After the Second World War, Britain was still a heavily regulated economy. In 1939, the country introduced exchange controls that were not lifted after the war.Footnote 1 Imports were limited by quotas and licences. The state oversaw all foreign exchange transactions. For example, when travelling abroad, the amount individuals could take was limited to £100.Footnote 2 In 1947, the Exchange Control Act formalised capital controls and divided the world into four sterling regions.
The state played a major role in everything from imports to production.Footnote 3 This applied to finance, where capital flows were controlled by the government. The Bank of England oversaw the exchange rate, and only authorised banks were allowed to deal in foreign currencies, within narrow official bands. The United States hoped that removing these controls would be a swift process, and American officials were keen to see markets in Europe develop. The process took much longer than policymakers anticipated.
Between 1948 and 1952, Europe was flooded with $13 billion in grants and loans from the United States.Footnote 4 Britain was the biggest recipient under the Marshall Plan (or European Reconstruction Plan, ERP). A key objective was to encourage convertibility and facilitate trade within Europe and with the United States. In that regard, the policy would boost demand for US products. After the war, as most countries had capital controls in place, trade was complicated.Footnote 5
In 1947, the United States forced the restoration of convertibility on Britain. This led to a currency crisis. The United Kingdom was not yet ready to open up to global markets, and its government was unprepared. Convertibility lasted as long as the government was able to use the Anglo-American loan to defend the pound. It ended as soon as the Bank exhausted the loan proceeds. As the Bank started to use its own gold reserves once the loan ran out, the United Kingdom chose to cease convertibility.
Convertibility meant ‘the freedom for individuals to engage in current account transactions without being subject to exchange controls’.Footnote 6 People could trade freely, transfer remittances and repatriate returns from existing investments. These operations are also known as current account transactions. Capital account convertibility was not included. This would have allowed foreign investment in securities or property, for example.
Britain attempted to restore current account convertibility at the insistence of the United States, but the government abandoned the policy after a mere thirty-seven days. Policymakers, markets and the press drew two conclusions from this: first, sterling was no longer the pre-eminent currency it had been before the war; second, re-establishing an international monetary system with free capital flows would take longer than anticipated a few years earlier at the Bretton Woods conference.
Chapters 2–6 trace the slow unfolding of a new international monetary system from 1947 to the introduction of convertibility in 1958 (current and capital account convertibility this time). The international monetary system was characterised by fixed exchange rates, limited capital mobility and relatively uncoordinated domestic monetary policies. Sterling convertibility was difficult because of the sterling overhang problem, as it was known at the time.
The Sterling Balances Problem
Contemporary debates on the question of sterling focused on the issue of the sterling balances.Footnote 7 These balances were in Britain but held by foreigners living in former colonies and overseas territories. They could not be converted into dollars due to British regulations.Footnote 8 If all the sterling holders abroad converted these balances into dollars, there would be a run on sterling. In 1947, the Bank for International Settlements (BIS) warned, ‘The balances and other short-term sterling assets which accumulated in London on overseas account during the war constitute a major factor in the United Kingdom’s external payments problem.’Footnote 9
Catherine Schenk has written extensively on these balances.Footnote 10 She argues that ‘Britain’s accumulation of enormous sterling liabilities to the Commonwealth and colonies profoundly affected the post-war configuration of the sterling area system.’Footnote 11 Sterling balances were a major issue. If the government liberalised capital flows, Britain’s creditors could request to convert their sterling balances into dollars at the Bank of England. Without US loans, this would deplete the Bank’s reserves and force the country into devaluation. Post-war sterling history is a story of progressively phasing out these foreign balances.
The amount of sterling in circulation tripled between 1938 and 1947, whereas gross national product (GNP) only doubled.Footnote 12 At the same time, private and public holdings of gold and dollars in the United Kingdom halved. Barry Eichengreen argues that, in these circumstances, restoring convertibility ‘was the height of recklessness’.Footnote 13 Paul Einzig, a currency expert and prolific author at the time, also claims that forcing Britain ‘to restore the convertibility of sterling involved grave risks’ as it would trigger a run on the pound as long as Britain maintained a large trade deficit.Footnote 14 Freeing capital flows would lead to an outflow of capital from Britain, especially if sterling creditors were to convert their reserves into dollars. Echoing Eichengreen, Schenk wrote, ‘[T]he amount of national currency in overseas foreign exchange reserves led to doubts about the ability of the issuing country to sustain their external position and seemed to threaten the stability of the international monetary system.’Footnote 15
The sterling balances were posing a systemic risk to the Bretton Woods system and participants in this new regime needed to unite if the international monetary system was to survive. The British exposure partly explains the lenient American attitude towards Britain and the numerous lines of credit opened to the Bank of England.
Different Sterling Areas
The Emergency Powers (Defence) Act 1939 introduced the legal basis for wartime exchange controls. At the end of the war, the 1947 Exchange Control Act extended the policy of controls. Sterling was too weak to be exposed to international capital markets. The United Kingdom had also amassed an onerous war debt. The responsibility for exchange control was held by the Treasury, which delegated its management to the Bank of England. In turn, the Bank entrusted commercial banks with some of the day-to-day management of controls. For UK residents, the Exchange Control Act stipulated that ‘no person, other than an authorised dealer’ was allowed to ‘buy or borrow’ or ‘sell or lend any gold or foreign currency’.Footnote 16 Authorised dealers were UK banks. They were allowed to deal only with customers holding a licence to import or export. British banks could also give limited amounts of foreign currency to people travelling abroad.Footnote 17
The controls classified sterling into four main categories, reflecting four geographic areas: the sterling area, the dollar area, the transferable-account countries and the bilateral countries.Footnote 18 All other countries were referred to as unclassified-account countries. More categories would emerge over time, but these four categories would remain in place until the introduction of convertibility in December 1958. Each category had its own set of rules. The goal was ‘to restrict convertibility of sterling into dollars in the context of the post-war dollar shortage and generally to conserve foreign exchange’.Footnote 19 These different types were a means to manage capital controls. The dollar area was centred on the United States. The USSR and other countries benefited from transferable sterling status. Finally, most of Europe and the European colonies benefited from bilateral status.
Sterling area members all pegged their exchange rates to the pound, ‘maintained a common exchange control against the rest of the world while enjoying free current and capital transactions with the UK’ and kept their central bank reserves in sterling.Footnote 20 Dollar area sterling was held by ‘residents of the United States, Canada, the Philippines, Liberia, and thirteen Latin American countries’.Footnote 21 These sterling reserves had the advantage of being ‘convertible into dollars, with no strings attached’.Footnote 22
The transferable-account sterling countries observed the following rules: ‘Payments of sterling from one transferable sterling account to another were allowed freely, as were payments between transferable accounts and sterling area accounts. Transfers were not permitted from transferable accounts to bilateral or American accounts.’Footnote 23 The Radcliffe Report noted that transferable sterling ‘transactions took place in unofficial markets at a discount on the official rate’.Footnote 24 The report continues by arguing that the price of transferable sterling ‘indicated and affected the state of overseas confidence in sterling’. It played the part of a confidence barometer.
Bilateral countries, the most restricted group of countries with regard to capital controls, could transfer sterling to the sterling area only. These countries had to ask the Bank of England for ‘administrative transferability’ to move capital from one bilateral country to another. However, with the introduction of the European Payment Union (EPU) in 1950, ‘administrative transferability for bilateral OEEC [Organisation for European Economic Co-operation, former OECD] countries became virtually automatic’.Footnote 25
Unclassified account countries were ‘small and relatively unimportant countries’, according to Coombs.Footnote 26 Sterling balances in these countries were generally not transferable to zones or countries.
Beyond these five main classifications, other types of sterling existed, among them security sterling and ‘cheap’ or ‘free’ sterling. Security sterling was created in 1940 when the British monetary authorities witnessed capital outflows from foreign-owned funds mainly to New York. Security sterling holders had the right to transfer these between residents of the same monetary area, and this type of sterling was traded at a discount.Footnote 27
Figure 1.1 presents the variety of exchange rates for different types of sterling. The data come from an internal memorandum at the Federal Reserve Bank of New York and show the complexity of the system over a few days. Figure 1.1 presents boxplots of eighteen transferable sterling rates and ten bilateral rates for different locations. Both parts of Figure 1.1 have the same scale on the left-hand side (from $2.4 to $2.9 per £1 sterling). Bilateral rates occupy most of the space on the panel, but the difference between the eighteen transferable rates is limited. The standard deviation for the bilateral countries sample is almost five times greater than for the transferable sample (0.10 for the bilateral countries; 0.02 for the transferable countries). This means that sterling markets in transferable sterling countries were much more integrated than in bilateral sterling countries.
Figure 1.1. Sterling exchange rates for ten bilateral countries (top panel) vs eighteen transferable sterling rates (bottom panel)
The Convertibility Crisis of 1947
Sterling convertibility was a condition of the Anglo-American Financial Agreement of December 1945 (Anglo-American loan, in short). This explains why the United Kingdom had to implement it despite the risks posed by the sterling balances. As the provider of the Anglo-American loan, the United States wanted to rebuild Europe as a trading partner as quickly as possible. Convertibility was a key step for US policymakers. Not only was it on the agenda of the Bretton Woods agreement, but it later became a condition of the Marshall Plan.
Bradford De Long and Barry Eichengreen attribute the success of European post-war economic growth to the conditionality of the Marshall Plan. It ‘pushed European political economy in a direction that left its post-World War II “mixed economies” with more “market” and less “controls” in the mix’.Footnote 28 Marshall Aid forced European nations to liberalise their economies. Forrest Capie noted that convertibility took place ‘at a time when only the United States had the productive capacity to supply the goods that were needed to rebuild the war-ravaged economies of Western Europe’.Footnote 29 For the United States, convertibility was self-serving.
The Anglo-American Financial Agreement was settled in December 1945. The agreement stipulated that current account convertibility had to be introduced one year after the effective date of the agreement, on 15 July 1946. A year later, convertibility of the pound was put in place. Within a month, the Bank of England had lost $1 billion.Footnote 30 Schenk found that between 10 and 15 August, $175 million in reserves were lost. This led the Cabinet to consider withdrawing from convertibility while still keeping the United States on their side, as not all the money promised in the Anglo-American loan had been disbursed.Footnote 31
On Sunday, 17 August, the Cabinet met to debate convertibility. The debate quickly moved ‘to when and how to inform the Americans’ about the British intention to halt convertibility.Footnote 32 On 19 August 1947, convertibility was suspended. It was not sustainable. The 1947 experience was well anchored in the minds of British and US policymakers. This event, along with the sterling devaluation in 1949, weakened sterling’s credibility as an international currency, though it was still the second most important reserve currency.Footnote 33
New data presented highlight daily movements on the Exchange Equalisation Account’s (EEA’s) gold and dollar accounts (see Figure 1.2). During the crisis there were three significant dollar inflows: on 15 July for $150 million; on 26 July for $300 million; and on 12 August for $150 million. These $600 million represented drawings on the Anglo-American loan, which were all lost during the thirty-seven days of the crisis. Gold reserves remained relatively unaffected until 18 August, dropping by only $18 million. On both 19 and 20 August, the Bank bought $50 million with gold to deal with losses arising from the crisis. Convertibility was stopped when the Bank started losing its own gold reserves. The Americans no longer funded dollar losses.
Figure 1.2. EEA gold and dollar holdings, April–December 1947
Note: The dollar holdings have been converted to sterling at the $4.03 per sterling parity.
Figure 1.2 highlights how the 1947 crisis did not trigger large gold losses for the EEA. Apart from a $100 million gold loss during the final days of convertibility, the crisis left gold reserves untouched. Unlike the 1949 devaluation, which drained gold reserves, as we will see, the 1947 crisis did not affect British reserves. The Anglo-American loan funded the currency crisis. This loan, unlike reserves, was no immediate concern for British policymakers. The United Kingdom’s dollar holdings persistently fluctuate (bottom line in Figure 1.2). These fluctuations are market losses followed by inflows from US loans or grants.
But what was the impact of the crisis on sterling’s exchange rate? Looking at different rates gives different answers. The Manchester Guardian rate reported in Figure 1.3 shows little sign of the crisis’ lasting negative impact. It looks as if the crisis only briefly increased the volatility of the currency, and the exchange rate then settled at a higher dollar rate, around the official parity of $4.03/£. The Bank of England controlled the official London spot rate, and it shows no variation. The picture is different when we look at the sterling–dollar cross-rate based on the Swiss franc–dollar and Swiss franc–sterling rates (scale on the right-hand side in Figure 1.3). Here I present new data on offshore rates from Switzerland collected from the archives of the Swiss National Bank. Using the free Swiss banknote cross-rate gives a more accurate picture of the crisis.Footnote 34 The offshore rate dropped by 1.5 per cent during the day of the crisis and fell further after the actual crisis. At the end of 1947 the offshore rate reached a low point of $2.30/£, representing a drop of almost 21 per cent in less than six months. Convertibility put pressure on the pound, even if this was not directly reflected in the controlled official exchange rates.
Consequences of the Crisis
What were the longer-term consequences of the convertibility crisis? Official exchange rates do not answer this question. They were tightly controlled until the opening of the foreign exchange market and so offer no information on the pressure on the exchange rate.
I use an Exchange Market Pressure (EMP) index to measure the pressure on the pound. I will use these indices on several occasions in this book, with variations that I will explain. In short, the higher the index, the more pressure on the pound. EMPs were used in the macroeconomic literature to determine stress on a currency with fixed exchange rates.Footnote 35 The belief is that if the price of the currency does not reflect market pressure, other indicators do.
Traditionally, the indices use the exchange rate, the central bank interest rate and central bank reserves. The Bank of England did not change the Bank Rate during the period in question, making the interest rate an uninteresting variable. Equally, using the official spot rate alone is misleading. It was highly controlled and displayed little volatility, as Figure 1.3 illustrates. I have computed a daily EMP index using offshore Swiss banknote cross-rates, EEA gold and dollar reserves and the official exchange rate. This latter rate is stable, but tends to move during crises. All three indicators are divided by their standard deviation to weight them equally.
The 1947 crisis shook the foundations of sterling. Exchange rate pressure started to increase immediately after. The British currency had lost its international currency status, and this was now public knowledge. The consequences were long-lived, as shown in Figure 1.4.
Figure 1.4. Sterling EMP index, 1947–49
Note: Data are indexed at March 1947 = 100. The shaded area highlights the convertibility crisis. The pressure is cumulative in this index.
After the 1947 crisis, the index follows an upward trend. The trend continues until the 1949 devaluation, when the chart stops. The aftermath of convertibility marked a continuous period of decline in reserves and lower exchange rates. This meant more pressure on the pound. Note that the sudden drop in the index in January 1948 was due to a Marshall Aid inflow. The United States disbursed $100 million, and this affected the reserve position and hence the EMP index.
The global financial community knew that the Bank of England did not have the firepower to defend a convertible sterling. Sterling declined from being an international currency with positive externalities for the United Kingdom to a concern for the international community to manage.Footnote 36 Now there was no longer a bipolar sterling–dollar paradigm; instead, sterling was a problematic international currency. The 1947 episode brought the United States and United Kingdom together to manage an orderly retreat for sterling. In the following pages I explore how the British currency was managed by the United States, the United Kingdom and the international community. All stakeholders hoped to avoid contagion from sterling to the dollar. If it was clear that sterling was only a shadow of its former self, contemporaries still believed that sterling was an important currency. And this mattered.
Open access