese exports had expanded rapidly, but the benefits to the people had been
circumscribed for several reasons. A rising share of exports went to the yen
bloc, which could neither pay in hard currency nor deliver the most needed
commodities. While foreign countries erected barriers against Japanese goods,
the terms of trade (relative world prices) worsened after 1930. No foreign
loans were available due to disorganized financial markets in America and
Europe and active discouragement of lenders by those governments because of
Japan's aggressions.
"The core of the
analysis," Hersey wrote, was a classification of Japan's imports
(typically 90 percent raw and semiprocessed materials
and foods and 10 percent manufactures) into two categories: commodities
required by factories that manufactured products for export and commodities for
final consumption within Japan. The latter, labeled "retained
imports," comprising 59 to 68 percent of all imports in the 1930s,
contributed directly to the standard of living. Another 25 percent of imports
were materials for processing and resale abroad, primarily raw cotton for
textiles, wood pulp and salt for rayon, and metals, chemicals, and fuels for
other manufactures. A final 8 percent of imports were offsets to exports of a
similar nature, swapped, in effect, because Japan both bought and sold in
various grades and processed forms, wheat, sugar, fish, coal, and fertilizer.8
Japan's greatest dilemma in the 1930s, Hersey believed, had been deterioration
of the "barter terms of trade," that is, weak export prices and high
import prices. Japan had to run harder to stay in the same place
internationally. For consistency he recast the data into indexes of
"constant yen" at 1930 terms of trade. (He also calculated 1936 and
1938 terms of trade although they were of less relevance to his conclusions.)
Hersey selected 1930 as a "best case" year, similar to the relatively
prosperous 1920s, and the last equilibrium year of Japan's international trade
before the turmoil of world depression, yen devaluation, the Manchurian
adventure, and foreign trade discrimination. In 1930 Japan's upscale products
enjoyed high prices abroad, notably raw silk and silk fabrics, premium
seafoods, fine pottery, and other consumer luxuries, while prices of raw cotton
and most other imported agricultural and forestry products were low. (Japan did
not yet import oil, metals, or minerals on a large scale in 1930, and not much
machinery).9
A terms-of-trade
index is not the same as the familiar domestic price index. It is a ratio of
relative prices, that is, an export price index divided by an import price
index. Hersey calculated data for twenty internationally traded product classes
that he aggregated into eight groups: food; fertilizer and fodder; coal and
petroleum; metals and minerals; cotton, wool, and pulp for rayon; lumber and
paper pulp; and manufactured goods. He assigned to the terms of trade in 1930
an arbitrary index number of 100. The index for any other year, actual or
predicted, was that year's export price index divided by its import price
index. A resulting index above 100 meant a favorable trend for Japan, and vice
versa for numbers below 100. Although any prediction was "pure
guesswork," Hersey admitted, a postwar Japan enjoying 1930 terms of trade
could fare adequately in the world, though not richly. "It is
doubtful," he opined, "whether Japan's terms of trade will under any
circumstances be more favorable than they were in the 1920s and 1930."
Hersey examined the
improvements of the Japanese standard of living before the war. Economists had
been awed by a surge of retained imports-79 percent higher in 1936 and 86
percent higher in 1938 compared with 1930but the benefit to ordinary Japanese
families was somewhat illusory. Yen devaluation, worsening terms of trade, and
a massive switch to importing and stockpiling of industrial and strategic goods
left the rise of retained imports for the benefit of the public at only 14
percent, barely more than population growth of 9 and 12 percent respectively
since 1930. Yet the Japanese standard of living had undeniably improved, by
about 10 percent per capita. Food consumption per capita was thought to be
unchanged; the rising population was fed from rising domestic farm output
through intensive fertilization. Gains in nonfood goods and services ranged
from 20 percent to more than 30 percent per capita. As with food, the gains
were mostly achieved by surges in production from domestic resources, notably
chemicals, electrical energy, and paper, and by the effect of rayon pulp (10
percent of textiles cost) substituting for raw cotton (50 percent of textile
cost). The experience implied that if postwar Japan could import consumer needs
for its populace at the 1930 rate in real terms per capita, reduction of the
standard of living would be tolerable, although disappointing for a population
used to improving conditions.10
For his "highly
tentative" postwar models of trade and living standards Hersey adopted the
hypothetical year" 1950" to represent a date a few years after the
war when physical reconstruction would be largely completed, production of most
domestic-sourced goods recovered, and crop yields normal. Population growth was
a crucial assumption. Japan's population had risen steadily at about 1 million
per year, from 64.4 million in 1930 to 72 million in 1938. Expecting
continuation of that rate, Hersey expected a population increase to 81 or 82
million in "1950," after adjusting for war casualties and
repatriation of Japanese émigrés from Asia.11 For a country that historically
found difficulty in feeding itself, millions of extra mouths would intensify the
dilemma of maintaining living standards in the face of weakened exports. Three
adjustments have been made here to adapt Hersey's data from "1950" to
represent Japan in, say, 1942, under a freeze but not at war with the Allies.
First, the population differences between the periods of seven to eight million
people are neutralized by converting trade to per capita values. Second, yen
are converted to dollars at appropriate exchange rates. Third, his eight
commodity groups are simplified into two: consumer commodities and other.
Japan's postwar
future was clouded by an anticipated vicious cycle of trade: uncertain markets,
adverse prices, and technological changes (notably the substitution of nylon,
reducing raw silk exports by an assumed 50 percent)12 resulting in a shortage
of hard currency to buy raw materials for factories that produced for export.
The uncertainties were profound. Rather than guess at world appetite for
Japan's specialized goods, Hersey found it easier and surer to calculate
imports essential for survival of an impoverished populace. He therefore set
imports as the independent variable and assumed two levels of
"retained" and other imports. He then "reverse engineered"
his models to determine the exports necessary to fund the purchasing abroad.
Japan's exporting capability became the dependent variable. Hersey developed
two scenarios of the Japanese standard of living in "1950" by
arbitrarily assuming two levels of nutrition, expressed as daily calories per
person, which set an upper limit on non-food imports. Case A assumed the 2,250
calories prevailing in 1930, which had not increased much if any in the
following ten years. Assuming, however, that Japan's capacity to harvest crops
and fish had topped out by 1941, a larger share of its limited postwar earnings
would necessarily have to pay for imported food, fertilizer, and fishing boat
fuel. Imports of materials for clothing, shelter, and infrastructure would have
to be severely constrained by government priority rules, leaving little or
nothing for other consumer goods such as foodstuff varieties. The procedure
resulted in reduced postwar living standards of 25 to 33 percent depending on
the details assumed.13
Case B envisioned a
horrendous outcome for the Japanese people because of an exporting capacity so
enfeebled that not even basic nourishment and health could be maintained.
Hersey arbitrarily assumed a 20 percent reduction in nutrition below Case A, to
1,800 calories per person per day. Food and fertilizer needs would overwhelm
other import priorities. Only minor imports could be financed for other
consumer needs and urgent infrastructure. Retained imports per capita would
slump 67 percent below 1930.14 Hersey also calculated terms of trade for 1936,
the last peacetime year and a "worst case" year for Japan. Raw silk
prices had fallen disastrously. Textiles and other wares were restricted by
U.S. tariffs and quotas and by British imperial preference. Although exports of
chemicals and mechanical products-bicycles, sewing machines, industrial
machinery-held up better, Japan mostly sold them to the empire for yen.
Meanwhile, prices of
imported commodities were propped up by dominant suppliers such as U.S.
government supports of cotton. (Strategic metals and fuels remained relatively
cheap but were minor items of import before the war in China.) Relative to a
1930 index of 100, export prices in 1936 dropped to 95 whereas import prices
soared to 129. Japan then had to sell 33 percent more goods to buy the same
basket of imports. Because Hersey assumed the value of other imports as equal
to Japan's residual buying power after meeting food and fertilizer needs, the
large difference between 1930 and 1936 terms of trade dictated a necessity of
much larger exports, but did not alter his Case A and B models of
"1950." For example, Case A, calibrated to the 1936 terms of trade,
required 55 percent more exports versus the 1930 terms of trade model, $1.62
billion versus $1.05 billion, to achieve the same standard of living
established by Hersey's assumptions. (Hersey did not itemize exports in detail
as he did for imports because of extreme uncertainty over the products Japan
could sell, and to which countries, after the war.) Despite Japanese censorship
of data from 1936 onward, Hersey calculated a somewhat improved 1938 terms of
trade index but did not rely on it because distortions caused by the war in
China, commodity stockpiling, and a renewed U.S. depression that lowered the
cost of Japanese imports rendered it irrelevant to his vision of
"1950."15
As indicated above,
Japan's de facto bankruptcy proved a crucial factor in the failure of
negotiations for a peaceful settlement with the United States. The diplomatic
maneuverings of 1940-41 have been exhaustively described in documents, memoirs,
diaries, interviews, postwar investigations, and war crimes trials.16 Thus we
briefly summarize the events, focusing on the significance of the dollar
freeze. U.S. resentment against Japanese aggressions began with the seizure of
Manchuria in 1931 and accelerated when Japan assaulted China in 1937. The'
country initially reacted with diplomatic scolding’s, aid to China, and
embargoing exports of a few arms-related products. In 1940 tensions grew acute
when Japan signed the Tripartite Pact with Germany and Italy whereby the three
powers pledged to assist each other in wars, under certain circumstances. The
United States, inching toward war in the Atlantic through pro-British policies,
grew concerned that it might have to fight Japan as well. Negotiations for a settlement
of tensions began in earnest in April 1941. All discussions were conducted in
Washington between Ambassador Kichisaburo Nomura
(assisted after 15 November by special envoy Saburo Kurusu) and Secretary of
State Cordell Hull. The Japanese diplomats also met directly with President
Roosevelt, and occasionally with civilian and naval officials Nomura new
personally.
Other U.S. officials
played relatively minor roles.17 Hull advanced four "principles" for
Asia: respect for the territory and sovereignty of all nations, noninterference
in their internal affairs, equal commercial opportunity, and maintenance of the
status quo in the Pacific-the principles established by the Nine Power Treaty
of 1922. For Japan the main stumbling block was surrendering its decade of
conquests by withdrawing from Indochina and China, perhaps even from Manchuria.
In Japanese eyes a retreat would mean giving up any possibility of gain from a
war that had cost two hundred thousand dead soldiers, required huge outlays of
national treasure, and caused economic hardships for its people. The United
States further demanded assurances that Japan would renounce the Tripartite
Pact, or at least refrain from fighting it as an ally of Hitler. To prod Japan,
Washington embarked on three programs: arms and financial aid to China, a
buildup of forces at Pearl Harbor and in the Philippines, and barring exports
of commodities needed for its own defense.
The Japanese position
was, simply, resistance to Hull's proposals: no U.S. interference in
China-Japan affairs, no military withdrawals from occupied territories,
maintaining ties with Germany, and continuing trade with the United States. In
1941 events in Europe emboldened Japanese leaders. Hitler's attack on Russia on
22 June quelled the army's fears of a Soviet attack on the empire, and it
joined the navy in favoring a war to seize the resources of western colonies in
Asia. On 24 July, Japan, having coerced Vichy France, occupied southern
Indochina, triggering the U.S. freezing orders two days later and those of the
Allies soon after. After 26 July 1941 Japan's priority shifted to demanding an
end of the dollar freeze, or at least an easing so that deliveries of oil and
perhaps other strategic commodities might resume. At first Tokyo phrased the
aim in generalities while its representatives in the United States searched for
loopholes. In August banking and consular officials petitioned for financial licenses
to pay for two shiploads of oil and probed the possibilities of paying with
dollars or gold held outside the frozen accounts. They were rebuffed at every
turn by the Foreign Funds Control Committee dominated by Dean Acheson.
Early in August Prime
Minister Prince Fumimaro Konoe launched an initiative
to meet with President Roosevelt personally, perhaps in Hawaii or Alaska, in
what later generations would call a summit meeting. To placate the generals and
admirals, Foreign Minister Teijiro Toyoda drafted
demands that the United States halt reinforcement of the Southwest Pacific,
mediate a peace settlement in China (a euphemism for abandoning aid to Chiang Kai-Shek), and restore normal commercial
relations (a euphemism for ending the freeze). In return Japan offered not to
advance beyond Indochina and to withdraw troops from China when the war ended
at some vague future date. FDR was intrigued but the State Department deemed
the tradeoffs unacceptable, especially because Japan refused to start
evacuating promptly. The United States declined the summit offer.
Japanese military and
naval leaders moved forward with plans to launch a war before the year was out.
On 6 September 1941 an imperial conference agreed to make a decision during the
first ten days of October about war against the United States, Britain, and the
Dutch Indies (a deadline gradually moved back to 29 November) unless Japan's
demands were met.3 On 18 September Acheson disclosed that the United States had
rejected Japan's last ditch barter scheme of oil for silk. Mobilizing for an
attack began in earnest in Tokyo in the second half of the month. Nevertheless,
Toyoda wished to test other avenues of negotiation. The deadlock between the
war hawks on one hand and Konoe and Toyoda, who favored some troop withdrawals,
on the other hand, led to the fall of Konoe and his replacement as prime
minister by General Tōjō Hideki on 17 October. Last-chance diplomacy
passed to a new foreign minister, Shigenori Togo.
Japanese agents had
continued to poke about desultorily for token financial licenses for oil or
minor freeze-evading transactions, without success. On 24 October, however,
Acheson told Counselor Tsutomo Nishiyama that the
looming insolvency of the Yokohama Specie Bank in New York, where Japan had
mobilized its dollars-a bank failure engineered by the U.S. government's
barring the bank from collecting money for silk delivered to the United States
before the freeze-would permanently lock up Japan's main holding of blocked
dollars. It was clear that oil cargoes would never sail. This casting away of
hope immediately preceded Tokyo's decision to demand financial relief, explicit
in time and very substantial in amount, countered by American musings of barter
concessions much below Japanese needs.
As resource
stockpiles dwindled, and with the military's reluctant consent, Shigenori Togo
proposed "Plan A," an offer reciting kinder words about free trade in
China but standing firm on the Axis pact and rejecting troop withdrawals for
twenty-five years. As expected, Hull rebuffed it. Togo followed with "Plan
an interim truce. Japan would evacuate Indochina if the United States kept its
nose out of China, resumed trade promptly at pre-freeze levels, supplied oil in
abundance, and prodded the Dutch to supply more.19 The army insisted on
amending Plan B so that "the United States will promise to supply Japan
with the petroleum it needs." On 14 November the generals defined their
terms:
The United States
must sell a tonnage of oil equivalent to 42 million barrels per year (converted
here at 7 barrels per metric ton), including 10.5 million barrels of avgas, and
ensure another 14 million barrels from the East Indies. If the Dutch did not
agree, Japan would occupy the Indies. If the United States did not comply one
week after signing an agreement, war would begin. Shigenori Togo and
General Tōjō Hideki scaled down the extravagant demands to 28 million
barrels of U.S. oil, still a wildly improbable figure 34 percent greater than
the annual rate of U.S. sales in January-July 1941. The amount was 259 percent
greater than the 7.8 million barrel annual quota based on 1935-36 that
Washington had contemplated in August for possible trade resumption. Avgas had
been effectively embargoed since December 1939. Nomura did not present the
exorbitant demand because Hull's response to Togo's first plan intervened.20
In November special
ambassador Saburo Kurusu arrived to assist Nomura, whose English was not the
best. As presented to Hull on 20 November, Plan B proposed evacuation of
Indochina, American noninterference in China matters, restoring pre-freeze
trade, including an undefined volume of oil, and helping obtain Indies
resources. Considering the plan "preposterous," Hull pondered a
response, urged by the military services to buy time for defense preparations
and by China and England not to go soft.6 On 18 October Hull had mused to Lord
Halifax, the British ambassador, about a minor swap of silk for cotton-not
oil-in exchange for a promise of a status quo in the Pacific. Anxious to avoid
a rupture, the Japanese envoys suggested another humble accommodation: small
quantities of U.S. rice and oil for Japan, far less than its full requirements,
with guarantees that none would go to its armed forces. Hull was willing to
think about it. Roosevelt informed Winston Churchill that the United States
might thaw the freeze slightly on quasi-barter terms, strictly for civilian
goods, for a three-month trial. The United States would license exports of food
products, ships' bunker fuel, pharmaceuticals, raw cotton worth up to six
hundred thousand dollars per month, and some petroleum for civilian needs while
encouraging the Dutch to supply more. Yet the United States would not unfreeze
Japan's dollars. Instead, it would buy Japanese products, two-thirds of which
was to be raw silk-about 5 percent of the pre-freeze rate of silk purchases-just
sufficient to finance the U.S. exports and to service Japanese bonds owned by
Americans.7 But the gesture, overtaken by the onrushing crisis, was never
offered to Japan. For six crucial days in November Hull played with notions of
a modus vivendi ("manner of living"), a standstill of three months
during which Japan would abandon southern Indochina, limit forces in the north,
and commence peace discussions with China. In return the United States would
unfreeze some Japanese dollars and resume some exports, although export
controls in effect "for reasons of national defense" would remain. It
would encourage the British and Dutch to act similarly. Between 20 and 26
November, Hull reviewed a slew of proposals and modifications from administration
officials that watered down his proposal. Acheson's boastful report of the
excellent results of the financial freeze arrived on his desk. By 24 November
Hull's draft conceded a barter-type exchange of raw silk for oil and other
goods, amounts not specified, but no release of blocked dollars.
The eviscerated modus
vivendi was never offered to the Japanese. Allied scouting planes spotted a
troop convoy heading for Thailand and Malaya. Landings there were sure to
provoke war. On 26 November Hull's definitive response, approved by FDR,
retreated all the way back to stiff-necked demands for the four principles and
unlinking from Germany. General Tōjō Hideki deemed deemed it an ultimatum.23 When six Japanese aircraft
carriers sorties from the Kurile Islands, Washington sent a war warning to
Pearl Harbor and other bases. An imperial conference of 1 December gave up on
negotiations and decided irretrievably that the empire would attack. On 4
December the southern invasion force sailed for Malaya from Hainan Island. On
the sixth Roosevelt made a futile personal appeal for peace to Emperor
Hirohito. On 7 December Japan attacked Pearl Harbor. The two nations were at
war.
Japan Sources of Dollars, Actual, 1939-1940, and
Projected, 1941-1943
Japan: Retained Imports per Capita, 1930s and “1950”
Projections
Sources Including For Further Research
The focus of this
case study is the United States' financial and economic sanctions against Japan
before Pearl Harbor, reconstructed primarily from official U.S. sources. Many
histories have been written about the run-up to the Pacific war, largely by
diplomatic historians, understandably in view of the centrality of the
Department of State in U.S.-Japanese negotiations and that department's
voluminous, well-organized files, which were declassified long ago, some as
early as 1943, supplemented by forty volumes of congressional hearings of 1946
about Pearl Harbor and precursor events.35
Not until 1996 did
the National Archives, at the prompting of a U.S. interagency group on Nazi
assets, declassify and make more readily available the worldwide papers of the
Treasury Department's Office of the Assistant Secretary of International
Affairs, established on 25 March 1938 and directed by Harry Dexter White.36
These records contain a trove of U.S. assessments of Japan's financial
problems, and U.S. proposals to exploit them, that have not appeared in other
histories. A similar wealth of information is in the records, first opened to
the public in 1996-97, of the Division of International Finance of the Board of
Governors of the Federal Reserve system, primarily from 1935 to 1955. The
Federal Reserve Bank of New York voluntarily sent to the National Archives
those of its records "that relate to the activity in accounts for foreign
governments" in the same era.37 The files of the U.S. Alien Property
Custodian, which include the 1880-1942 records of Japanese bank branches in the
United States seized in 1941, were closed until fifty years after seizure to
researchers lacking special permission and were inconveniently located until
transferred to the National Archives II in 1995-96 and "bulk
declassified." The records of the Tariff Commission (now the U.S.
International Trade Commission), with a wealth of studies on specific Japanese
products, were open but not properly described and arranged until 1992.4 The
planning records of the Administrator of Export Control, the office that led
the drive for sanctions against Japan during the crucial months of September
1940 to May 1941, were difficult for researchers to use until recently, when
they were rearranged and a finding aid was prepared at the National Archives.
That office was subsumed in September 1941 into the vast wartime bureaucracy of
the Foreign Economic Administration, which in turn was reorganized three or
four times during the war. Its boxed records extend 3,817 cubic feet and weigh
seventy-five tons. A comprehensive catalogue of all international records of
the era, which are mostly located at National Archives II in College Park,
Maryland, was completed in 1999 under the direction of Greg Bradsher and is
available online at http://www.archives.gov/research/ holocaust/finding-aid.
The main Japanese
sources are the excellent historical data published in bilingual tables by the
Japan Statistical Association, and Japanese commercial and diplomatic studies
published in English. Most of Japan's official records of 1931 to 1945 were burned
in the two-week interval between the surrender and the occupation in 1945 in
anticipation of war crimes trials. However, economic information for the last
prewar decade was reconstructed in detail and published by the U.S. Strategic
Bombing Survey and by investigators of the Supreme Commander of the Allied
Powers during the postwar occupation. Japanese financial and trade statistics
are usually presented for fiscal years beginning I April, so that, for example,
"1940" means the twelve months beginning 1 April 1940 and ending 31
March 1941. U.S. statistics are usually given for calendar years, making some
comparisons awkward. Physical trade units are stated here in U.S. measures such
as ounces, tons, or yards, or occasionally in metric measures. Some Japanese
figures have been converted from metric units or the ancient weights and
measures then used in trade.
Money figures are
stated in U.S. dollars, the dominant world currency then and now. The 1935-41
dollar was worth about $10 in 2007 dollars if measured by an average of U.S.
prices of goods, or about $25 if measured by average U.S. wages. In exchange
markets the yen was worth 49 to 50 cents from 1899 until devalued on 14
December 1931. It dipped as low as 20 cents in 1932-33, then stabilized at 28.3
cents until 24 October 1939, when it was devalued to 23.4 cents. There was no
organized exchange market after 25 July 1941; fragmentary trading in China
suggests that in late 1941 the yen's gray market value was much lower, perhaps
II or 12 cents.5 After a devastating wartime and postwar inflation, the yen was
stabilized at 0.28 cents (360 per dollar). It subsequently has risen to almost
I cent (l00 per dollar). The U.S. economy is roughly 150 times larger than in
1935-40 in unadjusted dollars and about 10 times larger adjusted for price
inflation. The Japanese economy is about 500 times larger in unadjusted U.S. dollars
and about 50 times larger adjusted for U.S. inflation. The prewar Japanese
economy was about 8 percent the size of the American. In 2006 it was about 40
percent as large. Japanese foreign trade is now about seven hundred times
greater in nominal value, $1.1 trillion versus $1.5 billion before the war, of
which half was within the "yen bloc." (Both figures are unadjusted
for inflation.) To grasp the relative significance in twenty-first century
terms of $100 million in 1941, a very large fraction of Japan's international
liquidity at the time, the reader may wish to multiply by a factor of one
thousand.
As seen above, the
United States forced Japan into international bankruptcy to deter its
aggression after Washington experts confidently predicted that the war in China
would bankrupt Japan. However, the United States did not know the Japanese
government had a huge cache of dollars fraudulently hidden in New York. Once
discovered, Japan scrambled to extract the money. But, in July 1941 President
Roosevelt invoked a long-forgotten clause of the Trading with the Enemy Act of
1917 to freeze Japan s dollars and forbade it to sell its hoard of gold to the
U.S. Treasury, the only open gold market after 1939. Roosevelt s temporary
gambit to bring Japan to its senses, not its knees, was thwarted, however, by
opportunistic bureaucrats. Dean Acheson, his handpicked administrator, slyly
maneuvered to deny Japan the dollars needed to buy oil and other resources for
war and for economic survival. So it is to the oil issue we now turn, continue...
1 Herbert Feis,
Economic Adviser, "Japan's Ultimate Foreign Exchange Resources," 20
September 1937, Box 21, File Japan Foreign Exchange Position, OASIA.
For updates
click homepage here